## ftnea2025011

## Source details

**Canonical URL:** [ftnea2025011](https://www.imf.org/-/media/files/publications/ftn063/2025/english/ftnea2025011.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/ftn063/2025/english/ftnea2025011.pdf.md)
- [Structured JSON version](/-/media/files/publications/ftn063/2025/english/ftnea2025011.pdf.json)

---

### Overview and key messages
- Central banks explore tokenized reserves to preserve the safety, liquidity, and policy role of central bank money within tokenized ecosystems and to support more efficient, automated, and resilient wholesale payment systems.
- Implementation choices (ledger designs, governance structures, stakeholder engagement) must align with policy objectives and desired levels of control, risk tolerance, and authority over issuance, access, data, and system continuity.
- Tokenization may not fundamentally affect central banks’ ability to implement monetary policy but could introduce enhancements such as programmable or automated liquidity management, requiring updates to governance, legal frameworks, and risk management.
- Alternative solutions include real-time gross settlement (RTGS) links, omnibus accounts, and privately issued tokenized money; each presents differing trade-offs in risk, cost, control, programmability, and alignment with policy objectives.
- Central banks are likely to adopt varying strategic approaches guided by policy goals, market readiness, institutional capacity, and supported by research, testing, and legal preparation.

### Definitions — What are tokenized reserves?
- Tokenized reserves, as used in this Note, refer specifically to reserves that:
  - Are a direct liability of the central bank.
  - Are issued using DLT.
  - Are accessible only to a predefined group of institutions—typically FIs.
- The Note uses "tokenized reserves" to emphasize the technological change (reserves issued on DLT) rather than implying a wholly new product.
- Technical assumption: tokenization occurs through DLT given its current prevalence.

### Policy objectives and wholesale use cases
- Two principal policy objectives cited:
  - Preserving the advantages of central bank money in a future tokenized financial system.
  - Enhancing the efficiency of the wholesale payment system.
- Preserving advantages of central bank money:
  - Central bank money is the most liquid and safe asset; reserves enable interbank settlement and are key to monetary policy implementation.
  - Principle 9 of the Principles for Financial Market Infrastructures recommends interbank settlement in central bank money where practical and available.
  - Tokenized reserves may help preserve central bank money as a risk-free settlement asset and mitigate credit and liquidity risks from privately issued money.
- Enhancing wholesale payment system efficiency:
  - DLT offers distributed architecture, potential resilience gains, and higher transparency (balanced against privacy considerations).
  - Smart contracts enable programmability and automation of conditional payments and workflows.
  - DLT experiments have moved from replicating RTGS to advanced use cases: delivery-versus-payment (DvP) and payment-versus-payment (PvP).
  - Key innovation: atomic settlement from having money and assets on the same ledger.
  - Multilateral settlement platforms (Project mBridge, Dunbar, Jura, Agorá) demonstrated real-time, peer-to-peer, atomic PvP transactions between commercial banks in different jurisdictions.

### Atomicity and atomic settlement
- Atomicity: technical guarantee that a multistep database operation executes in full or not at all.
- Atomic settlement: guarantees all legs of a multistep transaction are completed, or none are.
- Strict atomicity is guaranteed only when assets and payments are recorded and settled on a single ledger with a unified execution environment; smart contracts enforce simultaneous delivery and payment.
- Cross-ledger transactions can achieve at best "weak" atomicity (e.g., hash-time-locked contracts), introducing operational complexity, timeouts, and settlement failure risks.
- Throughout the Note, "atomicity" and "atomic settlement" refer to the strict form unless otherwise specified.

### Experiments, projects, and jurisdictional signals
- Notable projects and pilots:
  - Domestic-focused: Project Jasper, Ubin, Inthanon, Khokha.
  - Multilateral cross-border: Project mBridge, Dunbar, Jura, Agorá.
  - Project Helvetia Phase III — Digital SNB Bill (June 2024): issuance of 64 million Swiss francs in 7-day digital SNB bills, settled in tokenized reserves on the SIX Digital Exchange platform.
  - Project Pine (2025): prototype for implementing monetary policy in a fully tokenized environment; explored four monetary tools: 1. interest-bearing reserves, 2. open market operations, 3. collateral management, 4. asset purchases.
  - Project Ensemble sandbox (Hong Kong Monetary Authority, 2024) for industry-driven interbank settlement experiments.
- Survey signal:
  - 75 percent of surveyed central banks indicated interest in wCBDC [tokenized reserves] in the 2024 BIS survey on CBDC, citing preservation of central bank money as an important or very important reason for potential issuance.

### Emerging Markets — potential benefits, risks, and use cases (Box 2)
- Potential benefits:
  - May enhance efficiency of cross-border payments and trades and modernize FMIs.
  - Could improve capital market development—investor access, operational efficiency, liquidity, and tradability of assets.
  - Smart contracts can enable flexible fractionalization and lower investment minimums.
- Risks and caveats:
  - May complicate capital flow management and require robust legal and institutional frameworks.
  - Tokenization alone does not change fundamental asset characteristics or automatically make assets eligible as collateral.
- Cross-border and trade finance experiments:
  - Project Mandala: DLT for streamlining compliance for cross-border transactions.
  - Project Rialto: automated market makers for wholesale FX conversion using tokenized reserves.
  - Central bank collaborative pilots tokenized real-world goods and used tokenized reserves for atomic settlements in trade finance.
- Ledger model considerations for EMs:
  - Choice depends on institutional, technological, and market context.
  - Ledger models: single ledger; compatible ledgers.
  - Operating models for a single ledger (four basic types): Integration (central bank operated); Distribution (jointly operated); Separation (third party operated); Permissionless (decentralized).

### Ledger and operating models — technical opportunities and risks
- Ledger models and trade-offs:
  - Single ledger: feasible for advanced programmability and strict atomic settlement via smart contracts.
  - Compatible ledgers: can enable programmable settlements across ledgers but introduce operational and interoperability risks.
- Operating model descriptions:
  - Integration (central bank operated): central bank controls operation and governance.
  - Distribution (jointly operated): shared operation and governance with stakeholders.
  - Separation (third party operated): central bank issues tokenized reserves on a third-party ledger.
  - Permissionless (decentralized): tokenized reserves on a public, permissionless infrastructure; governance decentralized.
- Technical opportunities:
  - Single ledger enables atomic DvP, DvPvP and complex multistep transactions without third-party legal enforcement.
  - Smart contracts allow embedded eligibility checks, settlement logic, and risk parameters.
- Key risks and trade-offs:
  - Increased pre-funding and liquidity needs due to instantaneous settlement; instant settlement can eliminate netting.
  - Smart contract vulnerabilities and unintended automation.
  - 24/7 operations increase operational complexity, requiring stronger risk and liquidity management.
  - Integration model concentrates operational risk in the central bank; separation model reduces direct control but fosters private innovation.

### Critical central bank controls and functions across models
- Issuance and redemption:
  - Central banks should retain exclusive control over issuance and redemption; smart contract mint/burn controls should be restricted to the central bank in external-operator models.
- Access and eligibility:
  - Central banks should control who holds and transacts tokenized reserves and retain the ability to suspend accounts and freeze tokens.
- Data management and transparency:
  - Central banks should maintain visibility into transaction data for policy, oversight, and risk monitoring while respecting privacy.
- Ability to halt settlements:
  - Central banks should be able to halt or reverse settlements in emergencies; permissionless infrastructures make full-system halts nearly impossible.
- Continuity of settlement:
  - Models vary in contingency and recovery responsibilities; central-bank-operated ledgers enable direct contingency planning, while separation models rely on third-party arrangements.

### Macrofinancial implications and access policy trade-offs
- Single ledger environments can amplify contagion risks via transaction dependencies, faster settlement, and automation.
- Multiple non-interoperable ledgers risk market fragmentation, reduced liquidity, and price divergence across identical assets.
- Extending access to NBFIs or foreign FIs can promote competition and inclusion but introduces supervisory complexity, operational/financial risks, and potential funding-model impacts for banks.
- Granting access to foreign institutions may enhance cross-border payments but could increase capital flow volatility or currency substitution.
- These trade-offs require careful jurisdictional evaluation.

### Monetary policy implementation — general effects and scenarios
- General overview:
  - Tokenization would not fundamentally affect the central bank’s ability to implement monetary policy; core operations (liquidity provision, interest rate steering, collateralized operations) remain intact and could be enhanced via automation and programmability.
  - Expanded access or new wholesale use cases could change demand for central bank money and require liquidity management adjustments.
- Scenario 1 — Only tokenized reserves exist as wholesale central bank money:
  - Assumes tokenization widely adopted and tokenized reserves fully replace traditional reserves for interbank settlements.
  - Monetary policy implementation and market conventions need redesign for continuous, automated, composable settlement environments.
  - Programmability and 24/7 availability could allow monetary policy instruments to operate faster or beyond standard business hours.
  - Traditional market conventions (end-of-day cutoffs, day counts for interest calculation) may no longer apply; central banks must redesign reserve requirements and standing facilities for continuous-time systems.
  - Tokenized monetary policy instruments could:
    - Use smart contracts for modular upgrades and rapid implementation changes.
    - Automatically and continuously optimize liquidity management or trigger facilities based on predefined parameters.
    - Include tokenized central bank bonds on the same ledger to support efficient liquidity absorption.
  - Operational benefits must be integrated with human oversight.
- Scenario 2 — Coexistence of tokenized and traditional reserves:
  - Context: central bank issues tokenized reserves alongside traditional reserves.
  - Operational challenges:
    - Liquidity fragmentation from split reserve pools (RTGS and DLT-based).
    - If not seamlessly interchangeable, liquidity may become siloed and complicate steering short-term interest rates.
    - Design and policy differences (remuneration, access) could incentivize shifts in balances and create volatility.
  - Policy and operational safeguards recommended:
    - Coordinated liquidity monitoring.
    - Improved liquidity forecasting.
    - Interoperability for seamless transfers between reserve types.
    - Transitional measure: allow only intraday balances of tokenized reserves, requiring end-of-day conversion back to traditional reserves (as in Project Jura and Project Helvetia experiments).
    - Legal and economic equivalence across reserve types to preserve fungibility (e.g., overnight tokenized reserves should earn the same interest and count equally toward reserve requirements).
  - Policy choice: central banks could operate in both reserve systems if the tokenized segment grows to ensure effective policy transmission.

### Collateralization, programmable collateral, and risks
- Benefits of integrating tokenized reserves and tokenized collateral:
  - Embedded eligibility checks, settlement logic, and risk parameters in smart contracts.
  - Real-time margin calls, dynamic collateral substitution, and atomic DvP collateralized payments.
  - On-demand liquidity using collateral on a “just-enough” basis to reduce opportunity cost of locked assets.
  - Improved tradability, liquidity, price discovery for tokenized assets.
- Risks requiring redesign:
  - Cybersecurity and operational risks from erroneous smart contracts.
  - Contagion risk from tight integration of collateral and reserves.
  - Need for legal recognition of tokenized assets as eligible collateral and adjusted haircut frameworks.

### Alternative settlement solutions and core comparisons
- Solutions discussed: tokenized reserves (single ledger and compatible ledgers), RTGS links, omnibus accounts, privately issued tokenized money (tokenized deposits, deposit tokens, stablecoins).
- Comparative findings:
  - All six solutions support DvP settlement.
  - Atomic settlement (strict):
    - Enabled by tokenized reserves on a single ledger, omnibus accounts, and private tokenized money on single ledgers.
    - Not enabled (strict atomicity) by tokenized reserves on compatible ledgers, RTGS links, and private tokenized money on separate ledgers (they may achieve weak atomicity).
  - Central bank money settlement:
    - Tokenized reserves and RTGS links settle in central bank money.
    - Omnibus accounts reduce credit risk but may not constitute settlement in central bank money depending on legal treatment.
    - Privately issued tokenized money typically carries higher credit and liquidity risk unless fully backed by central bank reserves.

- Table summary (capabilities expressed as Yes/No in the Note):
  - Tokenized reserves—single ledger: DvP Settlement? Y ; Atomic Settlement? Y ; Central Bank Money Settlement? Y
  - Tokenized reserves—compatible ledgers: DvP Settlement? Y ; Atomic Settlement? N ; Central Bank Money Settlement? Y
  - RTGS link: DvP Settlement? Y ; Atomic Settlement? N ; Central Bank Money Settlement? Y
  - Omnibus account: DvP Settlement? Y ; Atomic Settlement? Y ; Central Bank Money Settlement? N
  - Private tokenized money—single ledger: DvP Settlement? Y ; Atomic Settlement? Y ; Central Bank Money Settlement? N
  - Private tokenized money—compatible ledgers: DvP Settlement? Y ; Atomic Settlement? N ; Central Bank Money Settlement? N

### Strategic approaches, sequencing, and implementation guidance
- Four strategic approaches for central banks engaging with tokenized asset markets:
  1. Inaction — deem initiatives not relevant or feasible now.
  2. Wait and see — monitor developments, gather data, consult stakeholders, conduct legal reviews or small-scale proofs of concept.
  3. Enablement — establish foundations for market-led developments via research, infrastructure, and regulatory frameworks; consider short-term solutions such as an RTGS link.
  4. Catalyst — proactive engagement to stimulate market adoption (e.g., develop RTGS links or make reserves available on DLT infrastructure).
- Sequencing and prioritization:
  - Prioritize short-term, lower-cost options (e.g., RTGS link) while developing complex solutions later.
  - Use a two-dimensional framework: suitability (alignment with policy goals) and feasibility (time, complexity, resources).
- Implementation guidance:
  - Use IMF’s “5P methodology” for project management: proof of concept, prototype, pilot, and potentially production.
  - Complementary activities: legal and regulatory work; technical and macrofinancial research; international monitoring and collaboration.

### Conclusion — policy questions and guidance
- Key policy questions:
  - If tokenization is widely adopted, should central banks issue tokenized reserves on DLT for payments and settlements of tokenized assets?
  - Which implementation approach best meets policy objectives and what factors must be considered?
  - How do alternative solutions (RTGS link, privately issued tokenized money) compare with tokenizing reserves?
  - Which strategic approach should a central bank adopt given policy direction, market demand, and capacity?
- Guidance emphasis:
  - Compare tokenized reserves with alternatives for DvP-based settlement using specified trade-offs.
  - Focus on implementation considerations, institutional constraints, market trends, and central bank views.
  - Recommend balanced assessment of benefits and risks based on country context, legal requirements, resource constraints, and evidence-based policymaking.
  - Reiterate that DLT and tokenized reserves complement—not substitute—strong institutions, sound policies, and robust legal frameworks.

*Source: Central Bank Exploration of Tokenized Reserves — NOTE 2025/011*

### References .............................................................................................................

### ftnea2025011 - References

### Overview
- This content unit is part of the Fintech Note titled "Central Bank Exploration of Tokenized Reserves."
- The Note examines opportunities, risks, implementation considerations, and policy implications of tokenizing central bank reserves (also referred to as wholesale central bank digital currency or wCBDC in some contexts).
- Structure of the Note: definition of tokenized reserves; policy objectives; implementation considerations and central bank roles; implications for monetary policy implementation; alternative solutions; timing, prioritization, and strategic choices; conclusion.

### Key messages and findings
- Central banks explore tokenized reserves to preserve the safety, liquidity, and policy role of central bank money within tokenized ecosystems by enabling risk-free settlement and support for more efficient, automated, and resilient wholesale payment systems.
- To ensure policy effectiveness, central banks must align implementation choices (ledger designs, governance structures, stakeholder engagement) with policy objectives and desired levels of control, risk tolerance, and authority over issuance, access, data, and system continuity.
- Tokenization may not fundamentally affect central banks’ ability to implement monetary policy but could introduce enhancements such as programmable or automated liquidity management, requiring updates to governance, legal frameworks, and risk management.
- Alternative solutions to tokenized reserves include real-time gross settlement (RTGS) links, omnibus accounts, and privately issued tokenized money; each offers differing trade-offs in risk, cost, control, programmability, and alignment with policy objectives.
- Central banks are likely to adopt varying strategic approaches guided by policy goals, market readiness, institutional capacity, and supported by research, testing, and legal preparation.

### Definitions — What are tokenized reserves?
- Tokenized reserves, as used in this Note, refer specifically to reserves that:
  - Are a direct liability of the central bank.
  - Are issued using DLT.
  - Are accessible only to a predefined group of institutions—typically FIs.
- The term wCBDC is sometimes used interchangeably with tokenized reserves, but definitions vary across institutions; this Note uses "tokenized reserves" to emphasize the specific technological change (reserves issued on DLT) rather than implying a wholly new product.
- Related terminology and cross-references:
  - wCBDC also described as "wholesale tokenized central bank money" (BIS and CPMI 2024) and as "a new platform for the distribution of wholesale central bank money" (Bank of England 2024a).
  - Bank of Canada and others (2025) use the term “wholesale central bank money tokens.”
- Technical assumption: the Note assumes tokenization occurs through DLT given its current prevalence, while acknowledging uncertainty about future technological adoption.

### Policy objectives and use cases
- Two principal policy objectives cited by central banks investigating tokenized reserves:
  - Preserving the advantages of central bank money in a future tokenized financial system.
  - Enhancing the efficiency of the wholesale payment system.
- Preserving advantages of central bank money:
  - Central bank money is the most liquid and safe asset; reserves enable interbank settlement and are key to monetary policy implementation.
  - International standards such as Principle 9 of the Principles for Financial Market Infrastructures recommend interbank settlement in central bank money where practical and available.
  - Central banks can inject liquidity via reserves during stress and act as lenders of last resort.
  - If financial institutions transact increasingly in tokenized assets, tokenized reserves may help preserve central bank money as a risk-free settlement asset and mitigate credit and liquidity risks from privately issued money (Bank of Canada and others 2025).
- Enhancing wholesale payment system efficiency:
  - DLT offers distributed architecture, potential resilience gains (reducing single points of failure), and higher transparency, balanced against data privacy considerations.
  - Smart contracts enable programmability and automation of conditional payments and workflows.
  - Early DLT experiments focused on replicating RTGS systems but often found limited advantages for simple interbank payment use cases, especially where efficient centralized infrastructures already exist.
  - DLT projects have shifted toward advanced use cases: delivery-versus-payment (DvP) for asset transfers and payment-versus-payment (PvP) for cross-border payments.
  - The key innovation with tokenized reserves is atomic settlement from having money and assets on the same ledger (see Box 1).
  - Multilateral settlement platforms (Project mBridge, Dunbar, Jura, Agorá) have demonstrated real-time, peer-to-peer, atomic PvP transactions between commercial banks in different jurisdictions with tokenized reserves linked to respective central banks.
  - Project Agorá additionally experiments with tokenizing commercial bank deposits to support credit intermediation while facilitating payments.

### Atomicity and atomic settlement (Box 1)
- Atomicity: technical guarantee that a multistep database operation executes in full or not at all.
- Atomic settlement: technology guarantees all legs of a multistep transaction are completed, or none are; prevents one party delivering while the other fails.
- Strict atomicity is guaranteed only when both assets and payments are recorded and settled on a single ledger with a unified execution environment; smart contracts can enforce simultaneous delivery and payment, with settlement finality provided by the consensus mechanism rather than legal arrangements.
- Cross-ledger transactions can achieve at best "weak" atomicity using mechanisms like hash-time-locked contracts; these require all ledgers to be live and synchronized and introduce operational complexity, timeouts, and risks of settlement failure or dispute—thus cannot replicate strict atomicity.
- Throughout the Note, "atomicity" and "atomic settlement" refer to the strict form unless otherwise specified.

### Experiments, projects, and jurisdictional considerations
- Notable experiments and projects mentioned:
  - Project Jasper, Ubin, Inthanon, Khokha — focused on domestic wholesale settlement.
  - Project mBridge, Dunbar, Jura, Agorá — multilateral cross-border settlement platforms demonstrating atomic PvP transactions.
  - Project Helvetia III described by the SNB as piloting "tokenized central bank money for wholesale use" (Jordan 2024).
  - Project Ensemble sandbox launched by the Hong Kong Monetary Authority (2024) for industry-driven interbank settlement use cases with experimental tokenized money.
  - Project Jura and Guardian demonstrated multi-leg, programmable transactions across currencies, asset classes, and jurisdictions.
- Survey data and institutional signals:
  - 75 percent of surveyed central banks indicated interest in wCBDC [tokenized reserves] in the 2024 BIS survey on CBDC; they cited preserving the role of central bank money as an important or very important reason for potential issuance and noted tokenized reserves could preserve settlement of tokenized assets such as tokenized securities (Illes, Kosse, and Wierts 2025).
- Jurisdictional distinctions:
  - Advanced economies: priority often to preserve the role of central bank money in future tokenized asset settlements rather than replacing already efficient interbank systems.
  - Emerging markets (EM): tokenized reserves and DLT may offer pathways to modernizing financial market infrastructures, but entail important cost and risk considerations (Illes, Kosse, and Wierts 2025; Box 2).

### Scope and focus
- The Note focuses mostly on tokenized reserves in domestic DvP settlement.
- In-depth analysis of cross-border and multicurrency settlement is left for future analysis.

*FINTECH NOTES  Central Bank Exploration of Tokenized Reserves, INTERNATIONAL MONETARY FUND*

### Box 2. Emerging Markets’ Exploration

### Box 2. Emerging Markets’ Exploration

### Potential benefits and risks
- Tokenized reserves could enhance the efficiency of cross-border payments and trades but may complicate capital flow management.
- Technology should complement—not replace—strong institutions, sound policies, and robust legal frameworks; poorly implemented technologies could undermine public trust and resilience.
- Associated costs and risks could arise and warrant an in-depth, separate analysis once sufficient experience has been accumulated.

### Better enforcement of rules or policies
- EMs may face less robust legal and regulatory enforcements compared to advanced economies; embedding rules at the infrastructure level may enhance policy implementation and ensure compliance from all participants.
- Example: In central securities depositories (CSDs), some treasuries set a minimum tradable amount of securities to diversify the investor base; integrating the CSD and tokenized reserves on a single ledger can enforce such rules by implementing the minimum tradable unit at the asset layer.1

### Cross-border payments and trade finance
- EMs often face greater friction in cross-border payments, such as high foreign exchange (FX) and compliance costs; DLT and tokenized reserves have shown potential to reduce some of these frictions.
- Project Mandala showed the potential of DLT for streamlining compliance for cross-border transactions by managing regulatory requirements in the context of cross-border tokenized reserves.2
- Project Rialto demonstrated how automated market makers for wholesale FX conversion, using tokenized reserves as a settlement asset, could support lower FX costs and secured instant cross-border retail payments.3
- Central bank-led collaborative projects:
  - Banco Central do Brasil and the Hong Kong Monetary Authority under Project Ensemble and Drex pilot program.
  - Bank of Ghana with Monetary Authority of Singapore.
  - These experiments tokenized real-world goods (for example, agricultural products) and used tokenized reserves for atomic settlements in trade finance to boost trust among importers and exporters, enhance transparency in supply-chain tracking, and improve access to funding for local firms.4

### Capital market development
- Interest in tokenized reserves and assets to support capital market development includes improving investor access, operational efficiency in asset issuance, and liquidity and tradability of assets (Leung and others 2023).
- Fully atomic settlement for money and assets within the same ledger could improve trust and address counterparty and settlement risks associated with collateral pledging and rehypothecation.5
- Tokenization may streamline the asset lifecycle by automating issuance, trading, and ownership transfers, reducing reliance on intermediaries and lowering associated costs.
- Smart contracts enable flexible fractionalization, lowering investment minimum thresholds, widening the investor base, and enhancing asset liquidity.
- Note: Tokenization alone does not automatically enable new eligible collateral or reduce the risk of certain types of collateral.5

### Implementation considerations
- Central banks must evaluate whether tokenized reserves should be issued on an exclusive ledger or on a ledger shared with other tokenized assets, and what framework should govern ledger operation and management.
- Each model involves trade-offs; the most suitable approach depends on each country’s institutional, technological, and market context.

### Ledger models (basic)
- Single ledger: Tokenized reserves and other assets such as bonds or securities are issued and exchanged on the same ledger.
- Compatible ledgers: Tokenized reserves are issued on a dedicated ledger without other asset types and can interoperate with other ledgers containing tokenized assets.
- Note: DLT = distributed ledger technology.

### Operating models for a single ledger (four models)8
- Integration (central bank operated): The central bank controls operation and governance of the ledger hosting both tokenized reserves and other assets; it is the main operator and determines governance, including access policies.
- Distribution (jointly operated): The ledger is operated and governed jointly by the central bank and other stakeholders (private institutions or other central banks); decision-making and operational responsibilities are shared.
- Separation (third party operated): The central bank issues tokenized reserves on a ledger operated by a third party, typically a private infrastructure provider; the central bank acts as participant and, where appropriate, as regulator or overseer.
- Permissionless (decentralized): Tokenized reserves are issued on a permissionless DLT infrastructure that is open and public; governance is decentralized with no single entity controlling the ledger and the central bank operates as a peer participant.
- The four models are not always distinct and hybrid models are possible.

### Notable experiments and analogies
- Integration example: Banco do Brasil as part of their Drex project.
- Distribution example: Project mBridge, developed collaboratively by four central banks, allows each participating central bank to issue its own tokenized reserves on a multicurrency platform with shared governance and operational responsibilities.
- Separation example: Project Helvetia, where the SNB issued tokenized reserves onto the SIX Digital Exchange, a DLT-based exchange operated by the private SIX Group; SIX Digital Exchange operates the DLT platform while the SNB provides tokenized wholesale central bank money for settlement.
- Permissionless model is technologically possible but entails operational challenges (transaction scalability and costs, governance, and accountability); implementing governance principles for FMIs in a permissionless environment is challenging because system management lies outside the control of any single institution.
- Principle 2 of the Principles for Financial Market Infrastructures emphasizes governance: “An FMI should have governance arrangements that are clear and transparent, promote the safety and efficiency of the FMI, and support the stability of the broader financial system, other relevant public interest considerations, and the objectives of relevant stakeholders” (CPMI-IOSCO 2012, p. 1).

*Source: ftnea2025011 - Box 2. Emerging Markets’ Exploration*

### Box 3. Non-distributed Ledger Technology Case Studies of Integration,

### Box 3. Non-distributed Ledger Technology Case Studies of Integration, Distribution, and Separation Models

### Case studies: integration, distribution, and separation
- Integration
  - T2S (Eurosystem), Fedwire Securities (United States), and BOJ-NET JGB Services (Japan) illustrate integration between cash and securities.
  - Rwanda: its only central securities depository (CSD) is owned and operated by the National Bank of Rwanda, holds both government and private securities and central bank money, and integrates private securities (Wendt, Katz, and Zanza 2019).
- Distribution
  - Canada: Lynx (large-value payment system) is owned and operated by Payments Canada (a not-for-profit, public-purpose organization) where the Bank of Canada is a member. The central bank provides integration of settlement accounts, enabling settlement in central bank money—reflecting a distribution model with similarities to separation.
- Separation
  - Switzerland: Swiss Interbank Clearing (RTGS) is operated by private entity SIX Interbank Clearing; the Swiss National Bank provides integration of settlement accounts and oversight but does not own or operate it.
  - Euroclear Sweden: private CSD settling private securities in central bank money through Riksbank accounts integrated into its platform; this separation setup is noted as rare.

### Operating models for compatible ledgers
- Central-bank-owned-and-operated model
  - Tokenized reserves’ ledger likely fully owned and operated by the central bank, similar to current wholesale payment system management and the integration model.
  - Enables direct control of critical rules and functions: setting eligible criteria for participants and interoperable ledgers, issuing, redeeming, and validating tokenized reserves.
- Outsourced/third-party operations (separation-like)
  - Central bank outsources ledger operations to a third-party provider while retaining strong influence over critical rules and functions.
  - Switzerland’s Swiss Interbank Clearing is cited as a comparable non-DLT example (Box 3).

### Technical opportunities and risks
- Advanced programmability and atomic settlement
  - Single ledger model:
    - Generally more feasible and operationally sound for advanced programmability (atomic settlement; complex multistep transactions such as DvP, PvP, DvPvP).
    - Can uniquely guarantee strict atomic settlements via smart contracts without third-party legal entities, mitigating counterparty and settlement risks.
    - If programmability and atomicity are priorities, a central bank may favor the single ledger model.
  - Compatible ledger model:
    - Can enable programmable settlements across different DLT ledgers but introduces operational and other technical and nontechnical risks.
    - May require an orchestrating entity, third-party insurance/legal agreements for failed executions.
    - Technical choices (consensus protocols, messaging/data/compliance standards) are critical to reduce friction and improve interoperability and safety.
    - Can still enable DvP or PvP settlement but typically relies on third parties to reduce operational and counterparty risks.
- Other risks and trade-offs
  - Simultaneous settlements can increase pre-funding requirements and liquidity needs; instant settlements eliminate netting possibilities (OECD 2025).
  - Smart contract vulnerabilities, poor governance, unintended automation, privacy concerns from increased data sharing/transparency.
  - 24/7 operations add complexity, requiring stronger risk and liquidity management frameworks.
- Impact of operating model on central bank risk and innovation
  - Integration model: central bank retains full control and directly implements safeguards; expanded operational role may crowd out private sector innovation and raise operational, cyber, reputational risks.
  - Separation model: day-to-day management delegated to third-party operator; central bank sets key policy parameters as issuer—reduces control but encourages private sector innovation.

### Macrofinancial implications
- Single ledger environment risks
  - Transaction dependencies, faster settlement speed, atomicity, and automation can amplify contagion risks and financial stability vulnerabilities (Danmarks Nationalbank 2025).
  - Programmability can create complex assets with returns/risks tied to multiple underlyings, increasing exposure across investors.
- Compatible ledgers and market fragmentation
  - Operating multiple ledgers without interoperability mechanisms and common standards can heighten market fragmentation, decreasing liquidity, increasing transaction costs, and causing information asymmetry.
  - Example: identical bonds across different ledgers may trade at different prices, undermining benchmark yield curves.
- Access policy trade-offs
  - Extending access to NBFIs or foreign FIs can promote competition, innovation, and financial inclusion but introduces operational/financial risks, supervisory complexity, and potential weakening of banks’ funding models (IMF 2020; BIS 2021; Bank of England 2024b; World Bank 2024).
  - Granting access to foreign institutions could enhance cross-border payments but may introduce increased capital flow volatility or greater currency substitution (IMF 2024; Reuter and others 2025).
  - These benefits and trade-offs apply to both traditional and tokenized reserves and require careful jurisdictional evaluation.
- Monetary policy implementation implications
  - Central banks may implement monetary policy through tokenized reserves, depending on tokenization adoption and access policy; effective transmission and implementation of monetary policy must be ensured as tokenization grows.

### Critical ledger management functions and controls (central bank priorities)
- Issuance and redemption
  - Central banks should retain exclusive control over issuance and redemption, managing conversions between traditional and tokenized reserves across all models.
  - In external-operator models, smart contracts controlling minting/burning should be restricted to central banks.
  - Distribution models: joint governance must ensure issuance authority rests solely with the central bank.
  - Separation models: issuance/redemption functions must remain under central bank technical or contractual control (example: Project Helvetia).
  - Permissionless models: central bank should control issuance and rules in related smart contracts to prevent unauthorized duplication/manipulation.
- Access and eligibility
  - Central banks should control who can hold and transact tokenized reserves and retain ability to suspend accounts and freeze tokens.
  - Integration models: strongest central bank control.
  - Distribution and separation models: access control must be negotiated and secured via technical tools or legal agreements.
  - Permissionless model: access must be secured through technical tools (e.g., whitelisting) for wallet or node criteria.
- Data management and transparency
  - Central banks should maintain visibility into transaction data for policy, oversight, and risk monitoring while respecting privacy.
  - Central bank-operated/integration models: full visibility.
  - Distribution model: data-sharing agreements or node access for supervisory functions.
  - Separation model: transparency via contractual rights or technical arrangements (audit trails, observer nodes).
  - Permissionless model: pseudonymity limits tracking; may require external analytics to derive meaningful insights.
- Ability to halt settlements
  - Central banks should be able to halt/reverse settlements in emergencies (cyberattacks, technical failures, participant insolvency).
  - Central-bank-operated ledger: clearest path with emergency stop under direct control.
  - Distribution/separation models: require pre-agreed processes, smart contract mechanisms, and contractual rights (example: SNB in Swiss Interbank Clearing).
  - Permissionless: full-system halts nearly impossible; administrative controls at token layer (freeze/revoke in token contracts) affect only token layer and require participant coordination for broader application-level controls.
- Continuity of settlement
  - Central-bank-operated ledger: direct operational control, contingency planning, recovery mechanisms.
  - Distribution models: shared contingency planning with partners.
  - Separation models: reliance on third party’s arrangements and regulatory/oversight authority; some central banks may need their own safeguards/contingency solutions.10
  - Permissionless: no operational control or regulatory influence over underlying infrastructure; resilience depends on network consensus and design.

### Additional considerations affecting model selection
- Legal foundations
  - Essential across all ledgers and operating models, including legal recognition of smart contract operations, tokenized money, and settlement finality.
  - Separation models with private ledgers may require additional legal adaptation to define central bank rights and liabilities.
  - In some jurisdictions, certain operating models may not be legally feasible.
- Capital flow management measures
  - Integration models enable central banks to enforce and retain direct control; separation or permissionless models may complicate implementation, requiring coordination with third parties or reliance on smart contract execution.11
- Institutional capacity, resources, and risk appetite
  - Central banks with strong operational capabilities and low risk tolerance may prefer compatible ledgers and integration models.
  - Greater openness to private collaboration may make distribution or separation models feasible if private operators meet regulatory/technical expectations and central banks can effectively oversee arrangements.
  - Cost considerations: integrated models generally place most development and operating costs on the central bank; external operators enable cost sharing but raise operational risk concerns.
- Financial market development
  - Less developed or smaller markets: private sector limited capacity may favor integration or public-led distribution (assuming sufficient public capacity).
  - Advanced markets with mature FMIs and robust regulation may find separation models viable.
  - Decisions are not static—transitions can occur over time.
- Nature of tokenized assets and coordination with fiscal authorities
  - If focus is on tokenizing government securities, coordination between central banks and fiscal authorities is essential; tighter public-sector control (integration or distribution) may be appropriate.
  - Where private entities manage core FMIs and private assets are tokenized on private ledgers, separation models may be practical.

### Monetary policy implementation with tokenized reserves (overview and Scenario 1)
- General overview
  - Tokenization would not fundamentally affect the central bank’s ability to implement monetary policy; core operations (liquidity provision, interest rate steering, collateralized operations) would remain intact and could be enhanced via automation, programmability, and composability.
  - Automated and instantaneous settlement could enhance capital allocation and affect liquidity dynamics across money and asset markets (Agur and others 2025).
  - Expanding access to tokenized reserves (e.g., to NBFIs) or enabling new wholesale use cases could change demand for wholesale central bank money and require adjustments to liquidity management practices.
- Two illustrative scenarios (overview; Scenario 1 detailed)
  - Scenario 1: Only tokenized reserves exist as wholesale central bank money
    - Assumes tokenization widely adopted and central bank fully replaces traditional reserves with tokenized reserves for interbank settlements.
    - Monetary policy implementation and market conventions need redesign for continuous, automated, composable settlement environments.
    - Programmability and 24/7 availability could allow monetary policy instruments to operate faster or beyond standard business hours.12
    - Traditional market conventions (end-of-day cutoffs, day counts for interest calculation) may no longer apply; central banks must redesign operational frameworks (reserve requirements, standing facilities) for continuous-time systems.
    - Tokenized monetary policy instruments could enhance flexibility and efficiency:
      - Smart contracts allow modular upgrades and rapid implementation changes (Schär 2021; BIS Innovation Hub and FRBNY 2025).13
      - Smart contracts could automatically and continuously optimize liquidity management or trigger fine-tuning facilities based on predefined parameters (market rate volatility, system liquidity).
      - Central banks might issue tokenized central bank bonds on the same ledger as tokenized reserves to support efficient liquidity absorption.
      - Commercial banks could automatically adjust reserve balances in real time to meet reserve requirements or manage intraday liquidity through tokenized collateral.
    - Operational benefits must be integrated with human oversight to ensure sound judgment and policy alignment.

*Source: Authors.*

### Box 4. Experimentation of Distributed Ledger Technology-B ased Monetary

### Box 4. Experimentation of Distributed Ledger Technology-Based Monetary Policy Implementation

### Prototypes and live pilots
- Project Pine (2025)
  - Collaboration: Federal Reserve’s New York Innovation Center and the BIS Innovation Hub.
  - Purpose: Developed a prototype for implementing monetary policy in a fully tokenized environment.
  - Demonstrated capabilities:
    - Tokenized reserves and securities facilitating monetary policy implementation.
    - Use of smart contracts to automate core functions.
  - Explored four monetary tools:
    1. interest-bearing reserves,
    2. open market operations,
    3. collateral management,
    4. asset purchases.
  - Included a smart contract toolkit testing functions such as creating facilities and exchanges and managing collateral.
  - Note: Project Pine experimented using smart contracts for central bank operations and showed that many smart contracts for automating asset servicing required both money and securities to be tokenized since the smart contracts needed to interact with both.
- Project Helvetia Phase III — Digital SNB Bill (June 2024)
  - The Swiss National Bank (SNB) executed a live monetary policy operation using DLT.
  - Issuance: 64 million Swiss francs in 7-day digital SNB bills, settled in tokenized reserves on the SIX Digital Exchange platform.
  - Demonstrated benefits: potential for improving transparency and internal workflows.
  - Revealed challenge: cash management complexity — tokenized reserves had to be exchanged for Swiss Interbank Clearing balances as part of tokenization before settling bill payments.
- Sources cited in the box: Bank for International Settlements 2025; Gerosa, Gloede, and Müller 2024.

### Collateralized transactions and tokenization: benefits and risks
- Potential enhancements from integrating tokenized reserves and tokenized assets used as collateral:
  - Embedding eligibility checks, settlement logic, and risk parameters directly into smart contracts.
  - Automation and improvement of collateral management processes, including:
    - real-time margin calls,
    - dynamic collateral substitution,
    - execution of collateralized payments through atomic DvP settlements.
  - Enabling on-demand liquidity access by allowing collateral to be used on a “just-enough” basis, reducing opportunity cost of locking assets as collateral (Bank of Thailand 2018; Lee, Martin, and Townsend 2024; Agur and others 2025).
  - Transforming less liquid, complex, or non-standardized assets—such as loan portfolios—into tradable digital tokens to improve tradability, liquidity, price discovery, and asset valuation.
    - Caveat: tokenization does not alter fundamental asset characteristics or automatically render assets eligible as collateral.
- New or amplified risks requiring rigorous evaluation and redesigned controls:
  - Cybersecurity risks.
  - Operational risks from erroneous smart contracts.
  - Contagion risks from close integration of collateral and reserves.
  - Need for redesigned risk management practices, including appropriate haircuts and a legal framework to recognize tokenized assets as eligible collateral (BIS and CPMI 2024).

### Scenario 2 — Coexistence of tokenized and traditional reserves
- Context: central bank issues tokenized reserves alongside traditional reserves.
- Operational challenges:
  - Liquidity fragmentation from central bank money split across two reserve pools (RTGS and DLT-based).
  - If not seamlessly interchangeable, liquidity may become siloed, complicating liquidity management to steer short-term interest rates.
  - Design/policy differences (e.g., remuneration, access) could incentivize shifts in balances and create volatility (Kunaratskul, Reslow, and Singh 2024).
  - Settlement integration shortcomings could hinder coordinated liquidity provision during stress.
- Policy and operational safeguards recommended:
  - Coordinated liquidity monitoring.
  - Improved liquidity forecasting.
  - Interoperability for seamless transfers between reserve types.
  - Transitional measure: allow only intraday balances of tokenized reserves, requiring end-of-day conversion back to traditional reserves.
    - Note: Tokenized reserves’ experiments including Project Jura and Project Helvetia followed such an intra-day setup.
  - Legal and economic equivalence (harmonized access, remuneration, and regulatory treatment) to preserve fungibility.
    - Example: overnight tokenized reserves should earn the same interest and count equally toward reserve requirements to avoid reserve demand volatility.
- Policy choice on where to implement monetary policy:
  - If tokenized segment grows, central banks could operate in both reserve systems to ensure effective policy transmission.

### Operating models, governance, and programmability
- Smart contract functionality depends on the ledger’s programmability standard; flexibility affects speed of updating policy logic.
- Governance determines how easily programmability standards can evolve.
- Operating models and implications for central bank control and flexibility:
  - Integration model: central bank retains full authority to write, deploy, and update smart contracts — rapid adjustments and custom features possible.
  - Distribution model: significant influence by central bank but coordination with partners required — slower updates.
  - Separation model: central bank depends on third-party operators for changes — reduced flexibility.
  - Permissionless model: protocol updates require broad network consensus — slow responsiveness but broader innovation potential.

### Alternative solutions for settling tokenized assets
- Real-Time Gross Settlement (RTGS) Link
  - Also called “a trigger solution” or “synchronization.”
  - Creates a technical bridge or synchronized operator to coordinate tokenized asset delivery on DLT with payment/settlement in traditional central bank reserves via RTGS.
  - Two legs (asset transfer and payment) are mutually conditional using smart contracts (DvP).
  - Examples: Bundesbank’s “Trigger Solution” (Deutsche Bundesbank 2021), Banca d’Italia’s “TIPS Hash-Link” proof of concept (Banca d’Italia 2023), Bank of England’s “RT2” (Bank of England 2025).
- Omnibus Account
  - Implemented by the Bank of England in 2021 for settlement of tokenized payments on private programmable asset platforms, fully funded by reserves (Bank of England 2021).
  - Licensed FMI operating a permissioned DLT platform holds funds in an omnibus account with the central bank; FIs prefund omnibus accounts and the FMI issues private money tokens for on-platform settlement.
  - Enables settlement using tokens outside RTGS hours and with added programmability, provided prefunding.
  - Legal arrangements vary; funds in the omnibus account may legally belong to the FMI or its participants (Bank of Canada and others 2025).
  - This Note considers tokenized money issued against omnibus account funds as settlement in private money, aligned with Bank of Canada and others (2025).
- Privately Issued Tokenized Money
  - Private institutions (commercial banks) or NBFIs (fintechs) issue tokenized money denominated in local currency for programmable wholesale settlement.
  - Three major types described:
    - Tokenized deposits: nontransferable claims on issuing bank; intra-bank transfers settle internally; interbank transfers require additional interbank settlement step.
    - Deposit tokens (“balance-sheet-backed” stablecoins): transferable between institutional clients of different banks; claims on issuing bank’s balance sheet; value can deviate from par; token transfers between holders do not affect bank balance sheets except at issuance/redemption (Garratt and Shin 2023).
    - Stablecoins (fiat-backed focus in this Note): transferable; backed by “reserve” assets (financial assets, commodities, or other crypto assets) to maintain stability; can be issued by nonbanks; this Note focuses on fiat-backed stablecoins denominated in existing currencies and backed by financial assets in the same currencies.
  - Ledger models: privately issued tokenized money can operate on single ledgers with tokenized assets or in compatible ledger arrangements (IMF 2024).

### Comparing tokenized reserves to alternative settlement solutions
- Core comparison dimensions: use of central bank money, ability to support atomic settlement, and ability to support DvP settlement.
- All six solutions discussed support DvP settlement.
- Atomic settlement:
  - Tokenized reserves and assets on a single ledger, omnibus account solutions, and private tokenized money on single ledgers enable atomic settlement.
  - Tokenized reserves on compatible ledgers, RTGS links, and private tokenized money on separate ledgers do not offer strict atomicity (they may approach it via techniques like hash time-locked contracts, but not strict atomicity).
- Credit and liquidity risk:
  - Tokenized reserves and RTGS links settle in central bank money — the safest settlement asset.
  - Omnibus accounts backed by funds at the central bank reduce credit risk but rely on robust oversight of the operator.
  - Privately issued tokenized money without direct central bank backing carries higher credit and liquidity risks reflecting issuer creditworthiness and underlying asset quality (OECD 2025).
  - Stablecoins’ risk depends on backing assets; if backed solely by central bank reserves, they would resemble omnibus solutions; otherwise they may be volatile.
  - Rehypothecation practices by stablecoin issuers can increase leverage and amplify risks (Adrian and others, forthcoming).
- Table 1 capabilities summarized (solutions and Yes/No across three goals)
  - Tokenized reserves—single ledger: DvP Settlement? Y ; Atomic Settlement? Y ; Central Bank Money Settlement? Y
  - Tokenized reserves—compatible ledgers: DvP Settlement? Y ; Atomic Settlement? N ; Central Bank Money Settlement? Y
  - RTGS link: DvP Settlement? Y ; Atomic Settlement? N ; Central Bank Money Settlement? Y
  - Omnibus account: DvP Settlement? Y ; Atomic Settlement? Y ; Central Bank Money Settlement? N
  - Private tokenized money—single ledger: DvP Settlement? Y ; Atomic Settlement? Y ; Central Bank Money Settlement? N
  - Private tokenized money—compatible ledgers: DvP Settlement? Y ; Atomic Settlement? N ; Central Bank Money Settlement? N
  - Note: Atomic settlement refers to “strict” atomicity and not “weak” atomicity. For omnibus accounts, some jurisdictions might consider final settlement in central bank money; treatment affects the atomicity and central bank money designation.

### Charting exploratory work and strategic approaches
- Central banks may pursue multiple solutions; sequencing can prioritize short-term, lower-cost options while developing complex solutions later (example: European Central Bank plans to first implement an RTGS link, with tokenized reserves as a possible next step).
- Prioritization framework: two-dimensional assessment — y-axis suitability (alignment with policy goals), x-axis feasibility (ease of implementation: time, complexity, resources).
- Four strategic approaches for engaging with tokenized asset markets:
  1. Inaction — deem initiatives not relevant or feasible now.
  2. Wait and see — monitor developments, gather data, consult stakeholders, conduct legal reviews or small-scale proofs of concept.
  3. Enablement — establish foundations for market-led developments via research, infrastructure, and regulatory frameworks; use pilots or prototypes; consider short-term solutions such as an RTGS link.
  4. Catalyst — proactive engagement to stimulate market adoption (e.g., develop RTGS links or make reserves available on DLT infrastructure); may include issuance of tokenized central bank, government, or corporate bonds.
- Complementary activities across strategies:
  - Legal and regulatory work to establish legal basis, clarity, and safety measures.
  - Ongoing research into DLT technical capabilities, macrofinancial implications, cybersecurity, and legal risks.
  - Monitor international developments and participate in global dialogue.
- Implementation guidance:
  - Use IMF’s “5P methodology” for project management if pursuing tokenized reserve issuance: proof of concept, prototype, pilot, and potentially production.
  - Analyze local market challenges to judge whether tokenization addresses specific goals or pain points; strategies and policy choices will vary by jurisdiction.

### Conclusion — policy questions and guidance
- Key policy questions highlighted:
  - If tokenization is widely adopted, should central banks issue tokenized reserves on DLT for payments and settlements of tokenized assets?
  - Which implementation approach best meets policy objectives and what factors must be considered?
  - How do alternative solutions (RTGS link, privately issued tokenized money) compare with tokenizing reserves?
  - Which strategic approach should a central bank adopt given policy direction, market demand, and capacity?
- Guidance emphasis:
  - Provide a framework for comparing tokenized reserves with alternatives for DvP-based settlement.
  - Discuss implementation considerations, institutional constraints, market trends, and central bank views on tokenization and DLT.
  - Outline four strategic approaches (inaction, wait and see, enablement, catalyst) to support internal alignment and transparency.
  - Stress balanced assessment of potential benefits and risks based on country context, legal requirements, resource constraints, and evidence-based policymaking.
  - Reiterate that DLT and tokenized reserves are complements to, not substitutes for, strong institutions, sound policies, and robust legal frameworks.

*Source: Box 4. Experimentation of Distributed Ledger Technology-Based Monetary Policy Implementation (FINTECH NOTES Central Bank Exploration of Tokenized Reserves, INTERNATIONAL MONETARY FUND).*

### References

### ftnea2025011 - References

### Tokenization, Stablecoins, and Tokenized Markets
- Adrian, Tobias, Parma Bains, Marianne Bechara, Eugenio Cerutti, Stephanie Forte, Federico Grinberg, Alessandro Gullo, Martina Hengge, Agnija Jekabsone, Kathleen Kao, Tommaso Mancini Griffoli, Soledad Martinez Peria, Marcello Miccoli, Marco Reuter, and Nobuyasu Sugimoto. Forthcoming. “Understanding Stablecoins.” IMF Departmental Paper.
- Agur, Itai, Germán Villegas-Bauer, Tommaso Mancini-Griffoli, Maria Soledad Martinez Peria, and Brandon Tan. 2025. “Tokenization and Financial Market Inefficiencies.” IMF Fintech Notes NOTE/2025/001, January. https://www.imf.org/en/-/media/files/publications/ftn063/2025/english/ftnea2025001.pdf
- Leung, Victor, Joe Wong, Alexander Ying, and Wilson Wan. 2023. “An Assessment on the Benefits of Bond Tokenisation.” Research Memorandum 04/2023, Hong Kong Monetary Authority, 28 November. https://www.hkma.gov.hk/media/eng/publication-and-research/research/researchmemorandums/2023/RM04-2023.pdf
- Lee, Michael Junho, Antoine Martin, and Robert M. Townsend. 2024. “Optimal Design of Tokenized Markets.” Federal Reserve of New York Staff Report, no. 1121, September.
- Budau, Victor, and Herve Tourpe. 2024. “ASAP: A Conceptual Model for Digital Asset Platforms.” IMF Working Paper No. 204/019, February. https://www.imf.org/en/-/ /media/files/publications/wp/2024/english/wpiea2024019-print-pdf.pdf
- OECD. 2025. “Tokenisation of Assets and Distributed Ledger Technologies in Financial Markets: Potential Impediments to Market Development and Policy Implications.” OECD Business and Finance Policy Papers, No. 75, OECD Publishing, Paris. https://doi.org/10.1787/40e7f217-en
- BIS and CPMI. 2024. “Tokenisation in the Context of Money and Other Assets: Concepts and Implications for Central Banks.” Bank of International Settlements and Committee on Payments and Market Infrastructures, Joint Report, October. https://www.bis.org/cpmi/publ/d225.pdf
- BIS Innovation Hub. 2025. “Project Agora: Exploring Tokenisation of Cross-Border Payments.” https://www.bis.org/about/bisih/topics/fmis/agora.htm

### Central Bank Digital Currency (CBDC), Wholesale CBDC, and Central Bank Money Innovation
- Cambride Centre for Analytical Finance. 2024. “Wholesale Central Bank Digital Currencies (wCBDCs): Approaches, Implementation Strategies, and Use Cases.” Judge School of Business and Cambridge Digital Assets Programme. https://www.jbs.cam.ac.uk/2024/report-exploring-wholesale-central-bank-digital-currencies/
- Bank of Canada, European Central Bank, Bank of Japan, Sveriges Riksbank, Swiss National Bank, Bank of England, Board of Governors Federal Reserve System, and Bank for International Settlements. 2025. “Wholesale Central Bank Money in the Context of Technological Innovation.” September.
- Reserve Bank of Australia. 2024. “Project Acacia—Exploring the Role of Digital Money in Wholesale Tokenized Asset Markets.” Consultation Paper, November.
- Gerosa, Rebecca, Oliver Gloede, and Philipp Müller. 2024. “Piloting Monetary Policy Implementation on a DLT-Based Infrastructure — Issuance of Digital SNB Bills.” Economic Note No. 04/2024. Zurich: Swiss National Bank. https://www.snb.ch/en/publications/research/economic-notes/2024/economic_note_2024_04
- Kunaratskul, Tansaya, André Reslow, and Manmohan Singh. 2024. “Implications of Central Bank Digital Currency on Monetary Operations.” IMF Fintech Note No 2024/007, October. https://www.imf.org/en/Publications/fintech-notes/Issues/2024/10/04/Implications-of-Central-Bank-Digital-Currency-for-Monetary-Operations-555883
- Patel, Manisha, Safari Kasiyanto, and André Reslow. 2024. “Positioning Central Bank Digital Currency in the Payments Landscape.” IMF Fintech Note No 2024/006, October. https://www.imf.org/en/Publications/fintech-notes/Issues/2024/10/02/Positioning-Central-Bank-Digital-Currency-in-the-Payments-Landscape-555743
- Tourpe, Herve, Ashley Lannquist, and Gabriel Soderberg. 2023. “A Guide to Central Bank Digital Currency Product Development.” Fintech Note No 2023/007, September. https://www.imf.org/en/Publications/fintech-notes/Issues/2023/09/08/A-Guide-to-Central-Bank-Digital-Currency-Product-Development-538496
- Illes, Anamaria, Anneke Kosse, and Peter Wierts. 2025. “Advancing in Tandem—Results of the 2024 BIS Survey on Central Bank Digital Currency and Crypto.” BIS Papers No. 159, August 22. https://www.bis.org/publ/bppdf/bispap159.htm
- Cambride Centre for Analytical Finance. 2024. “Wholesale Central Bank Digital Currencies (wCBDCs): Approaches, Implementation Strategies, and Use Cases.” Judge School of Business and Cambridge Digital Assets Programme. https://www.jbs.cam.ac.uk/2024/report-exploring-wholesale-central-bank-digital-currencies/

### Distributed Ledger Technology, Settlement, and Market Infrastructures
- BIS and CPMI. 2017. “Distributed Ledger Technology in Payment, Clearing and Settlement—An Analytical Framework.” Bank of International Settlements and Committee on Payments and Market Infrastructures, February. https://www.bis.org/cpmi/publ/d157.pdf
- Neuhaus, Holger, and Mirjam Plooij. 2023. “Central Bank Money Settlement of Wholesale Transactions in the Face of Technological Innovation.” ECB Economic Bulletin, Issue 8/2023.
- Deutsche Bundesbank. 2021. “DLT-Based Securities Settlement in Central Bank Money Successfully Tested.” Press Release, March 24. https://www.bundesbank.de/en/press/press-releases/dlt-based-securities-settlement-in-central-bank-money-successfully-tested-861444
- European Central Bank. 2025. “Eurosystem Expands Initiative to Settle DLT-Based Transactions in Central Bank Money.” Press Release, 20 February 2025, ECB, Frankfurt am Main, Germany.
- BIS. 2025. “Project Pine: Central Bank Open Market Operations with Smart Contracts.” Basel: BIS. https://www.bis.org/publ/othp95.pdf
- BIS Innovation Hub and Federal Reserve Bank of New York (FRBNY). 2025. “Project Pine: Central Bank Open Market Operations with Smart Contracts,” BIS Innovation Hub. https://www.bis.org/about/bisih/topics/fmis/pine.htm
- Lee, Michael Junho, Antoine Martin, and Benjamin Müller. 2022. “What Is Atomic Settlement?.” Liberty Street Economics, 7 November. https://libertystreeteconomics.newyorkfed.org/2022/11/what-is-atomic-settlement/
- Cabedo, Yaiza, Tommaso Mancini-Griffoli, Fabian Schär, and Nicolas Xuan-Yi Zhang. Forthcoming. “Financial Market Infrastructures Evolution in a Tokenized Economy.”
- Budau, Victor, and Herve Tourpe. 2024. “ASAP: A Conceptual Model for Digital Asset Platforms.” IMF Working Paper No. 204/019, February. https://www.imf.org/en/-/ /media/files/publications/wp/2024/english/wpiea2024019-print-pdf.pdf

### Policy, Access, and Regulatory Frameworks
- Bank of England. 2024a. “The Bank of England’s Approach to Innovation in Money and Payments.” Discussion Paper, July. https://www.bankofengland.co.uk/paper/2024/dp/the-boes-approach-to-innovation-in-money-and-payments
- Bank of England. 2024b. “Reviewing Access to RTGS Accounts for Settlement.” Discussion Paper, February. https://www.bankofengland.co.uk/paper/2024/dp/reviewing-access-to-rtgs-accounts-for-settlement
- Bank of England. 2021. “Bank of England Publishes Policy for Omnibus Accounts in RTGS.” News Release, 19 April.
- Bank of England. 2025. “Synchronisation.” https://www.bankofengland.co.uk/payment-and-settlement/rtgs-future-roadmap/what-is-synchro
- Banca d’Italia. 2023. “TIPS Hash-Link Service Description Overview.” European Central Bank Focus Session, December 15.
- BIS. 2021. “Building Block 10: Improving (Direct) Access to Payment Systems by Banks, Non-Banks and Payment Infrastructures.” Bank of International Settlements. https://www.bis.org/cpmi/cross_border/bb10.htm
- Wendt, Froukelien, Peter Katz, and Alice Zanza. 2019. “How to Organize Central Securities Depositories in Developing Markets: Key Considerations,” IMF How to Notes No. 2019/001, International Monetary Fund, Washington, DC.
- World Bank. 2024. “Granting Access to Real-Time Gross Settlement Systems for Nonbank Payment Service Providers: Implications for Fast Payments.” Project FAST, February. https://fastpayments.worldbank.org/sites/default/files/202402/Non%20Bank%20RTGS%20Access_Focus%20Note_Final.pdf
- IMF. 2020. “Chile: Technical Assistance Report—Central Bank Services to Non-Bank Financial Institutions.” IMF Staff Country Reports, May 13. https://www.imf.org/en/Publications/CR/Issues/2020/05/13/Chile-Technical-Assistance-Report-Central-Bank-Services-to-Non-bank-Financial-Institutions-49411
- IMF. 2024. “G-20 Note on Financial Platforms: What Are They and What Are Their Macro-Financial Implications?.” IMF Staff G-20 Note, October 2024. https://www.imf.org/en/-/media/files/research/imf-and-g20/2024/g20-report-2024-financial-platforms-macrofinancial-implications-imf-oct2024-final-board-publish.pdf

### Central Bank Speeches, Projects, and National Initiatives
- Jordan, Thomas. 2024. “Towards the Future Monetary System.” Speech at SNB Conference, April 8. https://www.snb.ch/en/publications/communication/speeches/2024/ref_20240408_tjn
- Maechler, M. Andréa, and Thomas Moser. 2023. “Swiss Payments Vision – An Ecosystem for Future-Proof Payments.” Speech at Swiss National Bank’s Money Market Event, March 30. https://www.snb.ch/en/publications/communication/speeches/2023/ref_20230330_amrtmo
- Hong Kong Monetary Authority. 2024. “HKMA Launches Project Ensemble Sandbox to Facilitate Interbank Settlement Use Cases.” Press Release, August 28.
- Danmarks Nationalbank. 2025. “New Technology and Settlement in Central Bank Money between Banks.” Analysis: Digitalization, No. 14, June 17.
- Sveriges Riksbank. 2024. “Investigation into the Riksbank’s Future Settlement Services.” https://www.riksbank.se/en-gb/payments--cash/the-payment-system---rix/ongoing-studies-and-projects/study-of-the-riksbanks-future-settlement-service/
- Bank of Ghana. 2024. “Bank of Ghana Announces Successful Completion of Cross-Border Trade Using Digital Credentials.” Press Release, June 12.

### Research on Stablecoins, DeFi, Payments Frictions, and Capital Flows
- Garratt, Rodney, and Hyun Song Shin. 2023. “Stablecoins versus Tokenised Deposits: Implications for the Singleness of Money.” BIS Bulletin No. 73, April 11. https://www.bis.org/publ/bisbull73.htm
- Schär, Fabian. 2021. “Decentralized Finance: On Blockchain- and Smart Contract-Based Financial Markets.” Federal Reserve Bank of St. Louis Review. https://berkeley-defi.github.io/assets/material/Fabian-Schar-decentralized-finance-on-blockchain-and-smart-contract-based-financial-markets.pdf
- Reuter, Marco, Itai Agur, Alexander Copestake, Maria Soledad Martinez Peria, and Ken Teoh. 2025. “Payment Frictions, Capital Flows, and Exchange Rates.” IMF Working Paper WP/25/171, August.
- He, Dong, Annamaria Kokenyne, Tommaso Mancini Griffoli, Marcello Miccoli, Thorvardur Tjoervi Olafsson, Gabriel Soderberg, and Herve Tourpe. 2023. “Capital Flow Management Measures in the Digital Age (2): Design Choices for Central Bank Digital Currency.” IMF Fintech Note 2023/009, International Monetary Fund, Washington, DC.
- Reslow, André, Gabriel Soderberg, and Natsuki Tsuda. 2024. “Cross-Border Payments with Retail Central Bank Digital Currencies.” IMF Fintech Note 2024/002, May. https://www.imf.org/en/publications/fintech-notes/issues/2024/05/15/cross-border-payments-with-retail-central-bank-digital-currencies-547195
- Garratt, Rodney, and Hyun Song Shin. 2023. “Stablecoins versus Tokenised Deposits: Implications for the Singleness of Money.” BIS Bulletin No. 73, April 11. https://www.bis.org/publ/bisbull73.htm

*Central Bank Exploration of Tokenized Reserves — NOTE 2025/011*

---


_Source: https://www.imf.org/-/media/files/publications/ftn063/2025/english/ftnea2025011.pdf_