Cross-Border & Interoperability

1. Global Landscape — Cross-Border CBDC Models

Cross-border payments remain one of the most frequently cited motivations for CBDC development, particularly among emerging market and developing economies. The BIS 2024 survey found that 91% of 93 surveyed central banks are exploring CBDCs, with cross-border payment efficiency ranking as an increasingly important driver. The IMF confirms that wholesale CBDC explorations are gaining prominence, with cross-border platforms at the forefront.

Source: BIS Papers No. 159, "Advancing in tandem — results of the 2024 BIS survey on central bank digital currencies and crypto," August 2025. https://www.bis.org/publ/bppdf/bispap159.htm

How the 20 jurisdictions distribute across cross-border activity (N=20):

• 2 of 20 are full multi-CBDC platform participants in a live wholesale arrangement (mBridge):
  China and
  Thailand (both founding members).
• 4 of 20 sit in the BIS Project Agorá cohort (the post-mBridge, Western-aligned flagship):
  Japan,
  South Korea, and the
  EU (via Banque de France representing the Eurosystem) are among the seven Agorá central banks;
  Canada (Bank of Canada) joins the real-value testing phase confirmed by the 27 May 2026 first-phase report.
• 3 of 20 are Project Dunbar alumni:
  Australia,
  Singapore, and
  South Africa (the only African participant — Nigeria did not take part).
• 1 of 20 led Project Icebreaker:
  Sweden (with non-CBDC-domain partners Norges Bank and Bank of Israel).
• Singapore is the single most cross-border-active jurisdiction of the 20 — Jasper-Ubin, Ubin+, Dunbar, Cedar×Ubin+, Mariana, Mandala, Agorá, and the Swift CBDC sandbox — yet, notably, is not an mBridge participant.
• India made its first confirmed BIS multilateral entries this cycle — Project Rialto and Project Mandala Phase 2 (RBI Annual Report 2025-26) — alongside UPI international linkages.
• At least 5 of 20 are domestic-only / have no operative cross-border CBDC corridor:
  Bahamas (Sand Dollar),
  Jamaica (JAM-DEX),
  Nigeria (eNaira — inbound-remittance capability approved but minimal uptake),
  the UK (cross-border interoperability explicitly excluded from the Oct 2025 digital-pound design notes), and
  Ghana (ECOWAS cross-border design announced 22 May 2026 but no live corridor).
• Eastern Caribbean's DCash was inherently cross-border (the only CBDC to operate natively across multiple sovereign territories — 8 ECCU states), but DCash 2.0 was suspended 13 Feb 2026 in favour of a Fast Payment System + CARICOM (CAPS) pivot.
• The US is the only jurisdiction to have formally withdrawn from active mBridge participation (over sanctioned-nation concerns) while remaining the anti-CBDC outlier; it retains a NY-Fed wholesale-research footprint (Project Cedar) and the NY Fed seat in Agorá.

Five distinct architectural models have emerged from experimentation to date. Each represents a fundamentally different approach to the governance, technology, and economic structure of cross-border CBDC settlement.

---

1.1 Model A: Multi-CBDC Common Platform (mBridge)

Concept: Multiple central banks issue their respective wholesale CBDCs onto a single shared distributed ledger platform. Participating commercial banks transact directly with each other using these CBDCs for cross-border payments and FX settlement.

Defining characteristics:

• Single shared ledger operated jointly by participating central banks
• Each central bank issues and redeems its own CBDC on the platform
• Commercial banks hold multi-currency CBDC wallets
• Real-time, peer-to-peer cross-border payments without correspondent banking intermediaries
• Atomic payment-versus-payment (PvP) FX settlement

Primary example: Project mBridge — People's Bank of

China (Digital Currency Institute), Hong Kong Monetary Authority, Bank of

Thailand, Central Bank of the UAE, Saudi Central Bank (SAMA). The BIS Innovation Hub co-developed the platform before withdrawing in October 2024.

Outcomes: mBridge has reportedly settled 4,047 cross-border transactions with a cumulative value of approximately USD 55.49 billion (as of November 2025) — self-reported and not independently audited. Transactions that previously took 3-5 days via correspondent banking settle in seconds.

Status: Operational post-MVP. BIS withdrew October 2024; platform now governed independently by participating central banks.

Sources:

• BIS, "Project mBridge reached minimum viable product stage." https://www.bis.org/about/bisih/topics/cbdc/mcbdc_bridge.htm
• BIS, "Project mBridge: connecting economies through CBDC." https://www.bis.org/publ/othp59.htm

---

1.2 Model B: Interlinking Domestic Systems — Hub-and-Spoke (Icebreaker)

Concept: Each central bank operates its own domestic CBDC system autonomously. A central "hub" coordinates cross-border transactions by linking these independent systems. A cross-border payment is decomposed into two domestic CBDC payments connected through the hub, with FX conversion performed by liquidity providers active in both systems.

Defining characteristics:

• No shared ledger — each CBDC stays within its domestic system
• The hub coordinates but does not hold funds
• FX providers bridge the two legs of the transaction
• Coordinated payment-versus-payment settlement using Hash Time Locked Contracts (HTLCs)
• Maximum preservation of central bank autonomy over domestic CBDC design

Primary example: Project Icebreaker — Sveriges Riksbank (

Sweden), Norges Bank (Norway), Bank of Israel, BIS Innovation Hub Nordic Centre. Completed March 2023.

Outcomes: Demonstrated that retail CBDC systems can be interlinked for cross-border payments without requiring a common platform. Cross-border transactions completed within seconds. Central banks retain "almost full autonomy" in domestic CBDC design while participating in the interlinking arrangement.

Status: Completed. Report published March 2023. No production deployment. Sweden's e-Krona cross-border trajectory is now closely tied to the EU digital euro timeline.

Source: BIS, "Project Icebreaker: breaking new paths in cross-border retail CBDC payments," March 2023. https://www.bis.org/publ/othp61.htm

---

1.3 Model C: Single-Platform DvP / PvP (Jura)

Concept: Two central banks issue their respective wholesale CBDCs on a third-party DLT platform operated by an independent entity. The platform enables both delivery-versus-payment (tokenised asset against wCBDC) and payment-versus-payment (wCBDC against wCBDC) between commercial banks domiciled in different jurisdictions.

Defining characteristics:

• Third-party operated platform (not operated by any single central bank)
• Subnetworks and dual-notary signing give each central bank governance comfort
• Regulated non-resident financial institutions can access wholesale CBDCs
• Enables both DvP (securities settlement) and PvP (FX settlement)

Primary example: Project Jura — Banque de France (BdF, for the

euro area), Swiss National Bank (SNB), BIS Innovation Hub, with Accenture, Credit Suisse, Natixis, R3, SIX Digital Exchange (SDX), UBS. Completed December 2021.

Transactions tested:

• DvP: Tokenised commercial paper issued under French law against EUR wCBDC
• PvP: EUR wCBDC against CHF wCBDC
• All settled between banks domiciled in France and Switzerland

Outcomes: Successfully demonstrated cross-border wholesale CBDC settlement on a third-party DLT platform. Confirmed that well-designed wCBDC can serve as a "safe and neutral settlement asset for international financial transactions."

Status: Completed. Exploratory; neither BdF nor SNB indicated plans to issue wCBDC based on the experiment.

Sources:

• BIS, "Project Jura: cross-border settlement using wholesale CBDC." https://www.bis.org/about/bisih/topics/cbdc/jura.htm
• BIS Press Release, December 8, 2021. https://www.bis.org/press/p211208.htm

---

1.4 Model D: Unified Ledger — Central + Commercial Bank Money (Agorá)

Concept: A single programmable ("unified") ledger brings together tokenised commercial bank deposits and wholesale CBDC in one integrated environment, so that cross-border wholesale payments settle atomically between central bank money and commercial bank money. Unlike mBridge (central-bank-to-central-bank CBDC exchange), this model places both money types on the same platform.

Defining characteristics:

• Tokenised commercial bank deposits settled against wholesale CBDC on a single ledger
• Strong private-sector orientation (large bank and payment-network involvement)
• Draws on the BIS "unified ledger" / "blueprint for the future monetary system" concept
• Targets correspondent-banking frictions in the cross-border wholesale layer

Primary example: Project Agorá — seven central banks (Banque de France for the

Eurosystem, Bank of

Japan, Bank of

Korea, Bank of Mexico, Swiss National Bank, Bank of England, Federal Reserve Bank of New York) plus ~40 private-sector firms. The 27 May 2026 first-phase report confirmed that atomic multi-currency wholesale settlement on a unified ledger is feasible and advanced the project to real-value testing, with the Bank of

Canada joining for that phase.

Status: Active. First-phase report published 27 May 2026; advancing to real-value experimentation. The post-mBridge flagship for the BIS.

Sources:

• BIS, "Project Agorá: exploring tokenisation of cross-border payments." https://www.bis.org/about/bisih/topics/fmis/agora.htm
• BIS Press Release, April 3, 2024. https://www.bis.org/press/p240403.htm

---

1.5 Model E: Bilateral CBDC Arrangements

Concept: Two jurisdictions establish a direct bilateral corridor for CBDC-based cross-border payments, typically building on existing trade relationships. One jurisdiction's CBDC infrastructure connects to the other's through a dedicated bridge or protocol.

Defining characteristics:

• Bilateral, not multilateral
• Often driven by specific trade corridors (e.g., China-Thailand, China-UAE)
• May serve as precursors to broader multilateral platforms
• Governance is simpler (two parties) but scalability is limited

Primary examples:

• Inthanon-LionRock (
  Thailand-Hong Kong): Bilateral corridor prototype between Bank of Thailand and HKMA. Thai baht-Hong Kong dollar corridor enabling peer-to-peer FX transactions. Served as a direct precursor to mBridge.
• HKMA-BCB (
  Brazil) tokenisation partnership (announced Oct 28, 2024): links the HKMA Ensemble Sandbox with Brazil's Drex; Brazil pursues interoperability through bilateral partnerships over multilateral platforms.
• China-led corridors: Active e-CNY cross-border pilots planned with
  Singapore, Thailand, Hong Kong, UAE, and Saudi Arabia under the PBOC's Shanghai International Operation Center.

Outcomes: Bilateral arrangements serve as effective stepping stones toward multilateral platforms. Inthanon-LionRock directly led to mBridge. However, bilateral approaches do not scale efficiently — each new corridor requires separate negotiation, legal framework, and technical integration.

Status: Most early bilateral CBDC arrangements have been absorbed into multilateral projects (principally mBridge). China's Shanghai International Operation Center and the HKMA-BCB partnership represent the current bilateral push.

Sources:

• PBOC, "e-CNY International Operation Center Officially Launched in Shanghai," September 24, 2025. https://www.pbc.gov.cn/en/3688110/3688172/5552468/5855057/index.html
• BIS, "Project mBridge: connecting economies through CBDC." https://www.bis.org/publ/othp59.htm

2. BIS Innovation Hub & Multilateral Projects Deep Dive

The BIS Innovation Hub has led or co-led more CBDC cross-border experiments than any other institution. This section provides detailed analysis of each major project and maps it to the 20 jurisdictions in this domain.

---

2.1 Project mBridge

Full name: Multiple CBDC Bridge

Participants:

• Digital Currency Institute of the People's Bank of
  China (PBOC/DCI) — founding member
• Hong Kong Monetary Authority (HKMA) — founding member
• Bank of
  Thailand (BOT) — founding member
• Central Bank of the United Arab Emirates (CBUAE) — founding member
• Saudi Central Bank (SAMA) — joined 2024 as full participant
• BIS Innovation Hub Hong Kong Centre — withdrew October 2024
• Observer / non-participating relationships among the 20:
  South Korea (observer status);
  India (interest expressed, not a participant);
  United States (withdrew from active participation over sanctioned-nation concerns).
  Japan,
  Singapore, the
  EU, and
  Russia are explicitly not mBridge participants.
  Brazil mBridge participation is NOT VERIFIED.

Technology — The mBridge Ledger:

• Purpose-built distributed ledger technology designed specifically for the project
• Compatible with the Ethereum Virtual Machine (EVM) for interoperability
• Supports real-time, peer-to-peer cross-border payments and FX transactions
• Commercial banks from China, Thailand, UAE, Hong Kong, and Saudi Arabia participated in testing

Timeline:

Governance after BIS withdrawal: The project operates under a bespoke governance and legal framework, including a rulebook tailored to the platform's decentralised nature. Following the BIS withdrawal, the participating central banks (led by PBOC/DCI and CBUAE) govern the platform independently. BIS General Manager Agustín Carstens stated publicly that "mBridge is not the BRICS Bridge," seeking to distance the BIS from geopolitical speculation.

Key outcomes:

• Demonstrated that multi-CBDC platforms can address cross-border payment inefficiencies: high costs, slow speeds, operational complexity, and correspondent banking retreat
• Transactions settling in seconds versus days
• China's e-CNY reportedly accounts for over 95% of total settlement volume on the platform (self-reported, not independently audited)
• Platform is technically capable of settling oil and commodity trades in digital national currencies, bypassing the SWIFT network (active use UNVERIFIED)

Sources:

• BIS, "Project mBridge reached minimum viable product stage." https://www.bis.org/about/bisih/topics/cbdc/mcbdc_bridge.htm
• BIS Press Release, June 5, 2024. https://www.bis.org/press/p240605.htm

---

2.2 Project Agorá

Full name: Project Agorá (from Greek agorá, "marketplace")

Participants — Central banks (7):

• Banque de France (representing the
  Eurosystem)
• Bank of
  Japan
• Bank of
  Korea
• Bank of Mexico
• Swiss National Bank
• Bank of England (see
  United Kingdom)
• Federal Reserve Bank of New York (see
  United States)

Joining for real-value testing: Bank of

Canada (per the 27 May 2026 first-phase report).

Private sector (~40 institutions): Including Visa, Swift, Citi, JPMorgan, UBS, Mastercard, MUFG, and others.

Concept: Explores how tokenised commercial bank deposits can be seamlessly integrated with wholesale CBDC on a unified ledger to improve cross-border wholesale settlement. Unlike mBridge (which focuses on central-bank-to-central-bank CBDC exchange), Agorá brings together central bank money and commercial bank money on a single platform.

Technology: Unified ledger concept — drawing on the BIS "blueprint for the future monetary system" (BIS AER 2023, Chapter III). Tokenised deposits are settled against wholesale CBDC in an integrated environment.

Timeline:

Significance: Agorá is the post-mBridge flagship for the BIS. Its participant list — including the Federal Reserve Bank of New York and Bank of England — positions it as the Western-aligned counterpart to mBridge. The inclusion of ~40 private-sector firms signals a stronger commercial orientation, and the May 2026 advance to real-value testing makes it the most operationally advanced wholesale-with-commercial-money cross-border experiment.

Sources:

• BIS, "Project Agorá: exploring tokenisation of cross-border payments." https://www.bis.org/about/bisih/topics/fmis/agora.htm
• BIS Press Release, April 3, 2024. https://www.bis.org/press/p240403.htm

---

2.3 Project Dunbar

Participants:

• Reserve Bank of
  Australia (RBA)
• Bank Negara Malaysia (BNM)
• Monetary Authority of
  Singapore (MAS)
• South African Reserve Bank (SARB) — see
  South Africa
• BIS Innovation Hub Singapore Centre

Timeline: Completed March 2022.

Technology:

• Two prototypes on different DLT platforms: R3 Corda and Partior/Quorum
• Three workstreams: high-level functional requirements and two concurrent technical streams

Key findings:

1. Financial institutions can use CBDCs issued by participating central banks to transact directly with each other on a shared platform
2. Addressed three critical design questions: (a) which entities hold/transact with CBDCs on the platform; (b) how to simplify cross-border flows while respecting regulatory differences; (c) governance arrangements for shared national infrastructure
3. Trust and shared control concerns can be addressed through "governance mechanisms enforced by robust technological means"
4. The approach "lays the foundation for the development of future global and regional platforms"

Significance: Dunbar demonstrated a Southern Hemisphere multi-CBDC model connecting advanced and emerging economies across three continents (Asia-Pacific and Africa). Among the 20, only Australia, Singapore, and South Africa took part —

Nigeria, by contrast, ran no comparable BIS cross-border pilot. South Africa's continued wholesale work (Project Khokha 2x, PK2x regional settlement) builds directly on the Dunbar foundation.

Sources:

• MAS Media Release, March 22, 2022. https://www.mas.gov.sg/news/media-releases/2022/experimental-multi-cbdc-platform-for-international-settlements
• BIS, "Project Dunbar: experimental multi-CBDC platform for international settlements." https://www.bis.org/press/p220322.htm

---

2.4 Project Icebreaker

Participants:

• Sveriges Riksbank (
  Sweden)
• Norges Bank (Norway)
• Bank of Israel
• BIS Innovation Hub Nordic Centre

Timeline: Completed March 2023.

The hub-and-spoke model in detail:

1. Payer initiates a cross-border payment in their domestic rCBDC
2. The "Icebreaker hub" identifies the cheapest FX rate among competing FX providers
3. An FX provider active in both domestic CBDC systems facilitates currency conversion
4. The payer's domestic CBDC is converted to the payee's domestic CBDC
5. Each leg settles in the respective national CBDC system
6. Settlement coordination via Hash Time Locked Contracts (HTLCs) ensures atomic PvP — either both legs complete or neither does

FX model:

• Competitive FX provision: multiple FX providers can quote rates, with the hub selecting the best price
• Transparent pricing for end users
• FX providers bear the currency risk for the brief settlement window
• Tested across three heterogeneous DLT stacks (Quorum / Besu / Corda)

Key outcomes:

• Retail CBDC systems can be interlinked without requiring a common platform
• Central bank monetary sovereignty and domestic CBDC design autonomy fully preserved
• FX conversion and liquidity provision identified as the key operational challenges
• Model is "readily scalable" — adding new jurisdictions requires only connecting their domestic CBDC system to the hub

Critical assessment: Icebreaker's hub-and-spoke model is the most sovereignty-preserving architecture tested. However, it depends on the existence of competitive FX providers willing to operate across multiple CBDC systems — a requirement that may not be met for smaller currency pairs or less liquid corridors. Of the 20, only Sweden is a direct participant; Sweden's own e-Krona path is now tied to the EU digital euro timeline.

Source: BIS, "Project Icebreaker: breaking new paths in cross-border retail CBDC payments," March 2023. https://www.bis.org/publ/othp61.htm

---

2.5 Project Jura

Participants:

• Banque de France (BdF, for the
  euro area)
• Swiss National Bank (SNB)
• BIS Innovation Hub Swiss Centre
• Private sector: Accenture, Credit Suisse, Natixis, R3, SIX Digital Exchange (SDX), UBS

Timeline: Completed December 2021.

Technical innovation — subnetworks and dual-notary signing:

• Subnetworks partition the shared DLT platform into jurisdictional zones, giving each central bank control over its issued wCBDC
• Dual-notary signing requires both the issuing central bank and the platform operator to validate cross-border transfers
• This architecture gives central banks the governance comfort needed to issue wCBDC on a third-party platform while maintaining control

Transactions tested:

• DvP: Tokenised commercial paper (French law) against EUR wCBDC
• PvP: EUR wCBDC against CHF wCBDC
• Settlement between banks domiciled in France and Switzerland

Key outcomes:

• Successfully demonstrated cross-border wholesale CBDC settlement on a third-party DLT platform
• Showed that regulated non-resident financial institutions can access wholesale CBDCs
• Confirmed wCBDC as a "safe and neutral settlement asset for international financial transactions"
• Continues experimentation under Project Helvetia and BdF's wholesale CBDC experiments initiated in 2020

Sources:

• BIS, "Project Jura: cross-border settlement using wholesale CBDC." https://www.bis.org/about/bisih/topics/cbdc/jura.htm
• SIX Digital Exchange, "BdF, SNB, BIS and SIX successfully complete cross-border wCBDC experiment," December 8, 2021. https://www.six-group.com/en/newsroom/media-releases/2021/20211208-bdf-snb-bis-sdx-jura.html

---

2.6 Project Mariana

Participants:

• Banque de France (BdF, for the
  euro area)
• Monetary Authority of
  Singapore (MAS)
• Swiss National Bank (SNB)
• BIS Innovation Hub Swiss, Singapore, and Eurosystem Centres

Timeline: Completed September 2023.

Technology — Automated Market Makers (AMMs):

• AMMs pooled liquidity of hypothetical EUR, SGD, and CHF wCBDCs
• Innovative algorithms enabled spot FX transactions to be priced and executed automatically with immediate settlement
• Deployed on a public blockchain (Ethereum-based)
• Three technical elements: (1) common token standard on public blockchain for interoperability; (2) bridges for seamless wCBDC transfer between networks; (3) AMM for automated FX pricing and execution

Key outcomes:

• Successfully demonstrated automated cross-border wCBDC trading and settlement
• DeFi elements (specifically AMMs) could form the basis for a "new generation of financial market infrastructures"
• Purely experimental; none of the involved central banks indicated intent to issue CBDC or endorse DeFi

Limitations identified:

• The report did not address the "commercial viability" of AMMs in cross-border transactions
• Monetary policy implications of deploying AMMs for sovereign currency FX were not explored
• Liquidity provision in AMM pools raises questions about central bank balance sheet exposure
• Brazil is sometimes cited as relevant to Mariana-style FX experimentation but was not a direct participant

Sources:

• BIS, "Project Mariana." https://www.bis.org/about/bisih/topics/cbdc/mariana.htm
• MAS, "BIS and central banks of France, Singapore and Switzerland successfully test cross-border wholesale CBDCs," September 2023. https://www.mas.gov.sg/news/media-releases/2023/bis-and-central-banks-successfully-test-cross-border-wholesale-cbdcs

---

2.7 Projects Mandala, Rialto & Other Emerging Multilaterals

A newer cohort of BIS Innovation Hub experiments addresses the compliance and FX/liquidity layers of cross-border settlement, and brought new jurisdictions from this domain into BIS multilateral work:

• Project Mandala (programmable compliance for cross-border transactions):
  Singapore (MAS) and, from Phase 2 (per RBI Annual Report 2025-26),
  India (RBI). Embeds AML/CFT and policy checks into the transaction itself.
• Project Rialto (instant cross-border settlement using wholesale CBDC + FX):
  India (RBI) — its first confirmed BIS multilateral participation.
• Project Cedar / Cedar×Ubin+ (NY Fed wholesale FX settlement; under-30-second cross-border settlement with MAS):
  United States (NY Fed, research only) and
  Singapore (MAS).
• Project Jasper-Ubin (early cross-border wholesale corridor across heterogeneous DLT):
  Canada (BoC),
  Singapore (MAS), and the Bank of England (see
  UK).
• TIPS interlinking: the
  EU is pursuing fast-payment interlinking via TIPS (starting with India) rather than a multi-CBDC bridge.

These projects show the cross-border CBDC frontier shifting from "can we settle?" (mBridge/Dunbar/Jura answered yes) toward "can we settle compliantly, with FX and liquidity, at production quality?" (Mandala, Rialto, Agorá real-value).

Sources:

• RBI, Annual Report 2025-26 (Project Rialto and Project Mandala Phase 2 participation). https://www.rbi.org.in/
• BIS, "BIS Innovation Hub work on central bank digital currency (CBDC)." https://www.bis.org/about/bisih/topics/cbdc.htm

3. Comparative Table

Key observations from the comparison:

1. Technology divergence: No standard technology has emerged. Projects span purpose-built DLT (mBridge), enterprise DLT (Corda/Partior in Dunbar/Jura), public blockchain (Mariana), unified ledger (Agorá), and hub-based coordination (Icebreaker). The BIS itself has not endorsed a single approach.

2. Governance models vary radically: From centralised hub (Icebreaker) to shared-ledger joint governance (mBridge/Dunbar) to third-party platform (Jura) to unified ledger with heavy private-sector involvement (Agorá).

3. FX settlement is the common challenge: Every project must solve the FX conversion problem differently — competitive providers (Icebreaker), atomic PvP (mBridge/Jura), AMMs (Mariana), or integrated FX/settlement (Rialto).

4. Two projects have reached (or are reaching) real-value: mBridge is operational; Agorá moved to real-value testing per its 27 May 2026 first-phase report. Dunbar, Icebreaker, Jura, and Mariana remain completed experiments.

4. Geopolitical Dimensions

4.1 mBridge and De-Dollarization Concerns

The BIS withdrawal from mBridge in October 2024 was the defining geopolitical event in cross-border CBDC development. The withdrawal occurred days after Russian President Vladimir Putin proposed a "BRICS Bridge" at the BRICS Plus summit (October 2024), raising concerns that mBridge — with

China reportedly accounting for over 95% of settlement volume (self-reported) — could be repurposed as sanctions-evasion infrastructure.

BIS General Manager Agustín Carstens sought to distance the institution, stating publicly that "mBridge is not the BRICS Bridge." However, the platform's operational characteristics — enabling commodity trade settlement in national digital currencies without SWIFT — make it functionally capable of serving de-dollarization objectives regardless of governance intent.

The post-BIS mBridge now operates as a China-led platform with Gulf state participation. The addition of Saudi Arabia (SAMA) in 2024 is geopolitically significant given Saudi Arabia's role as the world's largest crude oil exporter and the historical centrality of the US dollar in oil trading (the "petrodollar" system).

Sources:

• BIS, "Project mBridge reached minimum viable product stage." https://www.bis.org/about/bisih/topics/cbdc/mcbdc_bridge.htm
• BIS, "Project mBridge: connecting economies through CBDC." https://www.bis.org/publ/othp59.htm

4.2 BRICS Digital Currency Proposals

Russia, as 2024 BRICS chair, proposed the BRICS Bridge — a settlement platform using CBDCs for trade between BRICS nations (see

Russia).

India, in April 2026, proposed placing the linkage of BRICS-member CBDCs on the 2026 BRICS Summit agenda — that BRICS+ members link their central bank digital currencies to make "cross-border trade and tourism payments easier."

Current status (June 2026): BRICS Bridge remains largely conceptual. Key challenges include:

• Partner reluctance: China, India, and South Africa have reportedly shown limited enthusiasm for Russia-led financial architecture (NOT INDEPENDENTLY VERIFIED to primary diplomatic sources)
• Sovereignty concerns: China is unlikely to adopt Russian technology given its own advanced e-CNY infrastructure and sovereignty priorities
• Timeline slippage: Russian Deputy Foreign Minister Sergey Ryabkov clarified in October 2025 that the system is now targeted for operational status by 2030, a significant delay from earlier 2026 targets
• Sanctions reality: Counterparties in third countries still face secondary sanctions risk when transacting with Russian entities, regardless of the settlement platform

Sources:

• RBI, Annual Report 2025-26 (April 2026 BRICS CBDC-linkage proposal). https://www.rbi.org.in/
• Carnegie Endowment, "Can the Digital Ruble Shield Russia From Western Sanctions?" October 2024. https://carnegieendowment.org/research/2024/10/russia-digital-ruble-development

4.3 US Absence from Cross-Border CBDC Projects

The

United States occupies a unique position in cross-border CBDC development. In January 2025, President Trump issued Executive Order 14178 (Jan 23, 2025) prohibiting federal agencies from "undertaking any action to establish, issue, or promote a CBDC" — the only binding US prohibition currently in force. A statutory CBDC moratorium has passed both chambers in differing forms but has NOT been enacted: the House passed the Anti-CBDC Surveillance State Act (H.R. 1919) 219-210 on 17 July 2025, and the CBDC moratorium rode the 21st Century ROAD to Housing Act through the Senate (89-10, 12 Mar 2026) and an amended House version (396-13, 20 May 2026), but the versions remain unreconciled. The stablecoin-first alternative is anchored by the GENIUS Act (S. 1582), signed as Public Law 119-27 on 18 July 2025. Kevin Warsh was sworn in as Fed Chair on 22 May 2026 and reaffirmed his opposition to a retail CBDC.

However, the US is not entirely absent from the wholesale layer:

• The Federal Reserve Bank of New York participates in Project Agorá as one of seven central banks
• The Fed/NYIC ran Project Cedar (cross-border wholesale FX settlement) and was an observer in mBridge before withdrawing from active participation over sanctioned-nation concerns
• The US focus is on wholesale/infrastructure research, not retail CBDC issuance — leaning instead on stablecoins, FedNow, and SWIFT/correspondent banking (USD ~58% of global FX reserves)

The domestic CBDC prohibition combined with continued wholesale participation creates a bifurcated posture: the US opposes retail CBDC on civil-liberties grounds while engaging in wholesale cross-border experimentation to maintain influence over the global settlement architecture.

Sources:

• Federal Register, "Strengthening American Leadership in Digital Financial Technology" (EO 14178), Jan 23, 2025. https://www.federalregister.gov/documents/2025/01/31/2025-02123/strengthening-american-leadership-in-digital-financial-technology
• HRF CBDC Tracker, "United States." https://cbdctracker.hrf.org/currency/united-states

4.4 Russia's Cross-Border CBDC Ambitions Under Sanctions

Russia's digital ruble entered narrow live government/budget use from 1 January 2026, with a legally-mandated large-scale launch on 1 September 2026 (see

Russia). It is explicitly positioned as part of Russia's strategy to build alternative cross-border payment infrastructure independent of Western-controlled systems. Western sanctions imposed after February 2022 disconnected major Russian banks from SWIFT and froze substantial overseas assets.

Cross-border plans:

• Proposed BRICS Bridge settlement platform (largely conceptual as of mid-2026)
• Bilateral digital ruble integration with "friendly state" regulators
• From 2025, the CBR planned to allow foreign banks to open digital ruble accounts
• Russia is not joining the BIS mBridge platform

Assessment: The digital ruble is "not a panacea for circumventing sanctions" (Carnegie Endowment). Cross-border ambitions face significant headwinds from partner reluctance and the reality that counterparties face secondary sanctions exposure. The programme's greatest risk is becoming an effective domestic surveillance tool that fails to achieve its stated international objectives.

Sources:

• Carnegie Endowment, "Can the Digital Ruble Shield Russia From Western Sanctions?" https://carnegieendowment.org/research/2024/10/russia-digital-ruble-development
• Bank of Russia, "Large-scale introduction of digital ruble to begin on 1 September 2026." https://cbr.ru/eng/press/event/?id=25774

4.5 EU Digital Euro Scope — Euro Area First, International Later

The

digital euro is designed primarily as a euro-area domestic instrument. As of June 2026:

• Intended to be available throughout the 20 euro-area Member States; the ECB's preparation phase concluded in October 2025 and the Governing Council moved to the next phase on 30 October 2025 (PSP pilot call opened 5 Mar 2026; 12-month pilot in H2 2027; potential first issuance during 2029, contingent on the EU adopting the Regulation in 2026)
• Non-euro EU Member States may access via bilateral NCB-ECB agreement under COM(2023) 368 — relevant to
  Sweden, whose e-Krona path is now tied to the digital euro timeline
• No finalised framework for international interoperability with non-EU CBDCs; TIPS interlinking (starting with India) is the near-term cross-border vector
• The ECB is not a participant in BIS mBridge, though Banque de France represented the Eurosystem in Projects Jura, Mariana, and Agorá

The digital euro's inward focus reflects the EU's priority of European digital sovereignty — ensuring that a European payment instrument reduces dependence on non-European payment networks. International interoperability is a "longer-term objective."

Source: ECB, "Eurosystem moving to next phase of digital euro project," October 30, 2025. https://www.ecb.europa.eu/press/pr/date/2025/html/ecb.pr251030~8c5b5beef0.en.html

5. What Works and What Does Not

5.1 What Works

Multi-party platforms with clear governance produce results. mBridge and Dunbar demonstrate that central banks can share infrastructure while preserving monetary sovereignty, provided governance frameworks are explicit. mBridge's bespoke rulebook and Dunbar's governance-enforced-by-technology approach both addressed the fundamental trust problem. The key insight: governance design matters more than technology choice.

Atomic DvP and PvP settlement eliminates settlement risk. Projects Jura, mBridge, Icebreaker, and now Agorá all demonstrated that CBDC-based cross-border settlement can achieve atomicity — either both legs of a transaction complete or neither does. This eliminates the settlement risk (Herstatt risk) that remains one of the most significant hazards in correspondent banking. The CPMI PFMI Principle 8 (settlement finality) is inherently satisfied by atomic settlement designs.

Hub-and-spoke preserves sovereignty at the cost of complexity. Icebreaker showed that retail CBDC systems can be interlinked without surrendering design autonomy. For central banks unwilling to participate in shared-ledger platforms (whether for sovereignty, geopolitical, or technical reasons), hub-and-spoke is the viable alternative. The trade-off: operational complexity of coordinating FX providers and maintaining the hub infrastructure.

Cross-border CBDC can advance independently of domestic retail CBDC.

Thailand's experience is instructive: world-class cross-border wholesale CBDC participation (mBridge, Inthanon-LionRock) while concluding that domestic retail CBDC is unnecessary given PromptPay.

Singapore follows a similar dual-track approach — the single most cross-border-active jurisdiction of the 20, yet with no retail CBDC commitment. This decoupling is important for jurisdictions where domestic payment systems are mature but cross-border settlement remains inefficient.

Embedding compliance and FX into the cross-border layer is the new frontier. Project Mandala (

Singapore,

India) and Project Rialto (India) show that the question has shifted from "can CBDCs settle across borders?" to "can they do so with embedded AML/CFT, FX, and liquidity at production quality?" — a maturation of the field beyond pure settlement proofs-of-concept.

5.2 What Does Not Work

Bilateral arrangements do not scale. The progression from Inthanon-LionRock (bilateral) to mBridge (multilateral) demonstrates that bilateral CBDC corridors are useful as stepping stones but unsustainable as an end state. Each bilateral arrangement requires separate legal, governance, and technical negotiations. For N jurisdictions, bilateral arrangements require N(N-1)/2 separate integrations — a combinatorial explosion that multilateral platforms avoid.

Geopolitical weaponisation undermines the public-good rationale. The BIS withdrawal from mBridge illustrates how cross-border CBDC infrastructure can become entangled in geopolitical competition. When a cross-border platform is perceived as serving one bloc's strategic interests (sanctions circumvention, de-dollarization), it loses credibility as a neutral public good. This dynamic threatens all future multilateral CBDC initiatives.

Most proof-of-concept projects do not transition to production. Of the major BIS Innovation Hub cross-border CBDC projects, only mBridge reached operational deployment, and only Agorá has since reached real-value testing (27 May 2026). Dunbar, Icebreaker, Jura, and Mariana all produced valuable insights but concluded as completed experiments. The gap between "technically feasible" and "operationally deployed" is the widest challenge in cross-border CBDC.

Domestic-only launched CBDCs gain no cross-border benefit. The

Bahamas (Sand Dollar),

Jamaica (JAM-DEX), and

Nigeria (eNaira) all launched live retail CBDCs with no operative cross-border corridor. For jurisdictions where remittances are economically significant (e.g., Jamaica, ~20% of GDP), the absence of cross-border functionality forecloses one of the strongest potential value propositions — and none has demonstrably improved on existing rails.

AMM-based FX for sovereign currencies raises unresolved questions. Project Mariana demonstrated that DeFi-style AMMs can execute wCBDC FX transactions, but the report itself acknowledged it did not address commercial viability or monetary policy implications. Central banks providing liquidity to AMM pools would take on balance sheet exposure and potentially create novel monetary transmission channels. This approach remains experimental.

5.3 The Governance Challenge: Who Runs the Common Platform?

The fundamental unsolved problem in cross-border CBDC is governance of shared infrastructure. Every model creates a governance question:

The CPMI PFMI Principle 2 (governance) requires that FMIs 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." No multi-CBDC platform has yet undergone formal PFMI governance assessment.

6. Recommended Framework

6.1 Interoperability Standards for CBDC

Central banks designing or participating in cross-border CBDC arrangements should adopt or align with the following standards:

Messaging standards:

• ISO 20022: The CPMI published harmonised ISO 20022 data requirements for cross-border payments in October 2023. CBDC systems must be compatible with ISO 20022 to interoperate with existing global payment infrastructure. As of 2025, approximately 20% of Fast Payment Systems and 6% of RTGS systems are aligned with harmonised requirements; an additional 9% of FPS and 32% of RTGS systems plan to align.

Settlement standards:

• PFMI compliance: Wholesale CBDC systems should be designed to meet CPMI-IOSCO Principles for Financial Market Infrastructures from inception. Key applicable principles: legal basis (P1), governance (P2), settlement finality (P8), money settlements (P9), operational risk (P17), and communication procedures (P22).

Token standards:

• No universal token standard has emerged. mBridge uses EVM-compatible tokens; Mariana used public blockchain token standards; Jura used Corda tokens; Agorá uses a unified-ledger token model. Standardisation of token format and bridge protocols is a prerequisite for cross-platform interoperability.

Identity and AML/CFT standards:

• CPMI ISO 20022 harmonisation supports FATF travel rule compliance by ensuring cross-border CBDC transactions carry required originator and beneficiary data
• Project Mandala demonstrates embedding AML/CFT and policy checks directly into the cross-border transaction
• The IMF CBDC Handbook chapter on financial integrity (2025) provides guidance on embedding AML/CFT compliance into cross-border CBDC design

Sources:

• CPMI, Harmonised ISO 20022 data requirements, October 2023. https://www.bis.org/cpmi/cross_border/publications.htm
• CPMI-IOSCO, PFMI. https://www.iosco.org/v2/about/?subSection=cpmi_iosco&subSection1=publications

6.2 Multi-Lateral Arrangement Design Principles

Drawing from the outcomes of mBridge, Dunbar, Icebreaker, Jura, and Agorá, and informed by BIS and CPMI analysis, the following design principles should guide multilateral CBDC arrangements:

Principle 1: Governance by consensus with defined escalation. No single participant should exercise veto power over platform operations. Governance decisions should follow defined voting or consensus mechanisms with escalation procedures for disputes. mBridge's post-BIS governance, where China reportedly accounts for over 95% of settlement volume, illustrates the risk of de facto unilateral control in nominally multilateral arrangements.

Principle 2: Sovereignty preservation by design. Each central bank must retain full authority over: (a) issuance and redemption of its CBDC; (b) AML/CFT compliance for its jurisdiction's participants; (c) the decision to exit the arrangement. Icebreaker's hub-and-spoke model exemplifies maximum sovereignty preservation; common-platform models (mBridge, Dunbar) require explicit sovereignty protections in the rulebook.

Principle 3: Legal framework before deployment. A bespoke legal framework — covering the legal nature of CBDC on the platform, governing law for cross-border transactions, insolvency treatment, and dispute resolution — must be established before production deployment. The IMF CBDC Handbook chapter on key legal considerations (2025) and BIS/CPMI work on legal frameworks provide the reference framework.

Principle 4: Technology neutrality with minimum interoperability. Platforms should not mandate a single DLT or technology stack. However, they must specify minimum interoperability requirements: common messaging formats (ISO 20022), common token standards for cross-platform transfers, and standardised API specifications.

Principle 5: Graduated participation and off-ramp provisions. New jurisdictions should be able to join through defined accession processes — as the Bank of Canada did when joining Agorá's real-value phase. Equally important, participants must have clearly defined exit provisions — including settlement of outstanding obligations, data portability, and transition periods.

Principle 6: Neutrality and resistance to geopolitical capture. The platform should be governed by rules that prevent any single geopolitical bloc from directing the platform's operations for strategic purposes. Independent oversight, rotating governance roles, and explicit prohibitions on using the platform for sanctions enforcement or evasion should be enshrined in the rulebook.

6.3 FX Settlement Model Options

Cross-border CBDC platforms must choose an FX settlement model. The experiments to date have tested several approaches:

Recommendation: No single FX model is optimal for all contexts. Direct PvP suits high-volume bilateral corridors (e.g., China-Thailand trade). Unified-ledger atomic settlement (Agorá) suits wholesale corridors that must integrate commercial bank money. Hub-and-spoke suits multilateral retail cross-border payments where sovereignty concerns dominate. AMMs remain experimental and should not be deployed for sovereign currency FX settlement without extensive further research on monetary policy implications.

6.4 G20 Cross-Border Payments Roadmap Alignment

Cross-border CBDC initiatives should be evaluated against the G20 roadmap quantitative targets for end-2027:

CBDC-relevant building blocks:

• Building Block 13: Interlinking payment systems for cross-border payments — includes CBDC system interlinking (Icebreaker model) and fast-payment interlinking (EU TIPS)
• Building Block 17: New multilateral platforms and arrangements — directly encompasses multi-CBDC platforms (mBridge, Dunbar, Agorá)

Progress assessment (FSB 2025 consolidated report):

• Most international roadmap actions have been completed
• However, tangible user benefits remain limited — an "implementation gap"
• Key performance indicators show only slight improvement since 2023
• CBDC cross-border applications remain among the least mature building block areas
• It is "unlikely that the end-2027 targets will be achieved on time"

Recommendation: Central banks should use the G20 roadmap KPIs (cost, speed, transparency, access) as the benchmark against which cross-border CBDC effectiveness is measured. A cross-border CBDC platform that does not demonstrably improve on these metrics relative to existing correspondent banking or Fast Payment System interlinking offers insufficient justification for the governance and technology investment required.

Sources:

• FSB, "G20 Roadmap for Cross-border Payments: Consolidated progress report for 2025," October 2025. https://www.fsb.org/2025/10/g20-roadmap-for-cross-border-payments-consolidated-progress-report-for-2025/
• BIS, "Enhancing cross-border payments: state of play and way forward," BIS Bulletin No. 119. https://www.bis.org/publ/bisbull119.htm

7. Decision Matrix

Central banks evaluating participation in cross-border CBDC arrangements should use the following decision matrix:

7.1 Should We Participate in a Cross-Border CBDC Platform?

7.2 Which Model Should We Choose?

7.3 Readiness Checklist

Before joining a cross-border CBDC arrangement, a central bank should confirm:

• Legal authority: Enacted legislation or clear statutory basis for issuing CBDC that can circulate on a foreign or shared platform
• Governance review: Independent assessment of the platform's governance structure, including voting rights, dispute resolution, and exit provisions
• AML/CFT compatibility: Confirmation that the platform's AML/CFT framework is compatible with domestic requirements and FATF standards
• Monetary policy impact assessment: Analysis of how participation affects domestic money supply, FX reserves, and monetary policy transmission
• Operational resilience: Cyber security assessment of the platform and contingency plans for platform outages, including the ability to settle outside the platform
• Sanctions compliance review: For platforms with geopolitically sensitive participants, confirmation that participation does not create sanctions exposure for the central bank or its regulated institutions
• PFMI alignment: Assessment of the platform against CPMI-IOSCO PFMI principles, particularly Principles 1 (legal basis), 2 (governance), 8 (settlement finality), 9 (money settlements), and 17 (operational risk)
• Exit plan: Defined procedures and timeline for orderly withdrawal from the platform, including settlement of outstanding obligations

Wikilinks Reference

• Jurisdictions:
  China,
  Singapore,
  Thailand,
  Japan,
  South Korea,
  India,
  Australia,
  EU,
  Russia,
  Sweden,
  United Kingdom,
  United States,
  Canada,
  Brazil,
  Bahamas,
  Jamaica, Eastern Caribbean,
  Ghana,
  Nigeria,
  South Africa
• International frameworks: BIS, CPMI, IMF, FATF
• Guidance document cross-references: technology, legal framework, privacy, financial inclusion

Verification Notes

Sources Verified (against the canonical fact matrix)

1. Project Agorá 27 May 2026 first-phase report: Confirmed atomic multi-currency wholesale settlement on a unified ledger is feasible; project advances to real-value testing with the Bank of Canada joining. Seven founding central banks: Banque de France (Eurosystem), Bank of Japan, Bank of Korea, Bank of Mexico, SNB, Bank of England, Federal Reserve Bank of New York.

2. mBridge transaction figures: 4,047 transactions worth ~USD 55.49 billion (as of November 2025) and China's >95% settlement-volume share are self-reported and not independently audited — flagged accordingly throughout.

3. BIS withdrawal timing and Carstens quote: October 2024 withdrawal and the "mBridge is not the BRICS Bridge" statement are consistently reported.

4. US cross-border / anti-CBDC posture: Only EO 14178 (Jan 23, 2025) is in force. The statutory CBDC moratorium passed both chambers in differing, unreconciled forms (H.R. 1919, 219-210, 17 Jul 2025; ROAD to Housing Act — Senate 89-10, 12 Mar 2026; amended House 396-13, 20 May 2026) and is NOT enacted. GENIUS Act = S. 1582 / Public Law 119-27 (signed 18 Jul 2025). Kevin Warsh sworn in as Fed Chair 22 May 2026.

5. EU digital euro phase: ECB preparation phase concluded October 2025; Governing Council moved to the next phase on 30 October 2025 (PSP pilot call 5 Mar 2026; pilot H2 2027; potential first issuance during 2029, contingent on EU adoption of the Regulation in 2026).

6. Project participation map (N=20): mBridge full participants among the 20 = China and Thailand (founding); US withdrew; Korea observer; India interest only. Agorá among the 20 = EU (BdF), Japan, Korea (+ Canada joining real-value). Dunbar = Australia, Singapore, South Africa. Icebreaker = Sweden. Mariana = EU (BdF), Singapore. Mandala = Singapore, India. Rialto = India.

Items Requiring Further Verification

1. China's >95% share of mBridge settlement volume: Reported but not directly confirmed by mBridge governance. UNVERIFIED to primary source. Plausible given e-CNY scale (cumulative CNY 16.7 trillion by November 2025, self-reported) relative to other participants.

2. mBridge enabling oil and commodity settlement outside SWIFT: The platform has the technical capability; whether oil majors are actively using it is UNVERIFIED — no primary documentation of specific oil settlement transactions has been published.

3. BRICS partner reluctance (China, India, South Africa): Reported by Carnegie Endowment and secondary sources; NOT INDEPENDENTLY VERIFIED to primary diplomatic sources.

4. Brazil mBridge participation: NOT VERIFIED. Brazil pursues cross-border interoperability through bilateral partnerships (HKMA-BCB, Oct 28 2024) and the Drex Phase 2 trade-finance/FX themes rather than mBridge.

5. Project Agorá full private-sector list: ~40 institutions confirmed by BIS; individual names (Visa, Swift, Citi, JPMorgan, UBS, Mastercard, MUFG) verified via reporting, full list per the BIS website.

---

Last updated: 2026-06-04