Daily news

Project Meridian FX APAC: BoE, MAS and Bank of Thailand Test Synchronised FX Settlement

The Bank of England, the Monetary Authority of Singapore and the Bank of Thailand on Nov. 13, 2026 published results of Project Meridian FX APAC, a non-live experiment showing a Partior-built Synchronisation Operator could settle FX payment-versus-payment and multilateral net trades across an RTGS and two DLT/tokenised test systems. The BoE plans a live synchronisation service in 2028; MAS and BoT made no implementation commitment.

MyTraderBot DeskPublished: 9 min readخواندن به فارسی
Share
Bank of England building on Threadneedle Street in London (archive photo)
Photo: George Rex / Wikimedia Commons · CC BY-SA 2.0 (archive)
Table of Contents

The Bank of England (BoE), the Monetary Authority of Singapore (MAS) and the Bank of Thailand (BoT) on Friday, November 13, 2026 published the results of Project Meridian FX APAC, a joint experiment that tested whether a single coordination service could settle both currency legs of a foreign exchange (FX) trade at the same time across very different central bank systems, from a traditional Real-Time Gross Settlement (RTGS) system to distributed ledger (DLT) and tokenised platforms. According to the Bank of England's release and the matching MAS statement, the trial showed that this approach, called synchronisation, worked across all three test environments.

It is important to be clear about what this is: a proof-of-concept run in simulated, non-live environments, not a working cross-border payment system. No real customer money moved. The Bank of England says the findings will inform its own plan to launch a live synchronisation service in 2028, but its executive summary states that participation "does not imply an equivalent implementation commitment or a particular policy direction" by MAS or the Bank of Thailand.

What Project Meridian FX APAC tested

The project connected a prototype Synchronisation Operator (SO) to three simulated wholesale settlement environments with different architectures. Partior, the project's delivery and technical partner, developed, integrated and tested the SO prototype. The Reserve Bank of Australia (RBA) and Bangko Sentral ng Pilipinas (BSP) took part as observer central banks and contributed to the policy workstream.

The technical report describes the three systems:

  • Bank of England – Synchronisation Lab: a test environment that lets operators interact with a simulated, account-based RTGS system, including a prototype capability to reserve (earmark) cash and coordinate settlement.

  • Bank of Thailand – Project SAN: a prototype tokenised payment system with a centrally managed token ledger and an interoperability layer that routes transactions to a settlement layer.

  • MAS – SGD Testnet (non-production environment): a shared-ledger infrastructure supporting the issuance, transfer, programmability and redemption of Singapore dollar wholesale central bank digital currency (CBDC), where banks control their own wallets. The report notes SGD Testnet is an operational environment, but its use in this project was limited to a non-live configuration.

Three things were tested: bilateral payment-versus-payment (PvP) settlement of the two central bank money legs of a single FX trade; multilateral net FX PvP across all three systems, designed to cut liquidity needs by settling several trades together; and selected failure scenarios such as insufficient funds, invalid account identifiers and temporary system outages.

How the synchronisation workflow works

Every trade followed the same four stages. Each bank sent a settlement instruction to the SO; the SO matched the two legs of the trade; it then asked each system to earmark the needed funds; and once every system confirmed the funds were locked, it requested settlement. Until the first settlement instruction executes, the whole trade can still be cancelled simply by releasing the earmarks. After that "point of commitment," the SO no longer unwinds the trade and instead keeps retrying any leg whose system is temporarily unavailable. The goal is an all-or-nothing outcome: either both currencies move, or neither does.

Key findings

1. Different technologies can still interoperate

The BoE's executive summary says synchronisation worked across systems with "materially different technologies and operating models," suggesting that cross-border interoperability "does not necessarily require central banks to converge on a single infrastructure design." In plain terms, a country running a classic RTGS and another experimenting with tokenised central bank money could, in principle, still settle FX against each other safely.

2. Missing capabilities create workarounds and extra risk

The process was simplest where a system let an authorised third party reserve funds and trigger settlement. Where that was not available, the team used system-specific adapters. In the MAS test environment, the SO ran a dedicated escrow wallet to imitate fund reservation, and in Project SAN funds were placed in escrow within the SAN arrangement. The banks stress that this escrow set-up reflected the experimental systems and "is not an inherent feature of synchronisation or a proposed model for a future SO," because an operator that holds funds takes on custody, credit, liquidity and insolvency questions that a purely coordinating operator avoids.

3. Multilateral netting worked, but coordination is hard

Multilateral net FX PvP was technically achieved across all three environments, which could reduce the amount of liquidity banks must pre-position. The difficult part was not computing net positions but getting all three systems to agree on the same set of trades while funds stayed reserved. When one system rejected an earmark, the prototype removed a transaction belonging to the affected participant (generally its smallest one) and tried again until every system accepted the batch. In the trial, the SO itself calculated the net positions, a function the reports say brings it closer to a clearing house; one option for the future is for central bank systems to perform netting themselves.

4. Failure handling was encouraging, but not fully tested

In every exception scenario tested, the SO either completed settlement or fully unwound the trade without leaving one leg to settle on its own. Features such as idempotency (processing a repeated instruction only once), queryable transaction status and reliable event delivery were key. However, the reports say important real-world conditions were not tested, including non-overlapping operating hours, time-zone differences, earmark expiry and prolonged outages.

Policy questions: finality, governance and liability

The policy report examines the operator through five lenses: access and assurance; regulatory perimeter and oversight; settlement finality; governance and accountability; and data and information sharing. Its central warning is that technical atomicity is not the same as legal settlement finality: coordinating two ledgers so they move together "does not by itself establish a common, legally recognised settlement outcome" across separate jurisdictions. A live service would also need clear rules on who can pause, cancel or retry transactions, how losses and liabilities are allocated, and how data can be shared across borders.

The report uses existing arrangements as reference points rather than templates, including the RBA's Batch Administrator framework for access, the cooperative central bank oversight of Swift, and CLS for cross-border FX settlement finality. It concludes that existing frameworks offer "useful building blocks, but not a ready-made model," and that internationally developed principles could help, "if considered useful."

Why it matters for crypto, tokenisation and stablecoins

FX settlement risk, often called Herstatt risk, arises when one side of a currency trade pays but does not receive the other currency. Citing the 2025 BIS Triennial Survey, the BoE release says more than US$1.4 trillion, or 10%, of average daily FX settlement in April 2025 remained fully exposed to settlement risk. PvP removes that risk by making each leg conditional on the other.

For the digital asset industry, Meridian FX APAC is a sign of how central banks plan to connect tokenised money with existing rails. Instead of forcing everyone onto one shared ledger, the synchronisation model lets each country keep its own infrastructure, whether RTGS, a tokenised platform or a wholesale CBDC, and coordinates settlement across them. That matters for tokenised deposits, tokenised securities and, eventually, regulated stablecoins that may need to settle against central bank money. It fits a wider trend of traditional market infrastructure moving on-chain, from the DTCC tokenisation service connecting to Solana to NYSE choosing Avalanche to settle tokenised stocks. Thailand, one of the participants, is meanwhile tightening its rules for private digital money, as seen in the Thai SEC's proposed stablecoin transfer cap.

There is no direct short-term price effect for crypto assets: this is wholesale, central bank money infrastructure, not a token launch or a public blockchain integration. Notably, the SO prototype itself was a centralised system, not a DLT-based solution.

How it builds on earlier work

The project extends earlier Meridian experiments. The original Project Meridian showed an SO could coordinate a payment in an RTGS system with the transfer of an asset on another ledger. Project Meridian FX, led by the BIS Innovation Hub with the Bank of England, Bank of France, Deutsche Bundesbank, Bank of Italy and European Central Bank, applied the idea to FX PvP. Meridian FX APAC was announced in November 2025 and widened the scope to more currencies, technologies, infrastructures and time zones.

What central bankers said

Tom Mutton, Executive Director of Payments, Innovation and Fintech at the Bank of England, said the project "provides practical evidence of how synchronised settlement could help reduce risk and improve efficiency in cross-border FX transactions." Kenneth Gay, Chief FinTech Officer at MAS, said it "has shown how synchronisation could help enable PvP settlement across these diverse environments." Pucktada Treeratpituk, Director of Payment Systems and Financial Technology Policy at the Bank of Thailand, said it "successfully demonstrated that synchronisation can work across diverse wholesale settlement systems."

Limitations and what comes next

  • The trial did not reproduce the full FX trade lifecycle, live participants, realistic volumes or production conditions, and assumed overlapping system availability.

  • The reports are exploratory and do not recommend a common implementation or policy approach.

  • Possible next steps include delivery-versus-payment use cases, more jurisdictions and industry participants, a wider range of currencies and closer modelling of market participants.

  • The Bank of England will use the findings toward its live synchronisation service planned for 2028, announced as part of its Call for Input on the future of tokenisation with the FCA.

FAQ

Is Project Meridian FX APAC a live payment system?

No. It was an experiment in simulated, non-live test environments. Only the Bank of England has a stated plan for a live synchronisation service, targeted for 2028.

Does it use a blockchain or a cryptocurrency?

Two of the three test systems were DLT or tokenised platforms (Project SAN and SGD Testnet), but the operator prototype was a centralised system, and no public cryptocurrency or stablecoin was involved.

What does Partior do?

Partior was the technical partner that developed, integrated and tested the Synchronisation Operator prototype used in the trial.

Sources & further reading

Share

See the signals and their results

Browse publicly logged signals and how they performed, or create a free account to follow them.

Project Meridian FX APAC: Synchronised FX Settlement Trial