Connect with us

BUSINESS

Meridian FX Trial Links Three Systems Without Legal Finality

Project Meridian FX APAC linked three unlike payment systems, but the operator had to hold Singapore funds and legal finality is still missing.

Published

on

The Bank of England, the Monetary Authority of Singapore and the Bank of Thailand finished a cross-border FX settlement trial that linked three unlike payment systems. Their joint findings published on 13 November show the payments can move together in a lab. The same papers show that this still is not legally final settlement, and that only London has named a 2028 live date.

Partior, the project’s delivery partner, built the prototype Synchronisation Operator that matched both legs of each test trade. The Reserve Bank of Australia and Bangko Sentral ng Pilipinas sat in as observers on the policy work.

Three Unlike Systems, One Operator

Project Meridian FX APAC did not put sterling, the Singapore dollar and the baht onto one chain. Each central bank kept its own test money and its own ledger. The operator’s job was to make the two sides of a foreign-exchange trade move together, or not at all, which is payment-versus-payment settlement.

The three test beds were chosen because they do not work the same way. The Bank of England used its Synchronisation Lab, a simulated real-time gross settlement system that can earmark cash in a bank’s own account. The Bank of Thailand used Project SAN, a tokenised wholesale system with a central interoperability layer. MAS used a non-live build of SGD Testnet, a shared ledger for wholesale central bank digital currency with participant-controlled wallets.

THE THREE TEST RAILS

Central bank Test system How funds were reserved
Bank of England Synchronisation Lab (simulated RTGS) Earmarked in the participant’s own account
Bank of Thailand Project SAN (token ledger plus interoperability layer) Escrow inside SAN; the operator did not take custody
Monetary Authority of Singapore SGD Testnet (non-live wholesale CBDC) Tokens moved into an escrow wallet held by the operator

Adapters sat between the operator and each rail so the same four-stage workflow could run on all three. Instructions came in, legs were matched, funds were reserved, then settlement was requested only after every rail said it was ready. Until the first settlement instruction went out, the operator could still cancel the reservations and walk away. After that point of commitment, it stopped trying to unwind and kept retrying the remaining leg instead.

A second use case batched non-urgent trades and netted them across the three systems, so banks would not have to fund every gross payment. The prototype had to negotiate a batch that every rail would accept. If one name was short of cash, it dropped that name’s smallest trade and tried again. The central banks said this can cut liquidity needs. They did not publish a saving that a live desk could bank on.

The $1.2 Trillion That Existing PvP Cannot Cover

The Bank of England’s own release still cites the older BIS snapshot: more than $1.4 trillion a day, or 10% of average FX settlement in April 2025, sat fully exposed to settlement risk. The newer book is the Global Foreign Exchange Committee survey for April 2026, issued on 16 September 2026 from 49 firms.

APRIL 2026 DAILY FX SETTLEMENT

  • Gross volume: $14.4 trillion a day, up 8% from October 2025.
  • Settled PvP: $4.8 trillion, or 33%, of which 99.5% was in a CLS-eligible currency pair.
  • Intragroup: $6.1 trillion, or 42%, mostly inter-branch.
  • Gross bilateral, no mitigation: $1.2 trillion with no risk mitigation, or 8% of the pile.

Pre-settlement netting covered $2.0 trillion (14%), leaving $0.4 trillion to pay after the net. Accounts with timing controls covered another $0.4 trillion (3%). Of the unprotected bilateral slice, 84% could not use PvP at all. The committee said the blocks were an ineligible counterparty (51%), an ineligible currency pair (33%) and an ineligible trade type (33%), and that those reasons overlap. A thinner $191 billion was eligible for PvP and still went out raw.

That remainder is the market this design is aiming at. CLS already handles almost all of the PvP that runs in its currency pairs. The trades that cannot get in are the ones a coordinator between RTGS and token ledgers is supposed to reach, including names and pairs that never qualified for the old window.

Singapore’s Testnet Put Funds in Escrow

The policy baseline the three central banks want is a non-custodial operator. It matches instructions, watches readiness and tells each system when to settle. It does not run the ledgers, keep settlement accounts, or hold anyone’s money. Authorised third-party access and earmarking are what make that light model possible.

Singapore’s testnet did not let the operator earmark balances on a participant’s behalf. Tokens moved into a technical escrow wallet the operator held, under a bank’s authorisation and with participant digital signatures. The policy report treats that escrow, and the operator’s job of calculating net positions for the multilateral batch, as extras of this test rather than features of synchronisation itself. Both extras change the legal picture if they show up in a live service. Holding funds raises custody, credit, safeguarding and insolvency questions. Calculating the net moves the operator closer to clearing.

Thailand’s SAN also used escrow, but it stayed inside SAN, so custody never passed to the operator. The Bank of England lab earmarked in place. The light model survived where the rail was built for it. It did not survive, in this build, on SGD Testnet.

Partior already sells atomic settlement on a private ledger to banks. In this project it was the technical partner that had to grow the operator’s job to fit the weakest rail. The policy report says treatment should follow the functions actually performed, not the label on the box.

A Shared Legal Finish Line Is Still Missing

The technical report is careful on this point. The workflow’s point of commitment is not legal settlement finality. Each domestic system can define its own final event. A live FX service would still have to say how those events combine into one outcome the courts and the rulebooks will recognise, including when a rail is down.

Technical atomicity does not by itself establish a common, legally recognised moment of finality across participating systems.

Project Meridian FX APAC Policy Report, Bank of England

That sentence is the live-service problem in one line. Coordinating legs so they settle together does not, on its own, create a common legally recognised moment of finality across separate national systems. Rulebooks still have to say who may pause, cancel or retry a trade, and who eats a loss when the operator, a bank or a rail fails. Data has to cross borders in a form each privacy regime will accept.

The reports also refuse to treat the prototype as a product decision. They do not recommend a common build. They do not presume a live rollout. Further legal work is listed as future work, alongside more currencies and more countries.

Tom Mutton, the Bank of England’s executive director of payments, innovation and fintech, said the project gave “practical evidence of how synchronised settlement could help reduce risk and improve efficiency in cross-border FX transactions.” Kenneth Gay, MAS’s chief fintech officer, said synchronisation could help enable PvP settlement across systems built on different technologies. Pucktada Treeratpituk, director of payment systems and financial technology policy at the Bank of Thailand, called the trial a practical foundation for modernising wholesale settlement, not a go-live.

What the Tests Left Untried

The exception runs that did take place were short and controlled. The operator faced insufficient funds, bad account identifiers and temporary infrastructure outages. In those cases it either finished both legs or unwound the whole trade, so one currency never paid away alone. After commitment, an outage produced retries rather than a rollback, because reversing one settled leg would break PvP.

The papers are blunt about the conditions that never got a run. Those are the conditions a London-Bangkok-Singapore service would hit on a normal week.

OUTSIDE THE TEST SCOPE

  • Operating hours: The prototype assumed the three systems were open together, and did not test non-overlapping hours.
  • Time zones: Cross-zone recovery, including an overnight earmark held in one system while another is closed, was left for later work.
  • Earmark expiry: Reserved funds that time out were not modelled.
  • Long outages: Only temporary failures were run; prolonged downtime was not.

The experiment also used simulated instructions and configurable balances. It did not put live commercial trades through the operator. Idempotency, queryable status and reliable event delivery were the plumbing the reports say a live service would need, because a coordinator that cannot see a failed leg cannot keep PvP intact.

Only London Has Named a 2028 Date

The Bank of England has already said it intends to launch a live synchronisation interface in 2028, a pledge first set out with the Financial Conduct Authority in a joint call for input on tokenisation. The APAC findings, the Bank said, will feed that UK build. The executive summary draws a hard line around the other two names: participation “does not imply an equivalent implementation commitment or a particular policy direction” by MAS or the Bank of Thailand.

HOW THE MERIDIAN SERIES GOT HERE

  1. 2023: Project Meridian, with the BIS Innovation Hub’s London centre, tests synchronised settlement between RTGS and a digital land registry for housing.
  2. 2025: Project Meridian FX, with the Bank of England, Banque de France, Deutsche Bundesbank, Banca d’Italia and the European Central Bank, tests sterling-euro PvP against Eurosystem ledgers, including DLT.
  3. November 2025: The Bank of England, MAS and the Bank of Thailand announce the APAC follow-on.
  4. 2026: The Bank of England runs a Synchronisation Lab for several prospective operators and use cases, including Partior.
  5. 13 November 2026: The three APAC reports are published.
  6. 2028: The Bank of England’s target for a live synchronisation interface on its renewed RTGS service.

A UK interface on RT2 can be built with earmarking from the start. Connecting it to Asian token rails is a different product, because those rails do not share one legal moment of finality and, in this test, did not even share one way of locking cash. The next work the papers themselves list is more jurisdictions, more currencies, real operating hours, and the legal contracts that would make a completed technical workflow into a completed trade.

The Bank of England still plans to open that live interface in 2028. Singapore and Thailand have not named a matching date.

Harry is the editor of IAQABA, an independent publication he owns and runs. A decade in journalism, beginning as a reporter and now as the editor of his own titles, has left him with a clear test for what deserves a story: it has to change what a reader knows or decides, and it has to rest on something he can point to. That rules out recycled press releases, forecasts with no data behind them and rumours that no document supports. It leaves room for a great deal, and the site covers news, business, science and technology alongside sports, entertainment and lifestyle, with travel, auto and gaming given the same standard rather than lighter treatment. Sources are primary wherever possible: the regulator's filing, the company's own statement, the transcript, the dataset, or the product on Harry's desk. Figures are checked before they are published and rechecked if a reader questions them. Mistakes are corrected under a published policy. Readers across the world can reach him directly at support@iaqaba.com.

Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Trending