BIS Project Agorá: How 7 Central Banks and 40+ Institutions Proved Tokenized Settlement Works

Two years. Seven central banks. More than forty regulated institutions. The result: cross-border wholesale payments that settle atomically, without credit risk, around the clock. The Bank for International Settlements closed its Project Agorá prototype in May 2026 and is advancing to real-value testing.

This article is for informational purposes only and does not constitute financial, legal, or investment advice.

What Project Agorá Was

Project Agorá was a multi-year prototype run by the Bank for International Settlements Innovation Hub in collaboration with seven central banks: the Bank of France (representing the Eurosystem), the Bank of Japan, the Bank of Korea, the Bank of Mexico, the Swiss National Bank, the Bank of England, and the Federal Reserve Bank of New York. More than forty private financial institutions participated, selected through an open call managed by the Institute of International Finance (IIF).

The project's core thesis was that the bottleneck in cross-border wholesale payments is not speed — SWIFT messages can move quickly — but settlement risk. When payments require multiple correspondent banking hops across jurisdictions, each hop introduces a window of credit risk during which neither party has confirmed final settlement. Agorá aimed to eliminate that window entirely through atomic settlement on a shared tokenized ledger.

The Architecture

The prototype placed tokenized commercial bank deposits and tokenized central bank reserves on the same unified ledger — a concept the BIS calls a "unified ledger" architecture. When a cross-border payment is initiated, the debit and credit leg settle simultaneously as a single atomic transaction. There is no intermediate state where one leg has settled and the other has not.

This matters because it directly eliminates the principal risk that exists in today's correspondent banking chains: the possibility that a counterparty fails or a leg does not settle, leaving a bank exposed. The BIS characterized this as removing "credit risk at the edges" of the payment system.

The system operated on a round-the-clock basis rather than during fixed settlement windows. This contrasts with legacy infrastructure like Fedwire (US) or TARGET2 (EU), which have defined operating hours and batch settlement cycles that create overnight exposure.

Scale of the Problem Being Solved

The BIS report cited FXC Intelligence cross-border payment volume data showing $195 trillion moved through international payment rails in 2024, with projections of $320 trillion by 2032. The bulk of that volume runs through correspondent banking chains established decades ago, with average settlement times of one to five business days depending on corridor and currency pair.

The costs associated with that friction — float, pre-funding requirements, reconciliation overhead, nostro account maintenance — are estimated in the billions of dollars annually for major financial institutions. Agorá's architecture would eliminate most of that overhead by making pre-funding unnecessary when atomic settlement is guaranteed.

Moving to Real-Value Testing

As of May 2026, the BIS announced the prototype phase is complete and the project is advancing to real-value testing. This means actual transactions with economic value — not simulations — will be run through the architecture to validate it under live conditions. The exact timeline and jurisdictions for real-value testing have not been publicly specified, but participation from the Bank of Japan and the Federal Reserve Bank of New York signals that major currency corridors are included.

The transition from prototype to real-value testing is the key threshold in central bank digital infrastructure projects. Prototypes demonstrate technical feasibility; real-value tests expose operational, legal, and risk management gaps that simulations cannot replicate.

Implications for DLT Infrastructure Builders

Project Agorá's validation has direct implications for ledgers already positioned to handle tokenized wholesale settlement. The XRPL was built with atomic settlement as a core design property — transactions either settle fully or do not settle at all, with no partial execution. The ledger also handles multi-currency flows natively through its DEX and issued currency mechanism.

For context on how compliant tokenized asset issuance works on XRPL at the protocol level, including the trust line mechanism that governs investor participation, see our analysis of XRPL trust line compliance mechanics.

Agorá does not endorse any specific ledger. But its architecture — unified ledger, tokenized reserves, atomic cross-currency settlement, 24/7 operation — describes exactly the capabilities that public ledgers like XRPL have built since launch. The question for the next phase is whether central bank real-value testing will look to established public infrastructure or build permissioned alternatives from scratch.

Central Bank Participation Signals

The participation of the Bank of France representing the Eurosystem, the Bank of Japan, and the Federal Reserve Bank of New York in a project that concluded "the architecture works" is not a minor institutional data point. These three entities collectively oversee the settlement of the world's three largest reserve currency flows. Their technical endorsement of tokenized atomic settlement — even in prototype form — represents a meaningful shift in how global settlement infrastructure is being evaluated at the highest institutional levels.

The EU's parallel push to expand DLT pilot regime infrastructure, covered in our analysis of the 39-firm EU DLT reform coalition, reflects the same institutional momentum from the private sector side.