The deep dive
Project Agorá publishes: cross-border settlement gets a two-layer blueprint
The BIS Innovation Hub published the Project Agorá prototype report on 27 May, two years into a programme that now spans seven central banks and more than forty financial institutions, including 
JPMorgan, HSBC, Deutsche Bank, Swift, Mastercard and 
UBS. The Bank of Canada is joining for the next phase, with more institutions expected. The report matters less for any single number than for the architecture it settles on.
The load-bearing design decision was that central banks would retain full control of their own reserves. That single requirement drove the entire two-layer structure. Tokenised commercial-bank deposits sit on a shared unifying ledger where every participant can coordinate a payment. Tokenised central-bank reserves sit on separate jurisdictional ledgers, one per currency area, each operated under the relevant central bank's own authority. A smart contract called the payment coordinator, deployed on the unifying ledger, sequences the workflow but never directly invokes contracts on the jurisdictional ledgers, and there is no cross-ledger bridge: coordination runs through deterministic events and participant-operated middleware. This is a different model from mBridge, where a single shared platform hosts both central-bank and commercial money.
For an operator, three things stand out. First, the participant set is the major reserve currencies (Banque de France for the Eurosystem, plus the central banks of Japan, Korea, Mexico, Switzerland, the United Kingdom, and the Federal Reserve Bank of New York), and it is wholesale-only, which distinguishes it sharply from mBridge's graduated consortium where retail rails are in scope. Second, the prototype is a prototype: it addressed speed, efficiency, transparency and risk across nine friction areas participants had identified, and found them materially or partially improved, but no real value has settled and no production timeline is stated. The next phase is real-value testing. Third, participation from JPMorgan, HSBC and UBS reads as operational rather than symbolic; all three are balance-sheet committed to the testing phase.
The honest read is that Agorá has now produced the clearest public blueprint for how tokenised cross-border settlement can preserve monetary sovereignty while still settling atomically. It does not replace correspondent banking so much as re-plumb it: the report is explicit that the model preserves the correspondent-banking backbone and applies new technology to its performance. Whether the blueprint becomes infrastructure depends on the real-value phase, and on the governance questions the prototype deliberately left out of scope: the operating entity for the unifying ledger, the rulebook, and liability. For now it is the reference design the rest of the field will be measured against.
Worth watching next
- Whether the Bank of Canada's entry to Project Agorá widens the real-value phase beyond the original seven currency legs.
- The SEC's revised stock-tokenisation exemption text, and whether it admits third-party tokens carrying full ownership rights.
- Kelvin Wong's three post-consultation speeches (bodies not yet in the record) for the SFC's public framing on tokenised secondary markets.
Forward
Forward this to one person who should be reading it.