Japan’s Financial Services Agency has 43 companies inside a live test of how banks settle tokenised deposits between one another, and the design that deserves scrutiny is the one nobody defends: in the second of two models under evaluation, a private stablecoin does the job central bank money does today. The Bank of Japan speech the sponsors invoke as favourable context is, read in full, a warning about that substitution.
DCP Co., Ltd., GMO Aozora Net Bank and ABeam Consulting disclosed the roster on August 26, 2026. The FSA adopted the project on April 3, 2026 as the third case under the Payment Innovation Project, the payments-specific track opened inside its FinTech Proof-of-Concept Hub in November 2025, and the 14th case the hub has supported since 2017. Full-scale testing began on August 20, 2026, with the period given only as April 2026 “for the time being” — no end date.
Two designs, and neither settles in central bank money
The first model, branded the TD Lead Bank Model, appoints one private bank as lead bank and executes the user-to-user transfer and the interbank settlement inside that bank at the same moment. The second, the TD-SC Integration Model, conducts the interbank leg using a stablecoin. A third route, integrating with existing payment systems, will also be explored — but neither document names the Zengin System or BOJ-Net. The target is Real-Time Gross Settlement (RTGS) on every transaction, 24/7/365, tested for legality, usefulness and feasibility.
Read the two documents together and the gap sharpens. The FSA’s April description uses no branding: it says the experiment will test “a method using deposit accounts opened between banks” and “a method using stablecoins”. Both routes settle the interbank leg in a commercial liability — one bank’s deposit balance, or one issuer’s token. That is the design question for any custodian or treasurer that would later plug into the rail.
The 43 include no megabank
The count needs unpacking. The participant table runs to 37 lines — three applicants, 13 participating-bank entries and 21 observer entries. It reaches 43 because four lines are holding companies named alongside subsidiary banks: Tokyo Kiraboshi with Kiraboshi Bank and UI Bank, Fukuoka Financial Group with the Bank of Fukuoka and Minna Bank, plus Resona Holdings and Chugin Financial Group with their own banks. As legal entities: three applicants, 18 participating banks, 22 observers.
The participating tier runs to the Bank of Yokohama, Shizuoka, Joyo, Ashikaga, Higo and Hokuriku, plus AEON Bank and the Shoko Chukin Bank, the government-affiliated lender to small-business cooperatives. Observers include Chiba, Hiroshima, the Bank of Kyoto and au Jibun Bank. Mitsubishi UFJ, Sumitomo Mitsui and Mizuho appear nowhere in either tier. Japan’s tokenised deposit settlement layer is being specified by the users of the interbank system rather than its largest members, whose own on-chain work has run on separate infrastructure.
Note what the tiers commit to. Participating banks “present views on the verification hypotheses arising from the pilot study”; observers “voluntarily ask questions, share opinions, or raise any concerns”. Neither definition obliges a single institution to move money.
The release cites Bank of Japan Governor Kazuo Ueda’s March 3, 2026 address to FIN/SUM 2026 as supportive, on the strength of his plan to explore tokenising central bank money. The same speech says this: “Central bank money provides a foundation where money can be exchanged at par value for all payment instruments; in other words, the singleness of money is secured. Unless the deposits at different banks are connected through central bank deposits, there is a risk that people will perceive this as variation in the value of deposits among banks, as was the case during the wildcat banking era in 19th-century United States.”
That is the question the bank’s own announcement leaves alone. The lead-bank model concentrates the settlement asset on one commercial balance sheet; the stablecoin model moves it to an issuer’s. Both diverge from the netting-then-settle pattern bank consortia have favoured elsewhere and from the capacity arithmetic driving the US tokenised deposit push. Only the stablecoin-against-deposit-token route has a close precedent.
What follows is a legal review as much as a technical one. Japan is simultaneously rewriting how it classifies digital assets, and the FSA will publish the compliance and legal-interpretation issues the experiment surfaces once it concludes. With no end date and 40 bank entities in opinion-only roles, the signal to watch is which design the legal work eliminates first — and whether the route back into existing payment systems ever acquires a name.
This article is informational analysis only and is not financial, investment, or trading advice. Cryptocurrencies are highly volatile and can lose substantial value rapidly. Past performance and historical patterns do not guarantee future results. Do your own research and consult a regulated financial adviser before making any investment decision.
Karthik Subramanian
Karthik Subramanian is a founder, writer, and technology consultant with nine years in the crypto ecosystem. He covers token economics, L1/L2 infrastructure, DeFi protocols, wallets/custody, and the bridge between crypto and forex—broker technology, liquidity, and macro drivers. Karthik’s writing focuses on clear, practical frameworks that help professionals evaluate new products and on-chain innovation alongside FX market realities.
