Suit Up · Asia's institutional desk on tokenisation

TL;DR

Scope first: this playbook is for banks. Asset managers and fintechs cannot issue a tokenised deposit because they do not hold a banking licence; the right reading for non-banks is the Stablecoin licensing decision tree instead. A tokenised deposit is bank money on chain. The deposit liability stays inside the bank's balance sheet under the existing banking licence, and the chain entry represents transfer of beneficial interest in that liability. The decision tree turns on intra-bank vs cross-bank scope, in-house vs consortium platform, jurisdictional supervisor relationship, and use-case scope (wholesale settlement, treasury management, retail accommodation, cross-border). Most banks land at intra-bank-first with consortium membership for cross-bank capability; a small number of GSIBs justify in-house build across the full stack.

Decision frame

The question on your desk is whether and how to launch a tokenised-deposit programme. Four framings show up in early-stage internal memos that do not survive contact with the supervisor.

"Tokenised deposit is just a faster RTGS." It is not. Real-time gross settlement (RTGS) is central-bank-money settlement between bank reserve accounts; a tokenised deposit is commercial-bank money moving on a ledger the bank or a consortium operates, with the deposit liability staying inside the issuing bank rather than transiting central-bank infrastructure. The two coexist; RTGS clears the inter-bank cash leg, the tokenised deposit moves the customer-facing claim.

"We should issue a stablecoin instead." Sometimes. For a bank, the tokenised-deposit answer is structurally cleaner for most wholesale and treasury use cases because the instrument stays inside the existing banking licence, the prudential and resolution framework is the bank's existing one, and the customer relationship continues to run as a deposit relationship. The stablecoin path adds a separate licensing perimeter, often a separate issuer entity, and a credit-risk profile that runs to the issuer rather than to the bank. The stablecoin licensing decision covers the comparison directly. The choice is not exclusive: several large banks (HSBC in Hong Kong from H2 2026 under the Stablecoins Ordinance) are running both.

"Cross-bank interoperability comes free with the platform." It does not. Cross-bank transferability is an inter-bank arrangement layered on top of the chain, not a property of the chain itself. Without an explicit framework (DBS-Kinexys, Partior, Fnality, or a wholesale-CBDC settlement layer), a Bank A token sent to a Bank B customer leaves the recipient holding a claim on Bank A while sitting at Bank B.

"We can launch retail-facing tokenised deposits as a marketing differentiator." Typically not. Supervisor expectations on consumer protection compound when tokenisation is in the loop, and the wholesale and treasury use cases are where the early production cohort sits for structural reasons.

The seven design decisions

The meat of the playbook. Each decision is sequential, but the answers compound.

Scope: intra-bank vs cross-bank

Intra-bank tokenised deposit (the bank's own customers transact with each other on the bank's platform) is the operationally simpler launch. The chain is a transfer mechanism for the bank's own deposit liabilities; the reconciliation surface is the bank's own general ledger; the supervisor relationship runs through the home regulator on existing terms. Wells Fargo Digital Cash is the worked example of an intra-bank-only programme. The entire JPMorgan flow on Kinexys Digital Payments was intra-bank for the first several years post-launch.

Cross-bank tokenised deposit (the bank's customers settle against another bank's customers on a shared rail) requires either a consortium (Partior, Fnality, the Kinexys interoperability framework with DBS) or a bilateral integration. The November 2025 DBS-Kinexys interoperability framework is the named bilateral case at GSIB scale; it pairs DBS Token Services with Kinexys Digital Payments and JPMD on Base, with conversion logic the framework owns rather than the customer. The cross-bank path is harder because it touches inter-bank settlement (the Bank A liability has to extinguish and the Bank B liability has to issue, with the inter-bank cash leg settled separately) and because each participating bank needs its home supervisor comfortable with the arrangement. Most banks launch intra-bank first and add cross-bank capability through consortium membership in a sequenced phase 2.

Use-case scope

Four use cases dominate the early cohort. Wholesale FX settlement is the most common first launch, particularly in APAC where corporate treasury teams routinely move multi-currency liquidity across jurisdictions on RTGS-constrained windows; the tokenised-deposit rail collapses cut-off times into a 24/7 operating envelope. The Standard Chartered + Ant International Whale launch is the four-currency (HKD, CNH, USD, SGD) production worked example. Corporate treasury management (intra-day liquidity, programmable conditions, automated sweep logic between intragroup entities) is the second. Cross-border B2B payments is the third, with the FX leg either inside the platform (Whale, Partior PvP) or via a bilateral framework. Retail accommodation is the fourth and structurally the rarest.

The use case determines the integration scope. A wholesale-FX launch needs cross-currency handling and cut-off-time discipline; a treasury launch needs deep API integration with corporate ERPs and treasury workstations; a cross-border launch needs the inter-bank settlement architecture in place. Shipping more than one use case at launch is a common over-reach.

Platform: build vs consortium

GSIBs typically build. JPMorgan Kinexys, HSBC tokenised-deposit infrastructure on Orion, DBS Token Services, and Wells Fargo Digital Cash are all in-house builds inside the bank's existing licensing perimeter. The structural rationale is that the rail is the bank's franchise: the deposit liability is the bank's, the cross-bank dialogue with peer banks is the bank's, the operational risk capital around the chain is held inside the institution. Outsourcing the rail to a vendor is awkward when the franchise is the rail.

Regional banks and most non-GSIB players partner with consortium rails. Partior is the longest-running APAC option, with DBS, JPMorgan, Standard Chartered, and Deutsche Bank as founding or named participants and SGD, USD, EUR cross-bank settlement. Fnality is the UK-perimeter consortium with central-bank-money-backed settlement on the Bank of England Omnibus Account model. The Japanese megabank consortia run inside Progmat. Consortium membership gives a regional bank cross-bank capability without the in-house build cost; the trade-off is governance shared with the consortium. Cross-link Build vs partner for the layer-by-layer framework.

Chain or ledger choice

Permissioned ledger almost universally. Quorum-derived stacks dominate (Kinexys runs a permissioned EVM-compatible rail; Partior is on a Quorum-derived stack). Canton has emerged as the institutional alternative for participants in the JPMorgan / Daiwa / Mizuho cohort, with Digital Asset and JPMorgan announcing the intent to bring JPMD natively to Canton in March 2026. Bank-proprietary stacks cover the remaining set (DBS Token Services, Wells Fargo Digital Cash). Public chain is rare on the deposit-side because the supervisor-perimeter implications are uncomfortable: the bank cannot freeze a permissionless chain in a reconciliation incident. The DBS-Kinexys framework is the partial exception; JPMD is natively issued on Base (Coinbase's L2), but the structural posture is that JPMD on Base is a leg the bank actively governs rather than a free-flowing public-chain instrument. The chain choice is not the load-bearing decision: reconciliation discipline, the inter-bank arrangement, and the supervisor-relationship work matter more.

Cash-leg integration

The fiat on/off-ramp into and out of the tokenised-deposit balance has to clear through somewhere. RTGS integration is the conventional answer for intra-currency on/off-ramping. For consortium rails, an omnibus account at the central bank holds the consortium's pooled reserves; the Fnality UK design references the Bank of England Omnibus Account regime for sterling. Bilateral intra-day liquidity arrangements with correspondent banks cover the cross-border leg where the home central bank does not provide a cross-currency rail. In APAC, the HKMA EnsembleTX design contemplates a wholesale-CBDC settlement layer for the inter-bank leg of HKD tokenised-deposit flows. The cash-leg integration sits at the boundary between the tokenised-deposit programme and the bank's existing payments infrastructure; most early-cohort production failures have come from this seam.

Supervisor engagement

Pre-launch dialogue with the home regulator is non-optional. The Federal Reserve and OCC for US national banks, the PRA for UK banks, MAS for Singapore-incorporated banks, HKMA for Hong Kong-incorporated banks, FSA Japan for Japanese banks. The supervisor will pause the launch on questions the bank has not anticipated; surface the questions before the formal application, not after. Cross-link Pre-engaging your regulator for the four engagement modes and the sequencing.

For tokenised deposits specifically, the supervisor conversation usually covers chain governance and incident-response, the reconciliation discipline between chain and general ledger, the AML and sanctions screening posture at chain level (not just at boundary), the operational-risk capital sizing, and the cross-bank arrangement if any. Six to twelve months of bilateral engagement before launch is the lower bound, with the longer window applying to first-time launches and cross-bank arrangements.

Capital and resolution treatment

The deposit liability sits inside the bank. Capital treatment is the same as for a vanilla deposit on the same balance sheet, with the same Liquidity Coverage Ratio (LCR) and Net Stable Funding Ratio (NSFR) treatment, the same prudential capital, and the same deposit-insurance overlay where applicable. The token is not a separate asset class for capital purposes. Basel SCO60 Group 1a treatment is available where the standard's classification conditions are met, and for a tokenised deposit issued by a regulated bank under a designated system that bar is typically straightforward; cross-link Capital treatment for tokenised assets for the broader prudential picture.

This is a feature versus the stablecoin issuer SPV path. A stablecoin issuer is a separate legal entity with its own capital, its own resolution regime, and a credit-risk profile that runs to the issuer rather than to a bank. The bank-money tokenised-deposit construction skips that whole layer. Operational risk capital against the chain itself is the additional consideration, sized to chain complexity and ops maturity.

Reference programme tour

Five named programmes worth knowing in detail.

JPMorgan Kinexys. Formerly Onyx, rebranded at Singapore Fintech Festival in November 2024. Kinexys Digital Payments runs the JPMD deposit token across USD, EUR, and GBP; cumulative notional past USD 3 trillion and average daily volume above USD 5 billion as of late 2025. The longest-running and broadest single-bank programme. Cross-bank interoperability framework with DBS announced 11 November 2025; JPMD natively on Base; intent to bring natively to Canton announced March 2026.

DBS Token Services. Singapore's first GSIB-grade tokenised-deposit programme, SGD-denominated. Sits alongside Partior in the bank's multi-rail architecture (DBS is a Partior founding shareholder). The November 2025 DBS-Kinexys interoperability framework is the cross-bank bilateral case.

Standard Chartered + Ant Whale. The 18 December 2025 production launch on Ant International's Whale platform with HKD, CNH, USD, and SGD live. The first GSIB-issued multi-currency tokenised-deposit production capability and the first to ship four currencies on a single platform. Initial use case is Ant International's own intragroup treasury. See SC + Ant Whale tokenised deposit.

Wells Fargo Digital Cash. Intra-bank tokenised dollar settlement rail under the existing US national-bank charter. Closer to the early-Kinexys shape (single-bank, USD-only) than to the multi-rail constructions Standard Chartered and DBS run. Public volume disclosure is thin.

HSBC tokenised deposit. Built on HSBC Orion infrastructure, the same stack that runs the bank's tokenised bond issuance work. Used as the cash leg in the December 2025 UK DIGIT tokenised gilt pilot for atomic delivery-versus-payment. HSBC's posture is plural: tokenised deposits, tokenised gold, and an HKD stablecoin under the Hong Kong Stablecoins Ordinance (licensed 10 April 2026, launch H2 2026) all run inside the existing licensing perimeter rather than under a single rebrand.

None of the five programmes interoperate with each other directly today; cross-link Cross-bank tokenised-cash rails for the competitive landscape.

Worked example

A regional APAC commercial bank (a Korean, Thai, or Malaysian bank with significant corporate-banking and trade-finance flows but without GSIB-scale internal infrastructure) is deciding whether to build its own tokenised-deposit programme or join a consortium rail. Walk the seven decisions.

Scope: intra-bank first, with treasury and corporate-cash use cases on a single-bank rail. Cross-bank capability comes in phase 2 via Partior membership, which lets the bank settle against DBS, JPMorgan, Standard Chartered, and Deutsche Bank counterparties without building a bilateral framework with each one.

Use case: wholesale FX settlement plus corporate treasury intra-day liquidity at launch. Cross-border B2B payments come into scope after the Partior integration in phase 2. Retail is explicitly out of scope.

Platform: partner with Partior for the cross-bank capability; build the bank-internal API and UX layer in-house so the corporate-banking team owns the customer-facing experience and the treasury-workstation integration. The cross-bank settlement primitive is shared infrastructure best operated by a consortium; the customer experience layer is bank-specific differentiation that should not be outsourced.

Chain: use the underlying Quorum-derived stack that Partior runs. The consortium's choice governs the cross-bank leg; the bank's internal layer integrates above it.

Cash leg: RTGS integration through the bank's existing correspondent relationships in each currency. The Partior consortium's cash-leg settlement architecture handles the inter-bank piece.

Supervisor: six to twelve months of bilateral engagement with the home regulator before launch, covering chain governance, reconciliation discipline, AML and sanctions posture, operational-risk capital sizing, and the consortium-membership terms.

Capital: standard deposit treatment. The deposit liability is on the bank's balance sheet under existing prudential capital. No separate Basel SCO60 consideration because the tokenised-deposit issuance is a conventional deposit liability with the chain as a transfer mechanism.

Rationale: a regional commercial bank rarely justifies the in-house build cost of a full tokenised-deposit stack. The consortium rail gives the bank cross-bank capability faster and at lower capital outlay, with governance shared with the consortium. The phase 1 / phase 2 sequencing avoids over-reach; the bank ships intra-bank first, builds operational evidence with the supervisor, and adds cross-bank capability once the supervisor relationship and the internal ops discipline are in place.

Red flags

Designing the platform without explicit pre-engagement with the home regulator. The supervisor will pause the launch on questions the bank has not anticipated, and the pause typically extends the timeline by months. Pre-engagement is not a courtesy; it is the gating workstream.
Going public chain on the deposit-side. Permissionless chains have unbounded reconciliation surfaces and uncomfortable supervisor-perimeter implications. The early production cohort is universally permissioned for structural reasons, and a deviation from that pattern needs an unusually clear rationale.
Promising cross-bank interoperability before the consortium framework is signed. Wholesale customers care about the cross-bank capability; retail customers do not notice; supervisors care about the prudential structure. Promising the capability without the inter-bank arrangement in place is a marketing line that the operating model cannot deliver.
Conflating tokenised deposit with stablecoin in customer-facing marketing. Supervisors push back; legal teams scramble; the Chapter VII Part 4 confusions cover the substantive distinction. Calling the bank's tokenised deposit a "stablecoin" in front of a treasury team or a supervisor invites a long correction.
Building before reading the foundations chapter VII parts. The underlying primitives are non-trivial, and product papers that skip the definition, issuance models, and cross-bank reconciliation reading land on the supervisor's desk with structural gaps that the supervisor will surface in the first dialogue.

[Suit Up]

Tokenising deposits in practice (for banks)