The choice between wholesale central bank digital currency (wCBDC) and tokenised deposits is not a contest between substitutes. It is a layer choice within a tiered ledger, and the architecture that has emerged from serious APAC pilots treats the two as cooperating rather than competing. This part sets out the tiered-ledger thesis, then closes the chapter with the categorisation errors that surface most often when tokenisation product teams talk to bank counsel or to regulators.
Why it matters now: the tiered ledger thesis
The tiered ledger is the architectural pattern that has emerged from the serious APAC pilots, and the choice between wCBDC and tokenised deposits is a layer choice within it rather than a contest between substitutes. Project Ensemble in hong kong is the canonical example. The HKMA operates the wholesale settlement layer in tokenised central bank money. Participating commercial banks operate tokenised deposit layers above it for client-facing flows. Tokenised assets settle delivery-versus-payment (DvP) against either layer depending on the use case, with wholesale interbank legs landing in wCBDC and corporate or fund-level legs landing in tokenised deposits. The two layers cooperate. Neither tries to do the other's job.
The same pattern shows up in the Payment Innovation Project in japan, in MAS Project Guardian, and in the conceptual framing the BIS has set out through the unified-ledger work associated with Hyun Song Shin and Tommaso Mancini-Griffoli. The Annual Economic Report 2024 chapter, and the 2023 Blueprint for the future monetary system, are the policy documents the rest of the regulatory community is reasoning from (BIS AER 2024 Chapter III; BIS Blueprint, 2023). The Basel Committee, working under Ben Gully and previously Pablo Hernández de Cos, has set the prudential frame through the basel sco60 cryptoasset standard in a way that maps cleanly onto the tier architecture: tokenised central bank claims and tokenised deposits each inherit the prudential treatment of their off-chain analogue, and operational risk capital sits on top of the chain itself. CPMI and IOSCO have been working in parallel on the cross-border standards.
For an APAC tokenisation product team, the implication is that the choice of layer is a product decision, not a religion. Interbank settlement of large-value flows wants wCBDC: zero risk weight, central-bank-backed, no large-exposure issue. Client-facing programmable money flows (payroll, conditional B2B payments, agentic transactions covered in 09 agentic commerce tokenized rails) want tokenised deposits: the client can hold them, the bank's KYC perimeter applies, and the operational cadence of the bank's chain governs availability. Cross-border depends on whether the destination jurisdiction has a wCBDC, whether bilateral or multilateral arrangements (mBridge, Project Agorá) are operative for that corridor, or whether the flow routes via correspondent banking, shared deposit ledgers, or stablecoins.
Common confusions
A tokenised deposit on a permissioned chain operated by a central bank is not wCBDC. The chain operator and the issuer are two different roles. If the central bank runs the chain but the bank issues the token, the unit is commercial bank money. The tier follows the issuer. Treating "central bank involvement somewhere in the architecture" as making something wCBDC is the most common categorisation error in this space.
Tokenisation does not change the risk weight. A tokenised deposit at Bank A is the same claim on Bank A, with the same single-name credit risk, that a non-tokenised deposit at Bank A would be. The Basel framework is explicit that tokenisation of a traditional asset that meets the classification conditions inherits the underlying treatment. Hopes that tokenising a bank deposit might shift it into a different prudential bucket are misplaced.
wCBDC is not a stablecoin. It is central bank money, with the same legal character as reserves and physical cash, simply re-expressed on a programmable ledger. The reserve composition question that dominates stablecoin discussions does not apply to wCBDC, because the central bank issues the unit directly rather than holding reserves against it. See Stablecoin types for the stablecoin taxonomy and how it sits alongside the two-tier money system.
A wholesale CBDC that allows commercial bank holders only is still a wholesale CBDC even if the chain is technically permissionless. Permissioning at the chain level is a separate question from permissioning at the holder level. A central bank could run wCBDC issuance on top of public infrastructure, with allowlist controls limiting transfer to eligible institutions; the unit would still be wCBDC. No major central bank has chosen that architecture in production, and the operational and political reasons are well understood, but the categorisation does not depend on the chain.
Atomicity is a ledger property. The fact that a tokenised deposit can settle atomically against a tokenised security is not a special property of tokenised deposits relative to wCBDC. Both instruments support atomic settlement when the architecture coordinates the legs. Marketing language that frames atomicity as a wCBDC-specific feature, or as a tokenised-deposit-specific feature, usually does not survive a careful reading of the relevant pilot's design documents.
The next part walks through where the tiered model already shows up in production, with the worked examples from Hong Kong, Singapore, Japan, and the cross-border pilots that are testing the same architecture across central banks.