Suit Up · Asia's institutional desk on tokenisation

Tokenisation looks like one product category from outside and seven different ones from inside the regulatory perimeter. A tokenised money-market fund, a tokenised deposit, a tokenised bond, a tokenised equity, a tokenised private-credit interest, an asset-referenced token, and an e-money token each route through a different statute, sit under a different supervisor, and impose a different reserve, custody, and disclosure regime. The regulators that matter (the EU through MiCA and adjacent capital-markets statutes, the US through GENIUS plus existing securities and banking law, Hong Kong through the Stablecoins Ordinance, Singapore through the SCS framework, Japan through the PSA, Korea through VAUPA and the pending DABA, Basel through SCO60) have not converged on a common asset-class taxonomy and are unlikely to. For an institutional tokenisation operator scoping product, this page is the cross-cutting reference that names which regime catches which asset class and where the boundaries are sharp versus contested.

The seven asset-class buckets

Seven buckets capture the practical regulatory map. The split is not a Basel category list, it is the operating taxonomy that determines which statute controls.

Tokenised money-market funds (MMFs). The on-chain wrapper around a 2a-7-style fund (in the US), a UCITS short-term MMF (in the EU), or an MAS-authorised collective investment scheme (in Singapore). The instrument is a fund interest, the regulator is the securities supervisor (SEC, ESMA / national CAs, MAS), and the reserve composition is set by the underlying fund regulation rather than by the tokenisation regime. BUIDL, FOBXX, OUSG, and the smaller APAC-issued tokenised MMF wrappers all sit in this bucket.

Tokenised deposits. Bank-issued claims on the issuing bank, supervised under the existing banking statute (the US National Bank Act and OCC supervision, the EU CRR/CRD and national prudential regimes, the Singapore Banking Act, the Hong Kong Banking Ordinance, the Japan Banking Act). Tokenised deposits are explicitly outside the stablecoin perimeters in every major regime that has finalised one (GENIUS §4, Hong Kong Stablecoins Ordinance scope clause, Japan PSA EPI definition, the MAS SCS framework perimeter). Kinexys BDA, DBS Token Services, JPMD, Citi Token Services, and the Wells Fargo Digital Cash programme all operate in this bucket.

Tokenised bonds. The on-chain instrument is a debt security, the regulator is the securities supervisor in the issuing jurisdiction (SEC for US-issued, the relevant national CA for EU issuance under the Prospectus Regulation, the SFC for Hong Kong, MAS for Singapore, FSA for Japan via the Financial Instruments and Exchange Act). The HKMA-supervised tokenised-bond programmes from HSBC Orion, the LSEG and Bank of England DIGIT pilot, and the Singapore-issued tokenised commercial paper running inside Project Guardian all sit in this bucket. The MiCA carve-out for MiFID financial instruments matters here: tokenised bonds are outside MiCA and inside the DLT Pilot Regime where the issuer wants a specific market-infrastructure overlay.

Tokenised equities. Same regulator as bonds (securities supervisor) but the wrapper question gets sharper. A genuinely tokenised equity (the on-chain entry is the operative share register entry) requires the issuing jurisdiction's company law to recognise the on-chain record. Most jurisdictions have not made that change, which has pushed the live programmes (Robinhood EU's wrapped equities, Ondo Global Markets) into a derivative-reference structure where the on-chain instrument is not the equity itself but a contractual claim referencing it. The legal-control analysis from 01 3 tokenisation legal control is the operating distinction.

Tokenised private credit. The on-chain instrument is an interest in a private-credit fund or a direct credit position. The regulatory wrapper is whichever the underlying fund or vehicle uses (US 3(c)(7) and Reg D for the most-cited examples, Cayman or Luxembourg fund vehicles for cross-border issuance). Apollo's ACRED, the Centrifuge V3 RWA pools, Maple Finance's syrup products, Janus Henderson's JAAA, and Ondo Finance's adjacent credit products all sit here. The operating discipline is the underlying credit cadence rather than any tokenisation-specific reserve rule.

Asset-referenced tokens (ART). The MiCA Title III category for stablecoins referenced to a basket of assets, currencies, or commodities. Issuer set restricted to credit institutions or specifically authorised legal persons under MiCA Article 21, with own-funds requirements set at the highest of EUR 350,000, two percent of average reserve assets, or one quarter of fixed overheads from the prior year (Dechert MiCA Phase 1 note). No APAC equivalent: the Hong Kong Ordinance is fiat-referenced rather than basket-referenced, the SCS framework forecloses basket designs, and the Japan PSA EPI category is single-fiat. ART is an EU-only category by current design.

E-money tokens (EMT) and the parallel APAC single-fiat regimes. MiCA's Title III EMT category for single-fiat stablecoins, with parallel constructions in Hong Kong (the Stablecoins Ordinance for HKD and other fiat-referenced designs), Singapore (the SCS framework for SGD and G10), Japan (the EPI category across bank-direct, FTSP, and trust routes), and the US (GENIUS permitted-payment-stablecoin issuer set for USD). This bucket is the most converged in substance across the major regimes (cash-equivalent reserves, monthly attestation, par redemption, no holding-period interest) and the most fragmented on issuer eligibility.

Where the regimes converge

Three points of substantive convergence across the regimes that have shipped.

Reserve composition for stablecoin-class instruments converges on cash, cash equivalents, short-dated government securities in the pegged currency, and limited bank-deposit and repo allowances. MiCA EMT, the Hong Kong Ordinance, the MAS SCS framework, GENIUS, and the Japan PSA EPI bank-direct and FTSP routes all land within a similar asset-quality envelope. The differences are at the margin (which specific securities qualify, what concentration limits apply) rather than at the structural level.

Monthly attestation as the regulatory floor. MiCA, GENIUS, the Hong Kong Ordinance, the MAS SCS framework, and the Japanese EPI framework all impose monthly disclosure of reserve composition. Some issuers go further (Circle's daily near-real-time disclosures, the on-chain proof-of-reserves work several issuers have explored), but the regulatory floor is monthly. No major regime has demanded weekly or daily attestation as a baseline.

Par redemption on demand or within a tight window. The Hong Kong Ordinance imposes one business day, MiCA and GENIUS require par redemption with disclosed procedures, the MAS SCS framework requires par redemption inside a specified short window. The structural design point is that none of the major regimes permits a stablecoin to operate on a longer-than-one-business-cycle redemption cadence.

Where the regimes diverge

Five points of substantive divergence that an operator has to navigate explicitly.

Stablecoin scope. EU and US restrict to single-fiat (EMT and GENIUS payment stablecoins respectively). Hong Kong is fiat-referenced (broader, includes single-fiat and multi-fiat designs). Singapore is single-G10. Japan is single-fiat with three issuer routes (bank-direct, FTSP, trust). The basket-referenced ART category is EU-only. An issuer scoping a multi-fiat design has the EU as the only major venue.

Tokenised-deposit treatment. Tokenised deposits are explicitly outside the stablecoin perimeter in every major regime, but the supervisory texture differs. The OCC's January 2025 Interpretive Letter 1184 confirmed that US national banks may engage in tokenised-deposit activity under the existing National Bank Act, with the deposit characterisation holding end-to-end. MAS supervises tokenised deposits as bank money under the Banking Act. HKMA does the same under the Banking Ordinance. The FSA treats Japanese tokenised deposits as bank products under the Banking Act. The supervisory approach across the four regulators is broadly aligned, with the cross-border-perimeter question (when does a deposit liability transit a public chain across jurisdictions) the open piece.

Capital treatment. Basel SCO60 is the prudential standard, with Group 1a covering tokenised traditional assets meeting specified conditions (broadly the same capital treatment as the underlying), Group 1b covering tokenised arrangements meeting weaker conditions, and Group 2 covering unbacked crypto (sharply higher capital treatment with an exposure cap). Group 1a treatment is the gating condition for a bank to hold tokenised assets at viable capital cost. The targeted review of SCO60 was endorsed by GHOS in March 2026 and is being expedited per the November 2025 Basel Committee meeting; the substantive direction of the review is not yet public.

Cross-border passporting. MiCA passports a CASP licence across the EU. APAC has no equivalent. A licensed Hong Kong stablecoin issuer cannot passport into Singapore on the basis of the HKMA licence, and a MAS-licensed payment institution cannot rely on the SCS framework status to distribute into Hong Kong. Each APAC regime is a perimeter for its own jurisdiction, and cross-border distribution requires the target-jurisdiction perimeter to be cleared in addition to the home-jurisdiction one.

Foreign-issuer comparability. GENIUS imposes a reciprocal-comparability test for foreign stablecoins offered in the US (Latham GENIUS Act analysis). MiCA imposes its own perimeter for non-EU issuers offering crypto-asset services in the bloc. The Hong Kong Ordinance follows reference currency (any HKD-referenced stablecoin needs an HK licence regardless of issuer location). Singapore's SCS framework is silent on foreign issuance distributed in Singapore (the DPT regime under the PSA picks that up separately). The asymmetric foreign-issuer rules are one of the most consequential operating frictions in the cross-jurisdictional stablecoin landscape.

What is structurally underspecified

Three asset classes where the regulatory treatment is still being built rather than read.

Tokenised bonds across the cross-border layer. Domestic tokenised-bond regimes are now reasonably well-defined (Hong Kong under HKMA and SFC, Singapore under MAS, the EU under the DLT Pilot Regime). The cross-border treatment of a tokenised bond issued in one jurisdiction and held by an institutional investor in another has not been comprehensively addressed by any major regulator. The Project Guardian fixed-income workstream and the Project Ensemble cross-border tokenised-asset settlement work are the closest things to a worked example.

Tokenised equities. The legal-control problem remains unresolved in most jurisdictions. The EU's DLT Pilot Regime allows for on-chain share registers under specific conditions but uptake has been thin. The US has not made the company-law changes that would let the on-chain register be the operative one. The result is a market dominated by derivative-reference structures (Robinhood EU's wrapped equities, Ondo Global Markets) where the on-chain instrument is a contractual claim rather than the equity itself, with the operating regulatory wrapper being the broker-dealer or fund-vehicle rules rather than the equity-issuance rules.

Tokenised private credit. The asset class sits inside whatever the underlying fund or vehicle wrapper imposes (Reg D for US issuance, Cayman or Luxembourg for cross-border), with no tokenisation-specific reserve or attestation regime. The composability story (Aave Horizon's institutional-DeFi RWA work, the Centrifuge V3 deRWA pattern) is layering venue-level KYC on top of the existing wrapper rules. Whether a regulator eventually designates a tokenisation-specific perimeter for tokenised credit composable in DeFi is one of the open questions the institutional composability paradox page tracks.

Operating implications

For an institutional tokenisation operator picking product, the asset-class question is the first regulatory cut and the most consequential. A tokenised MMF, a tokenised deposit, and a single-fiat stablecoin can look superficially similar (cash-equivalent on-chain liability, daily yield or par redemption, multi-chain distribution) and route through three completely different supervisory perimeters with three different reserve, custody, and disclosure regimes.

The cross-jurisdictional cut is the second consequential question. A US-issued tokenised MMF distributed into Singapore institutional channels needs to clear the SEC perimeter for issuance and the MAS perimeter for distribution. A Hong Kong-licensed stablecoin distributed into Japan needs to clear the HKMA perimeter for issuance and the FSA EPI intermediary perimeter for distribution. The lack of passporting in APAC is the recurring operating constraint.

The capital-treatment cut sits behind both. Basel SCO60 determines whether a bank counterparty can hold the tokenised instrument at viable capital cost, regardless of which underlying perimeter the instrument was issued under. The bank-side reception of tokenised products is gated by Group 1a versus Group 2 classification under the existing standard, with the targeted review the load-bearing process for any horizon longer than six months.

Open questions

Whether MiCA's ART category gets any APAC analogue. The current APAC regimes have all foreclosed multi-fiat designs; whether a regulator eventually carves out a basket-referenced perimeter is the open piece.
The substantive direction of the Basel SCO60 targeted review. Will the Group 1a / Group 2 boundary be revisited or only the calibration around it.
Whether GENIUS comparability determinations will be granted to MiCA-compliant EMT issuers, and to the major APAC regimes (HKMA, MAS SCS, Japan EPI). The asymmetric foreign-issuer treatment is one of the biggest operating frictions in cross-border stablecoin distribution.
How the tokenised-equity legal-control problem resolves across major jurisdictions. The EU's DLT Pilot Regime is the closest worked example of a permissive perimeter; whether the US, Hong Kong, Singapore, or Japan moves their company-law treatment of on-chain registers in a comparable direction is the structural question.
The Korean DABA outcome on stablecoin issuer eligibility (deadlock between FSC and BoK on bank-ownership thresholds). Phase 2 will determine whether KRW joins the regulated single-fiat stablecoin perimeter or remains structurally outside.

[Suit Up]

Asset-class regulatory treatment across tokenisation regimes