Hong Kong virtual asset licensing is moving beyond exchanges. On 20 November 2026, the Government gazetted a bill that would bring virtual asset dealing, advisory, management and custody services under licensing or registration with the Securities and Futures Commission. That change matters because a digital asset market is not only a venue where buyers and sellers meet. It is a chain of recommendations, mandates, private key controls, settlement instructions, liquidity relationships and customer records. A regime that supervises only the trading platform can therefore regulate the visible marketplace while leaving important decisions outside the same perimeter.
The proposed law is best understood as an attempt to close those gaps. Hong Kong already regulates centralised virtual asset trading platforms, and the Hong Kong Monetary Authority already operates a separate regime for fiat referenced stablecoin issuers. The new bill would connect four additional service functions to that architecture. It would also require banks and stored value facility licensees supervised by the HKMA to register with the SFC if they want to provide covered virtual asset services. The result would be a market in which the legal character of an institution does not automatically remove the crypto function from specialist supervision.
That does not mean the bill is already law, that every detail is settled, or that licensing will guarantee safety. The bill is scheduled for First Reading in the Legislative Council on 2 December. Commencement, transition arrangements, detailed conduct standards and the eventual shape of implementation still matter. The deeper thesis is more important than the immediate headline: Hong Kong is testing whether regulation can make the service chain more coherent without making entry so expensive that the market consolidates around a small group of institutions.
What Hong Kong Actually Gazetted
The Hong Kong Government announcement says the Virtual Asset Services Legislation Amendment Bill 2026 would amend the Anti Money Laundering and Counter Terrorist Financing Ordinance. Any person carrying on a business of providing virtual asset dealing, advisory, management or custody services in Hong Kong would need to be licensed by or registered with the SFC and satisfy a fit and proper test plus other applicable requirements. The announcement also says the SFC and HKMA would receive powers to implement the regime according to their statutory responsibilities.
Three limits are essential. First, this is proposed legislation. The announcement says the bill will receive its First Reading on 2 December, so market participants should distinguish a gazetted proposal from an enacted and commenced regime. Second, the short announcement does not provide every operational rule. Capital requirements, transition periods, detailed exemptions, client classification, outsourcing controls and reporting expectations depend on the bill, later rules and regulatory guidance. Third, the four activities are related but not interchangeable. A dealer executes or arranges transactions, an adviser recommends action, a manager exercises discretion over assets, and a custodian controls access to property. Treating them as one generic crypto activity would hide the different ways a client can be harmed.
The proposal extends a process that began earlier. The SFC explains that centralised virtual asset trading platforms carrying on business in Hong Kong, or actively marketing to Hong Kong investors, are already required to be licensed and regulated. Under the existing framework, a platform can combine trading with custody that is ancillary to its venue. The new bill addresses activities that can exist outside that venue. A broker may route a client order without operating an exchange. A wealth manager may allocate to tokens without running an order book. A specialist custodian may hold keys for assets traded elsewhere. An adviser may influence the economic decision while never touching a token.
This is why the bill is a perimeter reform rather than an exchange rule. It asks which function is being performed, who controls the client relationship and where the relevant risk sits.
Hong Kong Virtual Asset Licensing Maps the Service Chain
The four proposed categories describe a sequence that can begin before a transaction and continue long after settlement.
| Service | Core function | Main risk channel | Evidence that matters |
|---|---|---|---|
| Dealing | Executing, arranging or intermediating transactions | Conflicts, pricing, routing, counterparty selection and market access | Execution records, best execution logic, fee disclosure and counterparty controls |
| Advisory | Recommending a transaction or strategy | Suitability, incentives, incomplete disclosure and narrative bias | Client profile, recommendation basis, conflicts and risk explanation |
| Management | Exercising discretion over a portfolio | Mandate drift, concentration, valuation, liquidity and hidden leverage | Mandate limits, valuation policy, risk reports and governance |
| Custody | Safeguarding assets or controlling private keys | Theft, loss, unauthorised transfer, insolvency and recovery failure | Key governance, segregation, approvals, reconciliation and recovery procedures |
The table shows why an exchange licence cannot solve every problem. A customer can receive a poor recommendation from a regulated adviser, obtain execution through a licensed platform and still suffer loss because a separate custodian failed to segregate assets. A manager can select sound assets but expose the portfolio to weak liquidity or an undisclosed related party. A dealer can route transactions to a venue that appears liquid while receiving incentives that the client does not understand. Regulation must follow the decision and control chain rather than stop at the interface where an order is displayed.
This functional logic also explains the proposal for banks. A bank is already subject to prudential supervision, but crypto activity creates additional questions about tokens, private keys, blockchain analytics, wallet architecture, transaction finality and interactions with public networks. Requiring an HKMA supervised bank to register with the SFC for covered services does not imply that bank supervision is inadequate. It recognises that the same institution can carry different risk functions that require coordinated expertise.
Hong Kong has already applied this layered approach to stablecoins. The HKMA stablecoin regime requires a licence for issuing fiat referenced stablecoins in Hong Kong. The issuer regime focuses on the liability, reserves, redemption and monetary stability. The new service regimes focus on how intermediaries deal, advise, manage and hold virtual assets. One licence should not be mistaken for permission to perform every adjacent activity.
That distinction is commercially important. An issuer can create a well governed token, yet distribution may still depend on dealers, platforms, advisers and custodians. A regulated dealer can provide access, yet the token may have weak redemption, uncertain legal claims or thin secondary liquidity. The service chain is only as strong as the interfaces between its separate legal and operational parts.
From Platform Regulation to Functional Regulation
The existing platform regime was a logical starting point. Centralised venues concentrate customer accounts, market access, order matching and often custody. They are visible control points. Hong Kong’s response to the JPEX episode also showed why platform claims, promotions and customer access cannot be treated as a purely technical question. Our analysis of the JPEX crypto fraud case examined how alleged unlicensed activity, marketing and the difficulty of tracing assets can combine into a broad investor protection problem.
Yet platform regulation creates a boundary problem. Services can be separated deliberately or emerge separately as the market matures. Institutional investors may use an external custodian rather than leave assets with a venue. A family office may appoint a manager that sends orders to several platforms. A bank may provide advice while another company executes. A wallet provider may offer software, recovery, routing and hosted services through different entities. Each arrangement can be legitimate, but the legal perimeter needs to identify which function creates which obligation.
Functional regulation reduces the incentive to design around labels. If custody is regulated only when performed by an exchange, a company might avoid exchange status while still controlling client keys. If advice is regulated only when attached to a securities product, a promoter might present a specific token strategy as education even when the economic function resembles a recommendation. A strong regime does not prohibit communication or self custody. It asks whether the business is exercising professional influence, discretion or control for clients.
The SFC and Financial Services and the Treasury Bureau made this direction clear before the bill. Their consultation conclusions for advisory and management services state that the proposed scopes would align with the conventional frameworks for advising on securities and asset management. The authorities reported 51 consultation responses and said the proposal received broad support. The comparison with traditional finance does not erase the technical differences of crypto. It supplies a familiar starting point for conflicts, competence, governance and client protection.
The earlier dealer and custodian consultation conclusions applied the same logic. Dealer rules were intended to resemble the framework for dealing in securities, while custody would focus on risks related to safeguarding private keys and client virtual assets. The important design choice is not to force every service into an exchange model. It is to adapt established conduct principles to the control structure of digital assets.
Custody Is Where Legal Rights Meet Cryptographic Control
Custody deserves special attention because crypto creates an unusually direct link between technical authority and economic loss. A private key, or a distributed signing process that can produce the same authority, may be sufficient to move an asset irreversibly. But possession of signing power does not answer the legal questions. Who owns the asset if the custodian fails? Is the client recorded as beneficial owner? Are assets segregated onchain, in internal books, or both? Can the custodian use a subcustodian? How are forks, airdrops, staking rewards and frozen tokens treated? What happens when law, protocol rules and the customer agreement point in different directions?
The custodian consultation conclusions reported 93 submissions and broad support for regulation following the platform regime. That level of engagement reflects the fact that custody is not a narrow wallet feature. It is an operational system involving access policies, people, hardware, software, reconciliation, legal agreements, incident response and recovery.
Our examination of Deutsche Bank’s crypto custody architecture reached the same conclusion from the institutional side. Moving crypto for clients must become a controlled financial process, not an informal wallet operation. A bank brand can support distribution and governance, but it does not remove the need to understand signing authority, subcustody, asset segregation and service boundaries.
Technology can strengthen this system without replacing governance. In the Blockstream and Thales custody design, distributed signing, hardware protection and policy controls were parts of one architecture. The critical insight was that multiple signature shares are not meaningful separation if the participants depend on the same administrator, credentials, cloud account or recovery secret. Regulation should therefore test control independence in substance, not count technical components.
Hong Kong’s challenge will be to translate that principle into proportionate requirements. A specialist institutional custodian, a bank, an exchange affiliate and a smaller technology provider may control assets in different ways. Rules need enough specificity to protect clients while allowing secure architectures that do not all look identical.
Advice and Management Regulate Influence Before Transfer
Advisory and management services address risk earlier in the chain. The client may suffer harm before any wallet instruction exists because the investment thesis, asset selection or portfolio structure was flawed.
Advice creates several conflicts that are especially acute in crypto. A provider may be paid by an issuer, a venue or a market maker. A token can generate affiliate revenue or improve the economics of another product. Historic returns can be presented without liquidity conditions, survivorship bias or drawdown context. A recommendation can describe network adoption while ignoring whether the token captures value. Technical complexity also lets marketing language appear analytical when the underlying evidence is weak.
Management adds discretion. The manager can decide position size, venue, custodian, staking provider, liquidity strategy and timing. That creates risks involving mandate drift, concentration, valuation and operational dependence. A portfolio that holds ten tokens can still be concentrated if they depend on the same stablecoin, bridge, custodian or market maker. A strategy can appear liquid during normal conditions but become difficult to exit when collateral falls and automated liquidations accelerate.
The proposed regimes are therefore not simply credentials for professionals. Their value depends on conduct standards, evidence and supervisory access. The record should show why a recommendation suited the client, how conflicts were handled, how a portfolio stayed inside its mandate and which assumptions supported valuation. The objective is not to eliminate loss. It is to make professional decisions accountable.
This is also where regulatory labels can mislead. A licensed adviser can recommend a high risk asset. A licensed manager can produce a negative return. Permission to provide a service is not a guarantee about the future value of a token. The regime can improve competence, disclosure and recourse, but it cannot convert uncertainty into safety.
The Competitive Effect May Be Consolidation
Comprehensive licensing can improve trust and still reduce the number of providers. Compliance requires experienced staff, systems, audits, capital, insurance, legal advice, monitoring and senior management attention. Those costs are easier for banks, large exchanges and established asset managers to absorb. Smaller firms may merge, specialise, become technology suppliers or leave the jurisdiction.
Consolidation can help when it removes weak operators and concentrates responsibility. It can also create new dependencies. If a small number of custodians hold most institutional assets, a technical incident or policy error can affect the entire market. If a few banks control fiat access, regulated liquidity can become expensive or selective. If advisers rely on the same research and approved product list, portfolio behaviour can become crowded.
Hong Kong should therefore judge the regime by market structure as well as licence count. A high number of applications can show interest, but it does not prove that competition is healthy. A small number of strong firms can improve oversight, but it can also make switching difficult. Useful measures include concentration by client assets, custody provider, banking relationship, market maker and stablecoin settlement route.
The experience of stablecoin licensing illustrates the tradeoff. The HKMA said it received 36 applications in its first batch and granted two initial licences in April 2026. Its explanation emphasised high standards, credible use cases and sustainable development. Our analysis of HSBC RedCoin argued that distribution can become the moat. A bank can connect a regulated token to existing customers and payment infrastructure, but that strength also means regulation may reinforce institutions that already control valuable channels.
The same pattern could appear across the new services. The decisive question is whether regulation creates open, interoperable access among licensed firms or builds several vertically integrated groups that internalise dealing, advice, management and custody.
Cross Border Activity Will Test the Perimeter
Crypto services are delivered across borders by design. A customer in Hong Kong can use an interface operated elsewhere, trade a token issued in another jurisdiction, settle through a global stablecoin and hold assets in a wallet whose infrastructure spans several countries. The blockchain transfer is global, while the legal relationships remain local.
The regime must therefore define when a service is carried on in Hong Kong and when overseas marketing creates an obligation. Consultation materials have already focused on active marketing and the treatment of existing providers. These boundaries are not minor drafting details. They determine whether an offshore company can reach Hong Kong users without equivalent supervision and whether a local licensed firm competes on comparable terms.
International standards offer a floor rather than a complete answer. The Financial Action Task Force guidance says virtual asset service providers should be licensed or registered and subject to effective monitoring for anti money laundering and counter terrorist financing compliance. Hong Kong’s proposal goes further by connecting that financial crime perimeter with conduct and investor protection concepts familiar from traditional financial services.
Cross border coordination remains difficult. A firm can satisfy customer due diligence rules while its foreign custodian follows different segregation standards. One authority may regulate the issuer while another supervises the dealer. A token can move outside the licensed environment after withdrawal. The regime can make local intermediaries more accountable, but it cannot make every foreign rule identical.
This is why the Türkiye crypto licensing monitoring test is relevant. Formal registration and custody separation are meaningful, yet global tokens, foreign venues and self controlled wallets continue to connect the domestic market with external risks. Licensing is the beginning of observation, not the end of market structure analysis.
What the Bill Could Improve
The constructive case for Hong Kong virtual asset licensing rests on four mechanisms.
First, it can make responsibility easier to locate. Clients should know whether a firm is acting as dealer, adviser, manager or custodian and which authority supervises that function. Clear role disclosure reduces the chance that a company uses one licence to imply approval for unrelated services.
Second, it can improve continuity. Custodians and managers should have documented procedures for incidents, departures, insolvency and asset transfer. A client should not discover during a crisis that the service depends on one employee, one signing device or one inaccessible record.
Third, it can improve evidence. Supervisors can require records of recommendations, execution, asset control and conflicts. That evidence supports both ongoing monitoring and reconstruction after harm. Crypto transactions can be visible onchain, but public transaction data rarely reveals the complete customer instruction, legal owner, recommendation or internal approval.
Fourth, the regime can make institutional partnerships easier. Banks, funds and payment companies are more likely to connect with crypto providers when the counterparty’s status, governance and control obligations are legible. The Government’s November policy update reported 17 retail tokenised investment products with tokenised classes holding about HK$10.3 billion at the end of September 2026, plus two licensed stablecoin issuers. Those figures show that Hong Kong’s digital asset strategy already extends beyond speculative token trading. A broader service perimeter is intended to support that institutional expansion.
None of these benefits is automatic. The quality of licensing depends on supervision after approval, enforcement against unlicensed activity, coordination between the SFC and HKMA, and the ability to update rules as technology changes.
Risks, Limits and Common Misreadings
The first misreading is that the bill bans self custody. The Government announcement addresses businesses that provide dealing, advisory, management or custody services. Holding one’s own keys is conceptually different from operating a custody business for clients. Final legal definitions still matter, but readers should not convert a service licensing proposal into a general claim that private wallets are prohibited.
The second misreading is that all crypto assets would become approved investments. Licensing governs service providers. It does not validate every token, protocol or strategy. A client can access a volatile, illiquid or poorly designed asset through a regulated chain.
The third misreading is that more licences necessarily produce more liquidity. Regulation can attract institutions, but liquidity also requires willing buyers and sellers, reliable settlement, market makers, banking access and confidence in the underlying asset. Compliance can reduce uncertainty while increasing operating costs. Both effects influence spreads and depth.
The fourth misreading is that one regulator will control the whole system. The bill explicitly contemplates roles for both the SFC and HKMA. Issuers, banks, platforms and service providers can sit under related but distinct frameworks. Coordination will be a strength only if responsibility is clear during routine supervision and crises.
The fifth limit is technology risk. A fit and proper test can assess people and institutions, but it does not prove that smart contracts, wallet software or signing systems are free from defects. Supervision can require testing, governance and incident response. It cannot make complex systems incapable of failure.
Finally, regulation can move activity rather than remove it. Some providers may stop serving Hong Kong, restrict product access or shift functions offshore. Users may then choose less transparent alternatives. Enforcement, public education and competitive licensed services must work together.
Three Scenarios for Hong Kong’s Digital Asset Market
Base scenario: a broader but concentrated licensed market
The bill proceeds through the Legislative Council, the authorities publish transition and conduct rules, and established providers begin pre application work. Banks, major exchanges, asset managers and specialist custodians obtain or prepare for the relevant permissions. The market gains clearer service boundaries, but compliance cost limits the number of full service operators. Partnerships expand gradually, especially around stablecoins, tokenised funds and institutional custody.
Evidence for this scenario would include orderly legislative progress, detailed implementation guidance, early engagement with existing providers, several credible applicants in each service category and no abrupt loss of customer access.
Constructive scenario: regulation creates an interoperable service network
Hong Kong combines high standards with practical recognition of different business models. Specialist firms can connect through clear outsourcing and delegation rules. Banks provide fiat and settlement access without forcing every service inside one group. Custodians support secure transfers among venues and managers. Advice and management rules improve institutional confidence. Liquidity grows because more participants can transact through accountable providers.
Evidence would include diverse licence holders, transparent service partnerships, falling spreads in regulated markets, growth in client assets without excessive concentration, and successful incident management. Product growth would be accompanied by clear redemption, custody and legal structures.
Adverse scenario: costs create closed vertical silos
Detailed requirements make independent operation uneconomic for smaller providers. A few groups combine dealing, advice, management and custody. They gain strong control over distribution and data. Product choice narrows, switching becomes difficult and liquidity fragments between approved networks and offshore alternatives. Unlicensed marketing continues through digital channels, leaving compliant firms with high costs while risk migrates elsewhere.
Evidence would include applicant withdrawals, rising concentration, restrictive banking access, wide pricing differences between local and global markets, reduced support for self directed transfers and repeated enforcement against offshore solicitation.
Indicators That Will Reveal the Real Outcome
The legislative calendar comes first. The 2 December First Reading is the next stated step, but later scrutiny, amendments, passage and commencement determine when obligations become real.
Transition arrangements will show whether authorities can protect clients without forcing abrupt exits. Existing providers need enough time to apply, restructure or transfer customers. Temporary status should not become a loophole, but immediate cessation can create operational risk.
Licence and registration data will show the competitive structure. Count applicants and approvals by service, but also measure common ownership and outsourcing relationships. Ten licensed brands can still depend on one custodian and two banks.
Client asset concentration will show whether custody risk is diversifying or accumulating. Public data on segregation, subcustody and incidents would improve market discipline.
Market quality matters. Track spreads, depth, execution reliability and access to banking and stablecoin settlement. More regulated entities should eventually improve usable liquidity, not merely the number of compliance announcements.
Enforcement will reveal whether the perimeter is credible. Authorities need to address false claims of licensing, unlicensed active marketing and firms that perform a regulated function under another label.
Operational incidents will test supervision. The important questions are whether providers detect failures, preserve assets, communicate clearly, recover service and supply evidence to regulators and clients.
Block2Learn Assessment
Hong Kong’s proposal is directionally strong because it regulates the service chain rather than treating the exchange as the whole crypto market. Dealing, advice, management and custody create different forms of influence and control. Bringing them into a coordinated perimeter can make responsibilities clearer and support institutional participation.
The decisive issue is implementation. A comprehensive regime can become either a network of accountable specialists or a set of closed institutional silos. The first outcome supports competition, resilience and innovation. The second may improve formal compliance while concentrating operational power and distribution.
The authorities should therefore evaluate more than the number of licences granted. They should monitor common dependencies, client asset concentration, outsourcing, banking access, cross border marketing and the ease with which customers can transfer assets among lawful providers. The goal should be a market where legal clarity improves mobility rather than trapping users inside one group.
Investors and users should apply the same discipline. Check which entity performs each function, what the licence covers, who controls keys, how assets are segregated, which law governs the relationship, and how the service can be exited. A regulated label is useful evidence, but it is not a substitute for understanding the control chain.
Conclusion: The Licence Is Only as Strong as the Links Between Services
Hong Kong virtual asset licensing has reached a more mature question. The city is no longer deciding only whether crypto platforms should be supervised. It is deciding how the full chain of dealing, recommendation, discretion and custody should fit together.
The 20 November bill creates a credible route toward a clearer market. It recognises that banks and crypto specialists can perform similar functions, that investor harm can occur before and after execution, and that private key control needs legal and operational governance. It also carries real risks involving cost, concentration and cross border migration.
The measure should therefore be judged by outcomes after legislation: whether clients can identify responsibility, whether specialists can compete, whether assets remain transferable, whether incidents are contained and whether regulated liquidity becomes deeper. The important transformation is not the creation of four additional permissions. It is the attempt to make the separate links of a digital asset market answerable to one coherent standard of conduct.
Continue Through the Block2Learn Learning Path
Understanding Hong Kong’s proposal requires more than memorising licence names. Readers need to connect regulation with market structure, custody, incentives, liquidity and operational risk. Free Start introduces the language of assets, risk and financial systems. Foundation develops the ability to compare claims, intermediaries and capital allocation. The Investor Operating System turns those concepts into a repeatable process for evaluating evidence rather than reacting to regulatory headlines.
The Trading layer explains execution, venue quality and liquidity. The Crypto layer goes deeper into wallets, private keys, token design, stablecoins, protocols and the boundary between network use and token value. Wealth Strategy connects asset access with portfolio construction, while the Framework integrates the separate disciplines into a coherent decision system.
Continue through the Block2Learn Learning Path to build the structure needed to evaluate how a licence, a service provider and a blockchain interact.
Information is abundant. Structure is rare.
This article is provided solely for informational and educational purposes and does not constitute financial or investment advice, a recommendation, or an offer or solicitation to buy or sell any financial instrument or digital asset. See our Financial Disclaimer.
This article was generated with the support of AI and reviewed by the Editorial Team. For more information, see our Terms of Service.









