Operating a digital-asset business across borders without a deliberate holding structure is one of the most expensive decisions a founder can make. The corporate entity, the jurisdiction where tokens are issued, the location of the treasury, the residency of the principals — each of these creates a tax event independently. Together they create a compounding exposure that no personal relocation alone can unwind. Crypto holding structure decisions are tax decisions, licensing decisions and exit decisions made simultaneously, not sequentially. This analysis maps the major jurisdictions, contrasts the structural options, and identifies the decision axes that matter most for a digital-asset operator choosing where to anchor a group.
Why Structure Decisions Come Before Everything Else
Getting the holding structure right before the first revenue event is the single most consequential step in the lifecycle of a digital-asset business. Once tokens circulate, once staking rewards accrue, once a secondary market forms, the jurisdictional character of each gain is already determined. Reversing a structure after the fact — through a migration, a redomiciliation or a founder relocation — is possible, but it triggers its own layer of exit taxes, controlled-foreign-corporation attributions and transfer-pricing scrutiny that may cost more than the original misstructure.
In our cross-border practice, we regularly advise founders who separated the personal and corporate decisions. They moved themselves to a low-tax jurisdiction and assumed the company's tax position followed. It rarely does. A holding company incorporated in a favorable jurisdiction that is effectively managed and controlled from a high-tax state — because the founder still chairs board calls from home — is treated as resident in that high-tax state under the laws of most OECD members. The corporate structure is only as effective as the substance that underlies it.
The relevant regimes here span the full spectrum: the EU's MiCA (the Markets in Crypto-Assets Regulation) framework, which imposes its own authorisation costs that interact with tax planning; Singapore's Payment Services Act and the MAS licensing regime; VARA in Dubai; ADGM in Abu Dhabi; FINMA in Switzerland; and the FCA regime in the UK. Each jurisdiction's tax treatment of digital-asset holding entities differs in material ways.
The structure question is always four questions in one: where does the holding entity sit, where is economic substance, where are the founders resident, and what is the exit plan? These four axes must resolve consistently or the structure fails its primary purpose.
A common assumption is that relocating personally is enough to change the group's tax position. This objection surfaces in almost every initial instruction we receive. The reality: personal tax residency changes the founder's personal liability on distributions and on gains in the entity's equity. It does not change the entity's own tax residency, the withholding tax on intercompany payments, or the treatment of intellectual property and treasury assets held above the operating level. Structure and residency must be designed together — or not at all.
To map the holding structure, banking and licensing stack for your specific build, contact OBOLUS at info@oboluslaw.com. The process above describes the standard analysis. Your facts — the entity chain, the user base, the token mechanics, the banking relationships — change the answer materially.
How the Major Jurisdictions Compare for Holding Entities
No single jurisdiction is universally optimal; each presents a different trade-off between tax efficiency, regulatory burden, substance requirements and exit flexibility. The right answer depends on the operator's profile, not on a ranking of tax rates.
The UAE has become the most discussed holding jurisdiction for digital-asset businesses over the past several years. Dubai's VARA regime and the ADGM framework in Abu Dhabi both provide regulatory homes for crypto-asset service providers. From a tax perspective, the UAE imposes no personal income tax and a corporate tax that — for qualifying free-zone entities — operates at a rate reported as zero on qualifying income under current free-zone incentive regimes, subject to substance requirements and the conditions attached to a qualifying income determination. Those substance requirements are real: directors must be physically present, board meetings must occur in-jurisdiction, and key decisions must demonstrably be made locally. An entity that holds a VARA licence in Dubai but is directed from London will not preserve its tax position.
Singapore offers a credible alternative for Asia-Pacific-focused groups. The MAS Payment Services Act provides a licensing path for digital payment token service providers. Singapore's corporate tax regime includes an exemption or partial exemption on foreign-sourced income under certain conditions, and the jurisdiction is a major hub for family office and fund structuring. Substance requirements under the Economic Substance framework apply to relevant activities including fund management and intellectual property. Singapore's value for a holding structure increases significantly when the principal group activity — exchange operations, fund management or token issuance — is genuinely run from the jurisdiction.
Switzerland remains attractive for token issuance specifically. FINMA's token taxonomy — distinguishing payment tokens, utility tokens and asset tokens — provides relative clarity for structuring the issuance vehicle. Swiss cantonal tax regimes vary considerably; cantons such as Zug have historically attracted crypto-native structures. Federal and cantonal tax together produce effective rates that vary by canton, activity and income type, so any Swiss structure requires a cantonal analysis alongside the federal framework.
Cayman Islands and BVI serve different purposes. A Cayman-domiciled fund holding digital assets benefits from the CIMA licensing regime under the Virtual Asset (Service Providers) Act and from Cayman's established exempt-company and exempted-limited-partnership regimes, which produce no corporate-level tax on most income. The BVI's FSC administers its own VASP Act 2022 regime. Both are used more as vehicles for the fund layer or the IP holding layer rather than as the entity that sits closest to the token-issuance function — because neither provides a domestic market or regulatory equivalence for a business seeking EU or UK distribution. They layer above or below an operating entity in a jurisdiction with a bilateral treaty network.
Lithuania and Malta provide the EU entry point. Under MiCA, a CASP authorised in one member state passports across the EU/EEA. Lithuania has historically been a relatively accessible EU registration jurisdiction for VASPs; Malta's MFSA-supervised VFA framework is transitioning to MiCA CASP authorisation. Neither Lithuania nor Malta offers a tax environment as low as the UAE or Cayman, but for a business that needs EU regulatory status, the relevant comparison is not against offshore structures — it is against the cost and timeline of obtaining MiCA authorisation in Germany, France or the Netherlands directly.
What Actually Drives the Structural Choice?
Three factors dominate the comparative analysis for most digital-asset operators: the nature of the group's primary income, the intended exit event, and the regulatory licences already held or planned.
Income type matters because most jurisdictions treat different categories of digital-asset income differently. Trading gains may be capital gains or revenue income depending on the jurisdiction and the holding period. Staking rewards may be income at receipt, capital on disposal, or neither — depending on how the relevant tax authority characterises the nature of the reward. Fees from exchange or custody services are almost universally ordinary income in the hands of the operating entity. A structure that is efficient for capital appreciation may be inefficient for fee income, and vice versa.
Exit planning is the most under-weighted factor in early-stage structuring. A founder who intends to sell the business to a strategic acquirer in five years needs a holding layer that produces a capital gains event — ideally in a jurisdiction that taxes that gain at a lower rate, or not at all — rather than an income event. A holding company that has directly issued tokens and earned protocol fees may find that an acquirer values those assets differently from equity in a clean holding vehicle. Separating the IP and treasury layer from the operating layer early is substantially easier than doing it at the point of exit.
We have seen this dynamic play out in practice. In a recent structuring engagement, a token-issuance group operating across multiple jurisdictions had placed its primary IP holding function in an operating entity in a high-tax EU member state. When acquisition discussions began, the proposed purchase price reflected an enterprise value that included embedded taxable gains at the operating entity level. Restructuring before exchange of contracts was technically possible but required a multi-month transfer-pricing exercise and a tax clearance process that compressed the deal timeline and introduced regulatory notification obligations under the MiCA framework's material-change provisions. A two-year earlier restructure would have addressed all of this at a fraction of the cost.
Regulatory licences create structural constraints. A VARA licence is granted to the entity that holds it. If the operating entity is the VARA licensee, it cannot be held in a Cayman SPV without VARA's approval of the ownership structure, which VARA scrutinises closely. FINMA and the MAS both require that the licensed entity maintain genuine substance in-jurisdiction. Structuring purely for tax without mapping the licensing permissions that attach to each layer is a common source of structural failure.
Decision Matrix: Which Profile Points Where
The right structure for a given operator depends on the combination of primary activity, founder location, intended user base and exit horizon. The following analysis maps four common profiles to their structural implications.
Profile A — Exchange operator, global user base, founder relocating to UAE: The natural structure anchors the operating entity under the VARA regime in Dubai or the ADGM framework in Abu Dhabi. The holding company may sit in the UAE free zone or, if treaty protection for cross-border cash flows matters, in a jurisdiction with a strong treaty network that interoperates with the UAE. The founder's personal residency in the UAE resolves the personal income tax question on distributions and equity gains, provided genuine UAE residency is established and maintained — not merely a visa. Key risk: if the exchange also serves EU users at scale, a MiCA-passportable entity may be required as a subsidiary, creating a second regulatory layer in a higher-tax jurisdiction. The group tax position is a blend, not a single rate.
Profile B — Token issuer, EU distribution planned, founders in EU: The natural structure uses a MiCA CASP authorisation in a favorable EU member state — Lithuania and Malta are the current lead options — with a Cayman or BVI holding layer for the treasury and the founders' equity. The token-issuance vehicle should be the MiCA CASP entity because token issuers of asset-referenced tokens and e-money tokens face specific authorisation requirements under MiCA that attach to the issuing entity directly. Personal tax: EU-resident founders face the full scope of EU member state personal income tax; treaty planning for distributions from the holding layer is the primary lever. Indicative timing from instruction to operational MiCA authorisation: varies by member state and the completeness of the application file, but experienced operators consistently describe a process measured in months, not weeks.
Profile C — DeFi protocol / treasury-holding LLC, no regulated service, founders with flexible residency: The cleanest structure for a founder with genuine flexibility is a Cayman exempted company or a BVI BC for the protocol treasury, with the founder establishing substantive residency in the UAE, Singapore or Switzerland. The critical question is whether the protocol's activities — liquidity provision, fee generation, governance — constitute regulated activities in any of the jurisdictions where its users sit. If they do, the structure must accommodate a regulated subsidiary. If they do not, the primary planning lever is the founder's personal tax position, which turns on establishing a genuine change of domicile and severing taxable connections to the prior high-tax jurisdiction. This process is more demanding than most founders anticipate: it involves social security analysis, pension tax, the treatment of unvested equity granted before the move, and the application of anti-avoidance rules targeting individuals who move shortly before a gain crystallises.
Profile D — Institutional fund investing in digital assets, LP base in multiple jurisdictions: The fund layer is typically Cayman or BVI, using the established limited-partnership structures recognised by institutional allocators. The management entity — which earns the performance fee and the management fee — should sit in a jurisdiction that offers regulatory recognition (Singapore, Switzerland, UK, ADGM) and a favorable carried-interest treatment. The management entity's jurisdiction is the primary tax-optimisation lever at the fund level. LP-level withholding tax on distributions turns on the treaty position between the fund jurisdiction and each LP's home state, which requires analysis at the investor-relations stage.
Substance Requirements: Where the Structure Lives or Dies
Substance requirements are the single most common point of structural failure for digital-asset groups that structure primarily for tax without concurrent regulatory analysis. Every major jurisdiction — UAE, Singapore, Cayman, BVI, Switzerland — now imposes substance requirements that must be satisfied for the entity to claim the tax or regulatory status it was designed to achieve.
The OECD's Base Erosion and Profit Shifting project, specifically the work on harmful tax practices and on country-by-country reporting, has changed the practical requirements significantly. An entity in a low-tax jurisdiction that cannot demonstrate that key decisions are made there — board meetings, investment decisions, risk-management approvals — faces the risk that a high-tax jurisdiction will claim the entity's profits under its controlled-foreign-corporation rules or under the concept of effective management and control.
In our practice, the substance question resolves into four operational checks. First: are the directors physically present in the jurisdiction a majority of the time? A UAE free-zone entity run by a director who spends most of the year in Germany is not a UAE-resident entity for most purposes. Second: are board meetings held in-jurisdiction, minuted, and supported by contemporaneous records showing that the decision was made there and then? Third: is the company's accounting, banking and operational infrastructure genuinely located in the jurisdiction, not merely registered there? Fourth: do the employees or service providers in the jurisdiction have the expertise and authority to conduct the entity's actual primary activity — not merely to stamp documents?
Failing these checks does not necessarily invalidate the structure; it changes its tax characterisation. The exposure is managed through early audit of the substance position before the tax authority raises it, not after.
If an existing structure has a substance gap, a second read can surface the structural reason and the route back. Write to OBOLUS at info@oboluslaw.com or message via t.me/oboluslaw to arrange a confidential review.
The IP and Treasury Layer: A Common Structural Weakness
Protocol IP — the smart-contract code, the brand, the governance frameworks — is often the most valuable asset in a digital-asset group and the most under-planned in the holding structure. Where the IP sits determines where royalties are recognised, how much is subject to withholding tax as they flow between group entities, and what transfer-pricing documentation is required.
Treasury assets — typically a mix of the group's own tokens, stablecoins, BTC and ETH — present a parallel question. A treasury held in the same entity as the operating licence is efficient when things go well. When they go badly — an enforcement action, a creditor claim, a hack — the treasury is exposed alongside the operating entity's regulated status. Separating the treasury into a distinct holding vehicle, appropriately capitalised and governed, is standard practice in mature digital-asset groups and increasingly expected by institutional banking counterparties.
The Travel Rule (the obligation to pass originator and beneficiary data with a virtual-asset transfer, derived from FATF Recommendation 15) creates an additional structural consideration for treasury management: intercompany transfers between group entities that are each VASPs are subject to the Travel Rule in most major jurisdictions. This means that the treasury entity and the operating entity cannot be treated as a single unit for compliance purposes simply because they share an ultimate beneficial owner. Inter-entity transfers must be compliant transfers.
For token-issuing groups, the IP layer also intersects with the MiCA whitepaper obligation. Under MiCA, the issuer of certain token categories must publish a whitepaper that discloses, among other things, the rights conferred by the token and the identity of the issuer. If the issuance function and the IP ownership function sit in different entities, the whitepaper must reflect this accurately or the structure creates a regulatory misrepresentation risk.
Founder Residency and Personal Tax: The Layer Most Often Mismanaged
Personal tax residency and corporate structure are decided together or not at all. This is the operating principle we return to most frequently in structuring engagements, and it is the area where the gap between what founders assume and what the law provides is widest.
A founder who moves to the UAE and establishes genuine residency there — spending the required number of days, maintaining a permanent home, severing social and economic ties with the prior jurisdiction — will generally cease to be taxable in the prior jurisdiction on future income and gains. The word "generally" is doing significant work in that sentence. Most high-tax jurisdictions impose an exit tax on unrealised gains at the point of departure. Some impose a period of continuing tax liability on gains from assets held before departure, particularly equity in private companies. The US taxes its citizens on worldwide income regardless of where they live, making a UAE move tax-neutral on US income without separate action.
The interaction between the founder's personal residency and the entity's tax residency is the primary structural lever at the personal level. A founder who is the sole director of a holding company, who exercises all significant decisions about that company, and who moves to a low-tax jurisdiction — but continues to exercise decisions from a home office in the prior jurisdiction — has not changed the holding company's effective management location. The personal relocation is necessary but not sufficient.
In our structuring practice, we align the founder residency analysis with the corporate substance analysis from the outset. The questions are: where will the founders actually spend their time, where will board decisions be made and recorded, and what activities will each entity in the structure genuinely conduct in its jurisdiction of incorporation? The answers determine the structure, not the reverse.
The timing of a relocation relative to a gain event is critical. Anti-avoidance provisions in major jurisdictions — including the UK, Germany, France, Australia and Canada — specifically target individuals who depart shortly before a gain crystallises. The applicable look-back periods vary. For a founder with a significant unrealised gain in a growing digital-asset business, the structuring analysis must model the personal tax liability at the proposed point of departure and compare it against the projected future liability of remaining resident. This is not a decision that should be made on the basis of a general conversation about UAE tax rates.
Common Structural Mistakes in Digital-Asset Holding Structures
Based on the structures we regularly encounter in our practice, the following errors recur with the highest frequency and the highest cost.
The most prevalent mistake is a single operating entity holding regulated licences, protocol IP, the treasury, and the founders' equity simultaneously. This creates concentration risk at every level: a regulatory enforcement action affects all four assets, a creditor claim reaches all four assets, and an exit transaction must untangle all four assets from a single entity. Layering takes time and transfers between related parties require transfer-pricing documentation, but the cost of not doing it consistently exceeds the cost of doing it.
The second most common mistake is failing to document the economic basis for intercompany arrangements. A holding company that charges the operating subsidiary a management fee, an IP royalty or a guarantee fee must do so at arm's length and must document the arm's-length analysis. Tax authorities in every major jurisdiction now specifically scrutinise intercompany arrangements between group entities in different tax jurisdictions. A management fee that lacks a contemporaneous transfer-pricing study is a management fee that a tax authority will reclassify as a dividend — with withholding tax applied retrospectively.
The third mistake is treating regulatory capital as structurally separate from tax planning. A VARA licence or a MAS licence requires the licensed entity to hold minimum regulatory capital. If that capital is contributed from above as equity, it is trapped in the licensed entity until the licence is surrendered or amended. If it is contributed as debt, the interest payments create a deductible expense in the licensed entity and a taxable receipt in the lending entity — which may or may not be efficient depending on where the lending entity sits. Modelling the regulatory capital requirement alongside the group's overall funding structure is a step that many early-stage groups skip.
The fourth mistake is ignoring the banking layer entirely. A structure that is tax-optimal on paper will fail operationally if the holding entity cannot open a bank account. Banks operating in the EU, the UK, Singapore and the US apply their own AML and reputational due diligence to digital-asset holding structures. An entity in a jurisdiction perceived as high-risk for FATF purposes — regardless of the quality of its local regulatory licence — will face correspondent banking problems that an entity in a jurisdiction with an equivalent standard but a stronger FATF profile would not. The AIFC in Kazakhstan, for example, offers a common-law framework and AFSA supervision, but operators regularly encounter a harder banking path than their counterparts in Singapore or the UAE.
Related at OBOLUS
Related at OBOLUS
- Tax and cross-border structuring for digital-asset businesses – practice overview covering the full spectrum of group structuring, treaty planning and exit design.
- Pre-exit tax restructuring under heightened scrutiny – how to restructure a digital-asset group ahead of a sale or token event when regulators and tax authorities are both watching.
- UAE (VARA/Dubai) vs UK: where to license a crypto business – a detailed comparison of two leading licensing jurisdictions and the tax implications of each choice.
FAQ
Where should a token-issuing entity be domiciled?
There is no universal answer. The domicile must satisfy three criteria simultaneously: it must support the regulatory authorisation required for the intended token category and distribution geography; it must provide a favorable tax treatment for the issuance proceeds and ongoing protocol income; and it must allow the entity to maintain genuine economic substance. Under MiCA, issuers of asset-referenced tokens and e-money tokens must obtain authorisation in an EU member state. Outside the EU, Singapore, Switzerland and the UAE each provide viable frameworks, subject to the specific token's regulatory characterisation in those jurisdictions.
How are staking rewards taxed?
The tax treatment of staking rewards is unresolved or explicitly unsettled in most jurisdictions. The core question is whether rewards constitute income at the point of receipt — taxable at ordinary income rates — or whether they represent the creation of a new asset, taxable only on disposal. Jurisdictions vary considerably: some treat all staking rewards as income on accrual, others distinguish between proof-of-stake validation rewards and liquidity-provision rewards. The entity's jurisdiction, its accounting treatment and the characterisation of the staking activity itself all influence the analysis. We advise that this question be resolved explicitly with qualified tax counsel before the first rewards accrue, not after.
Does remote working create tax residency risk?
Yes — consistently and materially. A director or employee working remotely from a jurisdiction in which the company holds no licence and has no registered presence can create a taxable nexus for the company in that jurisdiction under most domestic tax codes. The threshold is typically the existence of a dependent agent or a fixed place of business, both of which remote work can satisfy. For digital-asset groups with distributed teams, a remote-working policy that maps each team member's location against the group's treaty and nexus position is not optional — it is a core compliance requirement.
OBOLUS is an independent digital-asset law boutique acting only for businesses. We advise exchanges, custodians, token issuers and funds on licensing across 70+ jurisdictions, on disputes and on-chain asset recovery across 25+ forums, and on the tax, banking and compliance that sit around them. We align founder residency with the holding structure and exit plan — because the two decisions cannot be made in isolation. Our team advises crypto exchanges, custodians, token issuers and funds across more than seventy licensing jurisdictions. Digital assets are the whole of our practice. To discuss your situation, contact info@oboluslaw.com.
By Lydia Brennan, Tax & Structuring Analyst — specialising in cross-border holding structures, treaty planning and exit tax analysis for digital-asset operators and token-issuance groups.
This publication is general information about the law and does not constitute legal advice. It is not a substitute for advice tailored to your circumstances. OBOLUS accepts no liability for action taken or not taken on the basis of this material. For advice on your situation, contact info@oboluslaw.com.