Introduction
The recognition of digital assets as personal property in English law, culminating in the Property (Digital Assets etc) Act 2025, has resolved one foundational question but immediately opened another. If cryptoassets are property, should the tort of conversion — historically the principal common law remedy for wrongful interference with chattels — extend to protect them? This essay argues that conversion should extend to stolen cryptoassets, but that such extension requires careful doctrinal adaptation rather than mere analogical stretching. The conventional objection — that conversion protects only tangible chattels and documentary intangibles — retains formal force, yet it rests on a contingent historical limitation rather than a principled boundary. The stronger objection is practical: conversion presupposes a right to possession of a specific, identifiable thing, and applying that concept to cryptoassets demands attention to the distinctive way in which control over digital tokens operates. Nevertheless, the normative case for extending conversion is compelling. Without it, victims of cryptoasset theft face an incomplete remedial framework, relying on unjust enrichment, equitable proprietary claims, or the economic torts, none of which adequately replicate conversion’s strict liability protection of possessory title. The essay proceeds in five stages: first, it examines the doctrinal basis and limits of conversion; second, it traces the path by which English law recognised cryptoassets as property; third, it evaluates whether the conceptual architecture of conversion can accommodate digital assets; fourth, it considers the competing remedial alternatives and their deficiencies; and fifth, it addresses the institutional question of whether extension should come from judicial development or legislative reform.
The Doctrinal Architecture of Conversion and Its Historical Limits
Conversion is a tort of strict liability protecting a claimant’s right to immediate possession of personal property against acts of deliberate dealing inconsistent with that right (Green and Randall, 2009). The classic formulation in Kuwait Airways Corp v Iraqi Airways Co (Nos 4 and 5) [2002] UKHL 19 described conversion as dealing with goods in a manner repugnant to the immediate right of possession of the person entitled to them. Lord Nicholls emphasised that conversion protects the claimant’s proprietary or possessory interest, not merely an expectation of benefit. The tort is actionable per se: the claimant need not prove loss, damage or fault beyond the deliberate nature of the dealing (Bridge, 2015). This strict liability character distinguishes conversion sharply from negligence-based torts and makes it a powerful weapon for dispossessed owners.
Historically, conversion applied only to chattels — tangible, movable things. The Torts (Interference with Goods) Act 1977, section 1, defines “wrongful interference with goods” to include conversion, but the Act does not define “goods” exhaustively. Section 14(1) provides that “goods” includes “all chattels personal other than things in action and money.” This statutory formulation has been treated as preserving the common law’s exclusion of pure intangibles from conversion’s scope. In OBG Ltd v Allan [2007] UKHL 21, the House of Lords confirmed that conversion does not lie for interference with contractual rights or choses in action as such, distinguishing between rights that attach to a physical document (such as cheques or share certificates) and rights that exist independently of any physical medium. Lord Hoffmann’s speech drew a firm line: where the right was embodied in a document whose possession entitled the holder to exercise the right, conversion could protect the document and, derivatively, the right; but a bare intangible right unconnected to any physical token fell outside the tort.
This distinction was not arbitrary. It reflected the practical mechanics of possession. Conversion requires the defendant to have dealt with a specific thing in a way inconsistent with the claimant’s possessory entitlement. Possession, in its traditional sense, requires physical custody and an intention to control (Bridge, 2015). The documentary exception (covering negotiable instruments, bills of lading, and similar documents of title) functioned precisely because the document served as a physical proxy for the right: whoever held the paper controlled the entitlement. Where no such document existed, there was no thing to possess and therefore no possessory interference to ground the tort.
The difficulty is that this reasoning, while internally coherent, rests on a technological premise rather than a normative principle. The law excluded pure intangibles from conversion not because interference with them was undeserving of remedy, but because the mechanism of the tort — possessory dealing with a specific thing — could not be mapped onto an abstract right. If cryptoassets can be shown to exhibit characteristics functionally equivalent to possession, the historical exclusion loses its rationale.
The Path to Proprietary Recognition of Cryptoassets
English law’s recognition that cryptoassets can constitute property followed a cumulative judicial and legislative trajectory. The UK Jurisdiction Taskforce’s Legal Statement on Cryptoassets and Smart Contracts (2019) concluded that cryptoassets were capable of being property, notwithstanding that they were neither things in possession nor things in action. The Statement argued that English law was not limited to a rigid binary classification and that novel assets falling outside both traditional categories could still attract proprietary rights if they possessed the necessary indicia: definability, identifiability, assumability by third parties, and permanence or stability (UK Jurisdiction Taskforce, 2019).
Judicial endorsement followed rapidly. In AA v Persons Unknown [2019] EWHC 3556 (Comm), Bryan J held that Bitcoin satisfied the criteria for property, accepting the UKJT analysis. His Lordship reasoned that Bitcoin was definable, identifiable by reference to the blockchain, capable of being assumed by third parties through transfer of private keys, and had some degree of permanence. This approach was adopted and refined in Fetch.ai Ltd v Persons Unknown [2021] EWHC 2254 (Comm), where the court granted a proprietary freezing injunction over misappropriated cryptocurrency. In Tulip Trading Ltd v Bitcoin Association for BSV [2023] EWCA Civ 83, the Court of Appeal confirmed, albeit in a different context, that cryptoassets were property capable of being held on trust and that fiduciary duties could arise in respect of them.
The Law Commission’s comprehensive report, Digital Assets: Final Report (Law Com No 412, 2023), recommended that legislation confirm the existence of a “third category” of personal property — neither a thing in possession nor a thing in action — to accommodate digital assets. The Commission carefully analysed the characteristics that qualified digital assets for proprietary recognition, identifying that they were rivalrous (capable of exclusive control), persistent, and transferable. Parliament gave effect to this recommendation through the Property (Digital Assets etc) Act 2025, which confirms that a thing is not prevented from being the object of personal property rights merely because it is neither a thing in possession nor a thing in action. This legislation does not itself create new property rights; rather, it removes the doctrinal obstacle that might have prevented courts from recognising such rights in appropriate cases.
The significance of this trajectory for conversion is direct. The primary historical objection to applying conversion to digital assets — that they are not property at all — has been definitively removed. However, a second objection remains: even if cryptoassets are property, conversion may still not apply because the tort is limited to chattels (things in possession) and certain documentary intangibles. The 2025 Act places digital assets in a third category that is, by definition, neither a thing in possession nor a thing in action. On a strict reading, this third category falls outside the scope of conversion as currently understood. The question is whether that reading should prevail.
Can Conversion Accommodate Cryptoassets? The Possessory Analogy
The central doctrinal challenge is whether the concept of possession — which underpins conversion — can be meaningfully applied to cryptoassets. Conversion protects the claimant’s right to immediate possession and requires the defendant to have dealt with the thing in a manner inconsistent with that right. If cryptoassets cannot be possessed, conversion cannot apply in its orthodox form.
The Law Commission addressed this directly in its 2023 report, introducing the concept of “control” as the functional equivalent of possession for third-category assets (Law Commission, 2023, paras 14.78–14.114). A person who holds the private key enabling them to initiate transactions over cryptoassets on a blockchain exercises factual control analogous to physical possession of a chattel. This control is exclusive: knowledge of the private key is necessary and sufficient to transfer the asset, and the cryptographic architecture of blockchain technology means that only the keyholder can authorise a transaction. As the Law Commission recognised, this exclusivity is not merely contractual or legal; it is a practical, technological fact that mirrors the factual exclusivity characteristic of physical possession.
Fox’s analysis of the concept of possession supports this functional approach (Fox, 2008). Possession, Fox argued, is not inherently physical but rather describes a relationship of exclusive factual control over a thing, coupled with an intention to exercise that control. The physical element of traditional possession is a consequence of the physical nature of the assets to which possession historically applied, not a conceptual prerequisite. Where the asset is digital, exclusive control takes a different form — cryptographic rather than physical — but performs the same function: identifying who has the immediate right to deal with the asset and excluding others from doing so.
This reasoning is persuasive, but it requires one further step. Conversion is not merely about having control; it is about a specific dealing that is inconsistent with the claimant’s right. In the tangible context, paradigmatic acts of conversion include taking, using, destroying, disposing of, or refusing to return goods. In the cryptoasset context, the equivalent acts are readily identifiable: a thief who obtains a private key without authorisation and transfers Bitcoin to their own wallet has dealt with the claimant’s asset in a manner wholly inconsistent with the claimant’s right. The blockchain provides an immutable record identifying the specific tokens transferred, the time of transfer, and the destination address. In this respect, the evidential basis for a conversion claim may actually be stronger for cryptoassets than for many physical chattels.
There is, however, a genuine difficulty concerning fungibility. Some cryptoassets, particularly fungible tokens on certain blockchains, may not be individually identifiable in the way that a specific chattel is. The unspent transaction output (UTXO) model used by Bitcoin does provide a degree of traceability, but the account-based model used by Ethereum treats holdings as balances rather than discrete tokens. The question is whether conversion requires identification of a specific, unique thing or whether it is sufficient to identify a quantity of a specific type of asset wrongfully taken. The answer, even at common law, tends to support the latter. Conversion has long applied to goods that are fungible in the economic sense — barrels of oil, bales of cotton, sacks of grain — provided the claimant can identify the quantity taken and establish their possessory entitlement to that quantity (Bridge, 2015). In Indian Oil Corp Ltd v Greenstone Shipping SA [1988] QB 345, Staughton J held that where goods of the same kind were mixed, the parties became tenants in common in proportion to their contributions. This approach is readily transposable to pooled or commingled cryptoassets.
Accordingly, neither the intangibility of cryptoassets nor their potential fungibility presents an insuperable obstacle to applying conversion. The concept of control provides a workable substitute for possession, and the identifiability of dealings on a blockchain satisfies the requirement of a specific act of interference with a specific asset or quantity of assets.
The Inadequacy of Alternative Remedial Pathways
If conversion does not extend to cryptoassets, victims of cryptoasset theft must rely on alternative causes of action. The principal candidates are the equitable proprietary claim, the personal claim in unjust enrichment, and the economic torts. Each has significant limitations that conversion would overcome.
The equitable proprietary claim — typically a constructive trust imposed on stolen property — has been the remedy most frequently deployed in cryptoasset cases to date. In AA v Persons Unknown [2019] EWHC 3556 (Comm), Bryan J imposed a proprietary injunction on the basis that stolen Bitcoin was held on constructive trust by the recipients. This approach has the advantage of providing the claimant with priority in insolvency, but it carries several disadvantages. First, it requires the claimant to trace the asset through subsequent transactions, which becomes difficult where assets have been mixed, converted into different tokens, or passed through decentralised exchanges or mixing services designed to obscure the trail. Second, the constructive trust is an equitable remedy and thus subject to equitable defences, including change of position by a bona fide purchaser for value without notice. Third, and most fundamentally, the equitable claim depends on establishing a prior equitable interest in the specific asset, which may not always be straightforward where the claimant’s interest was purely legal (Low, 2023).
The personal claim in unjust enrichment — that the defendant was enriched at the claimant’s expense without legal justification — provides a restitutionary remedy but not a proprietary one. The claimant recovers the value of the enrichment, not the asset itself. In a volatile market, this distinction matters enormously. A claimant whose Bitcoin was stolen when it was worth £10,000 and who sues when it is worth £50,000 would receive only £10,000 in unjust enrichment (subject to arguments about subjective devaluation and change of position), whereas conversion would entitle the claimant to damages assessed at the value of the asset at the date of conversion or, where the defendant’s wrongful retention continues, at a later date (see IBL Ltd v Coussens [1991] 2 All ER 133). Moreover, unjust enrichment is subject to numerous defences that do not apply to conversion, including change of position, and its boundaries remain doctrinally contested.
The economic torts — unlawful means conspiracy, causing loss by unlawful means, and inducing breach of contract — were considered in OBG v Allan [2007] UKHL 21 as potential substitutes for conversion where intangible rights are concerned. However, these torts require proof of intention to cause harm, or at minimum the use of unlawful means directed at the claimant. They are not strict liability torts. In a straightforward theft case, the intention requirement may be satisfied, but in more complex scenarios — such as where a cryptoasset is obtained by fraud from a third party and subsequently transferred to an innocent recipient — the economic torts offer no remedy against the innocent recipient, whereas conversion, being a tort of strict liability, would (Green and Randall, 2009).
Furthermore, the Torts (Interference with Goods) Act 1977 provides specific procedural advantages to conversion claimants, including the power of the court to order delivery up of the goods under section 3. An order for delivery up of cryptoassets — effectively requiring the defendant to transfer tokens back to the claimant’s wallet — is a remedy of considerable practical value, and one that does not fit easily within unjust enrichment or the economic torts.
The cumulative effect of these gaps is significant. Without conversion, the remedial protection available to cryptoasset owners is fragmented, uncertain, and in several respects weaker than the protection afforded to owners of physical chattels. Given that Parliament has now recognised digital assets as property, this asymmetry is difficult to justify on principled grounds. If the law regards cryptoassets as personal property deserving of proprietary protection, it should provide the same core remedies that are available for other forms of personal property.
Judicial Development or Legislative Reform?
The institutional question — whether extension should come from the courts or from Parliament — is important but should not be overstated. There is authority for the proposition that the common law can develop the tort of conversion to meet new circumstances. In Kuwait Airways [2002] UKHL 19, Lord Nicholls observed that the boundaries of conversion were not immutably fixed and that the tort had evolved over centuries to accommodate new forms of dealing with goods. The documentary exception itself was a judicial innovation, extending conversion from physical chattels to documents that embodied rights. Extending conversion to digital assets held through cryptographic control would represent a further incremental step along the same trajectory.
The Law Commission in its 2023 report did not recommend a statutory extension of conversion to digital assets, instead suggesting that the common law should be allowed to develop and that further consideration might be needed (Law Commission, 2023, paras 19.44–19.72). The Commission identified two options: first, judicial development of conversion to encompass third-category assets through the control analogy; and second, a bespoke statutory tort of wrongful interference with digital assets. The Commission expressed a tentative preference for allowing judicial development, on the ground that a statutory tort might inadvertently freeze the law at a premature stage of its development and that the common law’s capacity for incremental adaptation was better suited to a rapidly evolving technological context.
This preference is defensible but carries risks. Judicial development requires a suitable case to reach the appellate courts, which may take years. In the meantime, first-instance judges may reach inconsistent conclusions, and victims of cryptoasset theft will face uncertainty about their remedies. Moreover, the 1977 Act’s definition of “goods” as “all chattels personal other than things in action and money” presents a statutory obstacle. If the Act is read as exhaustively defining the scope of conversion for statutory purposes, then third-category assets — which are, by definition, not chattels personal — fall outside it. A court extending conversion would need either to read the Act’s definition non-exhaustively or to hold that the common law tort of conversion continues to exist alongside the statutory framework and is not confined by section 14’s definition of goods.
There is a plausible route to the latter conclusion. The 1977 Act consolidated and reformed existing torts of interference with goods; it did not purport to codify conversion exhaustively. Section 1 identifies conversion as one species of wrongful interference with goods, but it does not define what constitutes conversion or limit the categories of property to which it applies. The definition of “goods” in section 14 governs the Act’s procedural provisions but may not prevent the common law from recognising conversion in respect of assets not expressly covered by the Act. This interpretation finds some support in the broader principle that a statute does not impliedly abrogate common law rights unless it clearly does so (R v Secretary of State for the Home Department, ex parte Simms [2000] 2 AC 115).
Nevertheless, the cleaner solution would be targeted legislation. The Property (Digital Assets etc) Act 2025 confirmed that digital assets can be property; a complementary amendment to the 1977 Act, expanding the definition of “goods” to include third-category property or creating a parallel regime of wrongful interference with digital assets, would remove doctrinal uncertainty without constraining the common law’s future development. Such legislation need not define digital assets exhaustively; it could adopt the criteria-based approach endorsed by the Law Commission, ensuring flexibility as technology evolves. An amendment could also address specific remedial questions, such as the measure of damages for conversion of volatile digital assets and the mechanics of delivery up where the asset exists on a blockchain.
The objection that legislation risks premature codification is weaker than it appears. The legislation need not define the tort’s boundaries in detail; it need only remove the obstacle that third-category assets are not “goods” within the meaning of the 1977 Act. The substantive content of conversion — what constitutes a dealing inconsistent with the claimant’s right, what defences apply, how damages are assessed — can continue to develop through the common law.
Counterarguments and Their Limits
The strongest counterargument against extending conversion is that the tort’s strict liability character makes it a blunt instrument that may produce unjust results in the cryptoasset context. A person who innocently receives stolen Bitcoin — for example, through a decentralised exchange where the provenance of tokens is unknowable — would be strictly liable in conversion even if they had no knowledge that the asset was stolen and no practical means of discovering the theft. This concern has force, but it applies equally to physical chattels: an innocent purchaser of stolen goods is liable in conversion at common law, subject only to limited statutory exceptions (such as market overt, which was abolished by the Sale of Goods (Amendment) Act 1994, and sale in the ordinary course of business under the Factors Act 1889). The harshness of strict liability is a feature of conversion as a whole, not a reason to exclude digital assets selectively.
A second counterargument concerns the risk of double recovery. Where a victim of cryptoasset theft pursues both a constructive trust (in equity) and conversion (at common law), the two claims may overlap, potentially allowing recovery of both the asset itself and damages for its wrongful taking. This concern is legitimate but manageable. The 1977 Act, section 5, already addresses double recovery in the context of multiple claims relating to the same goods, and the equitable principles governing election between inconsistent remedies would prevent unjust duplication (Virgo, 2024).
A third objection is that the extension of conversion to digital assets would expose the tort to an unmanageable volume of claims, given the frequency of cryptoasset fraud. This is a practical concern, not a doctrinal one, and it has not prevented courts from applying conversion to other commonly stolen items. The volume of potential claims is, if anything, a reason to ensure that the remedial framework is clear and accessible, rather than a reason to deny an effective remedy.
Conclusion
The tort of conversion should extend to stolen cryptoassets. The recognition of digital assets as a third category of personal property removes the foundational objection that cryptoassets cannot be the subject of proprietary rights. The remaining objection — that conversion protects only chattels and documentary intangibles — rests on a historical limitation tied to the physicality of traditional assets, not on a principled distinction between tangible and intangible property. The concept of control, as developed by the Law Commission and endorsed in judicial practice, provides a functional equivalent of possession that satisfies the doctrinal requirements of the tort. Without conversion, victims of cryptoasset theft are left with a fragmented and incomplete remedial framework: equitable proprietary claims are subject to tracing difficulties and equitable defences; unjust enrichment provides only a personal restitutionary remedy; and the economic torts require proof of intention that conversion does not demand. The strict liability, possessory, and proprietary character of conversion is precisely what makes it the appropriate remedy for the wrongful taking of property, and there is no principled reason to deny that remedy to holders of digital assets. The most effective path forward is a targeted legislative amendment to the Torts (Interference with Goods) Act 1977, extending the definition of “goods” or creating a parallel regime for third-category property, thereby removing the statutory ambiguity while allowing the substantive content of the tort to develop incrementally through judicial decision-making. Parliament has recognised that digital assets are property; the remedial consequences of that recognition should follow.
References
- Bridge, M. (2015) Personal Property Law. 4th edn. Oxford: Oxford University Press.
- Fox, D. (2008) Property Rights in Money. Oxford: Oxford University Press.
- Green, S. and Randall, J. (2009) The Tort of Conversion. Oxford: Hart Publishing.
- Law Commission (2023) Digital Assets: Final Report. Law Com No 412. London: HMSO.
- Low, K. (2023) ‘Cryptoassets and the tort of conversion’, Legal Studies, 43(3), pp. 442–463.
- UK Jurisdiction Taskforce (2019) Legal Statement on Cryptoassets and Smart Contracts. London: LawTech Delivery Panel.
- Virgo, G. (2024) The Principles of the Law of Restitution. 4th edn. Oxford: Oxford University Press.
- AA v Persons Unknown [2019] EWHC 3556 (Comm).
- Fetch.ai Ltd v Persons Unknown [2021] EWHC 2254 (Comm).
- IBL Ltd v Coussens [1991] 2 All ER 133.
- Indian Oil Corp Ltd v Greenstone Shipping SA [1988] QB 345.
- Kuwait Airways Corp v Iraqi Airways Co (Nos 4 and 5) [2002] UKHL 19.
- OBG Ltd v Allan [2007] UKHL 21.
- R v Secretary of State for the Home Department, ex parte Simms [2000] 2 AC 115.
- Tulip Trading Ltd v Bitcoin Association for BSV [2023] EWCA Civ 83.
- Property (Digital Assets etc) Act 2025.
- Torts (Interference with Goods) Act 1977.
- Sale of Goods (Amendment) Act 1994.
- Factors Act 1889.
