Digital Property law risks overreach after 2025 Act

Developments since the Property (Digital Assets etc) Act 2025 became law in December have added to concerns that the law on property in crypto-currencies and other virtual assets is developing in the wrong direction and is out of step with other jurisdictions.

As is now well-known, the Property (Digital Assets etc) Act 2025 states that a thing can be property even if it is not a thing in possession or a thing in action. None of these terms are themselves defined. 

The intention was to enshrine the Law Commission’s belief that digital assets such as crypto-currencies – bitcoin, ether and other like tokens – were property, but neither things in possession nor things in action, into English law. For a supposedly uncontroversial Law Commission bill, it proved oddly controversial, however. A classification in which things in possession are objects that can be possessed, and things in action are everything else, is perfectly coherent.

Some have even argued that it was wrong to grant digital assets proprietary status of any kind. The Act, however, passed into law and its consequences must be confronted. It would be a brave counsel who tried to argue that crypto-currencies were not property now, but what about other virtual assets? Are they property? If they are, are they things in action or something else? After all, the remedies that a claimant gets when someone interferes with a virtual asset may differ significantly depending on the answer. But English law is in danger of over-extension.

Over-Extending Property in Virtual Assets

R v Lakeman [2026] concerned the question whether “gold pieces” in a virtual role playing game, called Old School RuneScape which was marketed and run by a company called Jagex, could be stolen. Were they property for the purposes of the Theft Act 1968? The court decided yes. Further, the court drew a distinction between the private law definition of property and property under the Theft Act. It recognised that the two definitions of property need not correspond; something can be property under the Theft Act but fail to meet the definition of property under private law criteria. However, the decision still raises wider concerns about where English personal property law is going. 

Analogies with Crypto-Currencies

First, the court drew strong analogies between the “gold pieces” under consideration and crypto-currencies. Popplewell said (at [30]) “Although the gold pieces exist in a virtual world, they have a real world existence which manifests itself not only in their appearance and use on the screen but in real world trading. They are real functional things distinct from the code which creates them.”

Popplewell LJ saw an analogy with bitcoin (at [32-34]), saying: “The position is analogous to that which applies to cryptocurrencies, which are widely regarded as capable of constituting property… and for which the relevant criteria are applied to the functional concept of the cryptocurrency itself rather than to the software which creates it. Bitcoin, for example, consists of coded software which is entirely public and not the property of anyone: the code is public and is applied on a distributed ledger which itself is public. The only private element is the private key. There is no property interest in either the code or its manifestation on the blockchain.”

A bitcoin is also not just information; it has a larger functionality. It can be traded (indeed trading is all you can do with a bitcoin). Popplewell LJ argued (at [34]) that the way in which bitcoin is transferred on the chain has similar attributes to the way in which gold coins in the game are transferred. When transferring a bitcoin, the original bitcoin attributed to Alice is destroyed, and a new one attributable to Bob and represented by a different alphanumeric sequence is created. In the same way, when transferring an in-game gold coin, the code instantiating the coin as Alice’s is deleted, and another set of code attributing it to Bob is created. In both cases, therefore, the property is functionally distinct from the data and is transferred in the same way. Popplewell LJ proceeded to draw the obvious inference. If a bitcoin is property – and can be stolen – so are these gold pieces. 

The Ainsworth Criteria 

Secondly, Popplewell LJ drew on the criteria set out in National Provincial Bank v Ainsworth [1965]. These were referred to in Yeates v The King [2025], a decision of the Victoria Court of Appeal that bitcoins were property capable of being stolen under section 72 of the Crimes Act 1958. That section is in identical terms to section 1(1) of the Theft Act 1968 and the definition of property in section 71 is the same as under section 4 of the Theft Act. There are four criteria, although the third does not apply in Australia. A right or interest in property must:

• Be definable
• Be identifiable by third parties
• Be capable in its nature of assumption by third parties
• Have some degree of permanence and stability 

The gold coins met all four. It is possible to define what the gold piece is. One gold piece is then identifiable and distinguishable from another gold piece. They can be bought and sold and will remain in existence indefinitely unless cancelled by Jagex. The Ainsworth criteria are, however, viciously circular. The reason why you – the reader – are not an object of property is that you are not capable of assumption by third parties. You can’t be bought and sold. But, of course, the reason you can’t be bought or sold is that you are not property. English law does not recognise chattel slavery. The circularity should be obvious. Yet the criteria have been regularly applied as a guide to the private law proprietary nature of things – including bitcoin. The Court of Appeal, having said – repeatedly in fact - that what is property for criminal law purposes is not necessarily so for private law purposes, promptly reached for a (flawed) private law tool to assess proprietariness.

Rivalrousness

Thirdly, the Court of Appeal examined the concept of rivalrousness. Counsel – Mr Eissa - had suggested that meeting the Ainsworth criteria was not sufficient. For digital assets to be property, they must also be rivalrous. These gold coins were not, he argued. Popplewell LJ rejected this and made extensive reference to the Law Commission reports and to the 2025 Act. He referred to the Law Commission’s report, but we believe his interpretation does not fully capture the argument. 

His Lordship set out the three criteria provisionally suggested by the Commission’s consultation paper for an object to fall within their third category of personal property. 

• It is composed of data represented in an electronic medium, including in the form of computer code, electronic, digital or analogue signals
• It exists independently of persons and exists independently of the legal system
• It is rivalrous 

Having quoted these criteria (at [73]), Popplewell LJ went on to say (at [75]) that the gold pieces met the Law Commission’s criteria. We respectfully disagree; they do not meet the Law Commission’s criteria. The Law Commission, in fact, concluded that in-game assets were not independent of the legal system and do not fall within the scope of their proposed category of personal property. (LCCP Ch 7) In-game assets exist solely because of a proprietary ecosystem owned by a developer and the web of licences governing what players are able to do. 

Popplewell LJ, however, concentrated his discussion on rivalrousness and noted that in Tulip Trading v Bitcoin Association [2023] Birss LJ states that bitcoins are rivalrous. Rivalrousness refers to the fact that one party’s use of a thing necessarily impairs or excludes another’s. Possession is inherently rivalrous. If I possess my pen, you do not. It is imperfect control because you might grab the pen from me, but this does not affect its rivalrousness. The point is that control and rivalrousness are conceptually different. Popplewell LJ’s argument was that the username and password secured the rivalrousness of the gold pieces by ensuring only Alice could use Alice’s gold. But they do not. They may secure Alice’s control, but control is, as we have said, different from rivalrousness. Control is not binary; rivalrousness is binary. 

This is why Mr Eissa’s objection “…that Jagex have both the practical ability and contractual right to alter the holding by player A or player B of their gold pieces, either by action taken directly against the player or by changing the rules of the game. Jagex control the game code and so can rewrite the code to destroy the player’s holding.” (discussed at [76]) is so important. Popplewell LJ said of Jagex’ ability and the possibility that it would rewrite the code, that “it does not stop the gold pieces being rivalrous at a time before it occurs.” 

But this reasoning is, with respect, mistaken. 

To stress, the gold pieces are either rivalrous, or they are not. It is not possible for a thing to be rivalrous before a party’s exercise of control and not afterwards. That is simply not what rivalrousness means. 

A distinction must be drawn between crypto-currencies and in-game assets. There is no doubt about crypto-currencies being rivalrous, but the analogy should not apply to in-game coins, unless they use the same technical functionalities. Crypto-currencies utilise distributed ledger technology wherein the functionality of the network by design allows the crypto-assets to be rivalrous. The use of a crypto-currency necessarily inhibits the use of that same crypto-currency by another user. If I spend the bitcoin, you cannot. Indeed, even if things go wrong and there is a 51% attack where a bad actor takes control of enough of the network to cancel transactions already made, only one person at a time can use the bitcoin. Contrarily, an in-game asset is simply a data entry in the game’s database. Its perceived ‘rivalrousness’ is illusory, whereas a crypto-currency is inherently rivalrous. Alice has unique control of the bitcoin unless (being imperfect) it is wrested from her. By contrast, Alice’s control of her virtual gold coins is inherently subject to Jagex’s control. It is not rivalrous. 

There is a slippery slope here. The logic is that because my username and password give me access to the game and control of an account, I could advance as far as I wish, then sell the account information, and this should count as property. Email accounts should also, by the same logic, be property. They too are protected by usernames and passwords, but they are not property. Attempts have been made to make email accounts property, but they have – rightly - been unsuccessful (Fairstar Heavy Transport Ltd v Adkins [2013]). This suggests a fundamental – albeit not uncommon - misunderstanding of rivalrousness as identical to control.

English Law is Out of Step 

Will virtual property like these gold coins come to be seen as attracting private property rights? If so, will such virtual property be seen as part of the third category implicitly recognised by the 2025 Act? Despite Popplewell LJ’s insistence that this is a criminal law question and not a property law question, we cannot help but think that the case will be influential beyond its bounds. If these questions are answered in the affirmative, the law is developing out of step with other major jurisdictions. For example, the American approach does not support such an extension. Controllable Electronic Records are recognised as a new category of intangible property under Article 12 of the Uniform Commercial Code, but the type of asset it covers solely refers to assets that can be subject to exclusive factual control such as crypto-assets. In-game assets do not fall within that category. 

In conclusion, early indications after the Act suggest cause for concern. They indicate continued misunderstandings and misconceptions about rivalrousness and a willingness to expand the category of property in a way that stretches it past breaking point. 

Copyright & permissions