In my previous article, I attempted to make a case for a “taxonomy of tokens” by drawing attention to the practical, regulatory implications of an uncertain definitional framework. In this article, I seek to focus more on the defining features of different types of tokens, and how the usage of precise terminology enables us to better understand how each type of token fits within the broader framework of personal property law. As a corollary of this exposition, I will conclude by arguing that an alternative proprietary basis for virtual currencies, by the recognition of a third category of personal property outside choses in action and in possession, should be adopted.
A working definitional framework may be gleaned from Ang et al (2019) in a recent article for the Business Law International periodical1. The table below sets out these different types of tokens, and two key features which might have implications on the proprietary analyses we apply to them: whether the token is 1) blockchain based, and whether it is 2) asset-backed.
It is important to separate digitalised fiat from other types of digital tokens because the specific item of property concerned differs from that at issue for digital tokens which are blockchain-based.
This first category is “asset-backed” in the sense that it represents in digital form state-issued, fiat currency. Accordingly, the legal analysis pertaining to the proprietary rights attaching to these digital tokens does not differ much from that relating to physical fiat2. Briefly, however, I would add that the analysis does depend on the type of digital wallet that we access the fiat through.
Wallets may be ‘pass-through’ or ‘staged’, and this has implications primarily on the party against whom a claim of debt is brought. Debit cards, for example, function simply as mechanisms through which customers – or whoever bears the payment authorisation number (“PAN”) associated with the card – ‘pay’ by exercising rights in direct relation to their ‘fluctuating chose in action’ – the debt the bank owes them represented by their bank ‘balance’3. The same is true of ‘pass-through’ wallets such as ApplePay or SamsungPay4, which offer customers the opportunity to exercise this right as against their bank through the platform provided by the intermediary: Apple, Samsung, or any other service provider as the case may be.
Staged wallets, however, do not perform that bare intermediary function. In the case of a staged wallet such as SquareCash, a customer ‘tops up’ the wallet by injecting funds directly into the wallet-provider5. Accordingly, it is the balance in the wallet which represents the debt owed to them by the wallet-provider; and it is this debt owed by the wallet-provider which is the item of property over which rights are exercised at the point of payment.
What, then, is the item of property that the holder of the asset has in the context of blockchain-based digital tokens?
Security tokens and tokenised assets
Broadly speaking, a security token is “asset-backed” in that it carries with it other rights of ownership or control in relation to the system provider or otherwise the issuing entity. A tokenised asset usually refers to units of items of property, in practice mostly commodities6 or real estate7, which have been “tokenised” for distribution using blockchain technology.
Recognising an item of property over these tokens is arguably less problematic than in relation to non-asset-backed tokens. A security token purporting to give its holder voting rights in respect of certain matters regarding the issuing firm can arguably be treated similarly to a share as an item of property8. Accordingly, the token is merely an embodiment of the rights vis-à-vis the issuer, no different from shares recorded solely electronically by the Certificateless Registry for Electronic Share Transfer (“CREST”). Its proprietary significance per se is derived completely from the rights it represents.
The main difficulty with this ‘share-analogy’ approach, however, is that not all asset-backed tokens are the same. Not all of them purport to confer rights as against an issuing firm. A tokenised asset may be ‘backed’ by gold bullion, or real estate, or sugar; but it may not provide a right to these commodities in specie, legal ownership of which remains with the issuing firm. Is it correct, then, that whether or not a holder of an asset-backed token has a proprietary interest, depends so much on the kinds of rights the token embodies?
Fortunately, there is an alternative basis for recognising a proprietary interest in asset-backed tokens. This proprietary basis is applicable also to non-asset-backed tokens so long as they are blockchain-based. This next section sets out the definition of such non-asset-backed, blockchain-based tokens, and applies the reasoning put forward by Professor David Fox, arguing in favour for recognising this separate basis.
The paradigm example of a non-asset-backed, blockchain-based digital token is a virtual currency, or “crypto-currency” in a narrow sense. The term refers to a digital token on a distributed ledger used primarily as a medium of exchange for goods and services. A virtual currency is distinct from the other four types of digital tokens set out above in the main respect that it is not asset-backed. Accordingly, a virtual currency proper has no intrinsic value – its value is determined by market forces, extrinsic to itself.
A key trait of any digital token native to a blockchain is that the “token” exists per se as a ‘data string recording transaction’: an item of intangible “property”, namely, information9. Recall also that a blockchain is, in substantia, a public “ledger”, which means that each cryptographically secured transaction is identified by reference to a prior cryptographically secured transaction, and that transaction to the one prior. In other words, when a token, x,is moved from one individual (“A”)’s cryptographically secured wallet to another (“B”)’s, the information on that ‘data string’ changes from xto x + y.10
The result of this is that each transaction gives rise to a new item of property; token xis not token x + y. It also means that, even when stored in a cryptographically secured wallet, it is not the fund or the value of his Bitcoin balance that person A has a proprietary interest in, in the same way that a customer at a bank has an interest in his bank account as a chose in action.11 On the contrary, A’s proprietary right must be understood as corresponding to each individual digital token as an item of property in itself.
Does this mean that all virtual currency tokens are infinitely specific and not fungible? In this respect, are virtual currencies different from fiat (state-denominated, “real world”) currencies?12
The answer to these questions must be in the negative. Professor Fox13 tells us that:
“fungibility or specificity of a thing is not an absolute property of the thing itself. The characterisation of the thing as fungible or specific depends instead on the perspective of the parties to a transaction and the kinds of legal right at issue between them.”14
On what basis does this right exist? Professor Fox directs us to the dicta in Armstrong v Winnington 15: a blockchain based digital token is neither a chose in action or in possession, but part of a third category of personal property, since it is 1) definite; 2) identifiable by third-parties; 3) derives its value from trade; 4) permanent and stable; and 5) subsisting over time16.
Recognising this category constitutes a major reform to the law, but I think it is right to do so. In the first place, bereft of this basis, understanding virtual currency digital tokens as part of the larger private law framework of personal property becomes impossible. It seems wrong to me that a merchant transacting solely in virtual currency is in a different legal position than one transacting solely in digitalised fiat.
In the second place, adopting this reasoning in general provides coherence to the category of asset-backed tokens as a whole. It removes the legal uncertainty attending holders of security tokens or tokenised assets, who may be uncertain as to their private law rights in relation to the digital tokens they ‘hold’.
A rose by any other name is not a rose. I admit the danger in adopting rigid definitional categories, but what I have tried to show is that referring to digital tokens as a whole indiscriminately as ‘cryptocurrencies’ obscures more than it reveals. The differences between the four main categories of digital tokens, between tokens which are ‘blockchain-based’ and those which are not, as well as the distinctions between tokens which are ‘asset-backed’ and those which are not, affect the legal analyses we employ in determining how these assets fit in our understanding of personal property law.