I. Introduction
Tokenisation enables the creation of platforms that reference assets, manage rights in those assets, and simplify their trading. Many businesses and regulators, especially in financial services, have launched projects based on this technology.Footnote 1 The value of applications leveraging this technology is substantial and its growth accelerating.Footnote 2 In response, the European Union has enacted the new Markets in Crypto-Assets Regulation.Footnote 3 This Regulation creates a harmonised European regulatory framework for crypto-assets.Footnote 4 However, as Garcia-Teruel and Simón-Moreno point out, this regulation does not cover the legal nature, effects and admissibility of using tokens to transfer property rights.Footnote 5 This creates uncertainty as these technological implementations, both in terms of ownership and transfer of the token, rely on private law concepts. The law of property and the law of contract provide the legal basis for financial contracts and for financial entities, as was demonstrated by Martino, Nabilou and Pacces.Footnote 6 Kiskis demonstrates that the introduction of tokenisation technology in financial services necessitates an analysis under private law.Footnote 7 This article contributes to that private law analysis and aims to deepen our understanding of the impact of token technology on existing theories on transfers of property. This is important as the private law foundations that form the legal underpinnings of financial services play a crucial role in safeguarding the stability of markets and the protection of investors. As such, the consequences of an unsuitable private law foundation might be severe. For instance, Kokorin examines the recent collapses of prominent crypto trading and lending firms to illustrate the impact of uncertainties in private law within the crypto context and shows how this contributes to existing vulnerabilities.Footnote 8 Similarly, one might ask whether transferring a token that represents a right could, in effect, constitute the transfer of the right itself – an important consideration in decentralised finance, where tokenised assets and rights are frequently exchanged on blockchain platforms.Footnote 9 From a technical perspective, programming software to achieve exactly this is possible.Footnote 10 However, if such transfers are not recognised by the private law frameworks of the relevant legal system, it remains to be seen whether such transfers are capable of generating legal effects.
This paper uses the token container model to examine whether current European private law frameworks on transfers of rights could effectively facilitate token-based transfers. This examination contributes to a better understanding of current applications of token technology, its impact on existing systems on transfers of rights, and its capacity to produce transfers that hold generate effect, particularly in the context of decentralised finance. The token container model, detailed in section III, forms the foundation of the analysis in this paper. It addresses one central question: how can tokens be used to transfer rights? This question will be explored using the analytical framework proposed by Van Vliet: he identifies two dividing lines that can be used to classify transfer systems.Footnote 11 This framework enables a qualitative analysis of consensual, causal tradition, and abstract tradition transfer systems, which, in turn, offers insight into whether the token container model provides a model on the basis of which broader European legal frameworks can give effect to token transfers, thereby determining whether tokens can actually be used to transfer rights.
In order to transfer a token in a manner recognised by the law, it must be recognised by the legal system in question as a transferable object within property law. Furthermore, tokens with certain specific contents might be qualified as a specific financial instrument. As far as the private law rules on transfer are concerned, this is primarily a matter of national laws.Footnote 12 However, if the token is qualified as a financial instrument, harmonised European rules regarding the transfer of such instruments might be applicable.Footnote 13 In such contexts, the previously mentioned Markets in Crypto Assets Regulation should be considered. It is for these reasons that the qualification of the token is relevant and deserves ample consideration. However, in order to ensure a certain degree of conciseness, this paper will take Principle 3(1) of the Unidroit Principles on Digital Assets and Private Law as its starting point and treat tokens as digital assets that can be subject to proprietary rights.Footnote 14 It follows from this that tokens fall within the scope of the law on property as transferable objects. Hence, while the relevant national and European rules stipulate how a specific token must be qualified, this paper takes the premise that tokens are objects for the scope of property law and, as such, can be transferred in a manner recognised by the law. Differentiating between the different types of tokens, analysing how these might be qualified, and examining whether the applicable (harmonised) rules on transfers are compatible with token-transfers is beyond the scope of this current research. Rather than qualifying a token as more than a transferable object within the law of property, this paper approaches tokens as technical instruments that could streamline and simplify the processes of transferring right. This enables an analysis of whether national rules on transfers of movables, immovables, and claims are compatible with the use of tokens as instruments for transferring rights in these objects. Both the qualification of the token itself under national law or under European law, including MiCAR, is therefore beyond the scope of this research.
The central questions considered in this paper are strongly intertwined with questions of private international law. In this regard, the question on applicable law is especially relevant.Footnote 15 It could, for example, be argued that, if the applicable law has adopted the token container model, the rules on transfer of that legal system apply. This paper is concerned with the preliminary matter of whether the context provided by the national rules on transfer are compatible with such token-based transfers. This question is relevant to the risk analysis that industry stakeholders experimenting with this technology make, as well as policymakers who might want to create a regulatory environment that stimulates innovations based on tokenisation technology. As the adoption of this technology increases, the relevance and the importance of that private international law question increases in tandem. This relates first and foremost to the questions whether, and in which circumstances, the applicability of national laws that have adopted the tokenisation model also results in the rules on transfer of that legal system being applicable and whether this offers an effective solution. However, these matters of private international law are beyond the scope of this paper.
Additionally, the scope of this paper is limited in three ways. First, the focus is limited to property law aspects of token transfers alone: the broader private law context is out of scope. Second, this paper is concerned with transfers of movables, transfers of immovables, and assignments of claims. Third, this paper is only concerned with voluntary transfers through legal agreements. Other transfers, such as those in the context of gifts, inheritances, and insolvencies, are beyond the scope of this paper.
Four steps will be taken to address the central research question. First, section II will explain what tokens are and how token technology functions. Second, section III will illustrate the token container model through recent legislative initiatives in the Principality of Liechtenstein and the Republic of Belarus. Third, section IV will provide an overview of European transfer systems as far as absolute rights are concerned. Following this, in subsections IV.3 and IV.4 respectively, an analysis will be performed on whether these rules are such that they can give token-based transfer legal effect. Fourth, section V will provide an overview of European transfer systems as far as the assignment of claims is concerned. Subsequently, subsection V.1 will analyse whether these rules are such that they can give token-based assignments of claims legal effect. By taking these four steps, this paper identifies two fundamental legal obstacles to token-based transfers of rights in European systems of private law. Section VI provides the conclusion.
II. Technology
Regulators typically differentiate between three types of tokens: payment tokens, utility tokens, and asset tokens.Footnote 16 This final category is central to this contribution. Tokenisation is commonly associated with blockchain technology; however, its underlying goals can also be achieved without relying on a blockchain. Given that tokenisation has gained traction in this specific context, this article will explain tokenisation as a continued development of this technology.
1. Blockchains
A blockchain platform constitutes the foundational layer upon which a tokenisation application is built. The technology was developed by individuals who subscribed to extropianism, an early school of transhumanism. Extropianism asserts that “an anarchistic market creates free and dynamic order, while the state and its life-stealing authoritarianism is entropic.”Footnote 17 This is illustrated, for example, by the fact that the idea for smart contracts was initially proposed in Extropy Magazine, a journal described as ‘a journal for transhuman thought’.Footnote 18 Many of the individuals who contributed to the early developments of blockchain technology identified as “crypto-anarchists” or “cypherpunks.” Consequently, they designed this technology to facilitate interactions and transactions beyond the reach of governments.Footnote 19 This technology is deeply rooted in this ideology, exemplified by the decentralised nature of the platform and its relatively anonymous operation. This unique background, combined with its specific goal, differentiates a blockchain platform from centralised platforms and databases in three ways.
First of all, the system is immutable. The cryptographic principles that underpin the platform ensure the integrity of the transactions and provide non-repudiation among the parties operating on that platform.Footnote 20 Furthermore, these same cryptographic principles offer “record immutability” due to the system’s method of ensuring consensus among network participants regarding the state of information on the network.Footnote 21
Second, the system is transparent. In a decentralised environment, the network participants cannot rely on a trusted third party to maintain the state of information. Doing so would go against the overarching goals. Instead, the parties rely on the consensus mechanism to collectively maintain the state of the database. To do so, parties require information about transactions and information contained within transactions. Hence, the decentralised nature of the platform, along with how the state of information in the database is maintained, results in transparency being a key characteristic of the system.Footnote 22
Third, the system is pseudonymous. This is a direct result of the asymmetric cryptography underlying these platforms, indicating that transparency is not absolute. Parties operate on the platform by way of a key pair, which consists of a public and a private key. The public key functions as both an account and an address that can be accessed by using the private key.Footnote 23 This is analogous to a person’s bank account: the public key can be compared to the bank account number, while the private key can be compared to the PIN.Footnote 24 The public key therefore functions as a pseudonym. Whilst it is very difficult to determine the person associated with a public key, it is not impossible. For this reason, the system is considered to be pseudonymous rather than anonymous.
For this contribution, the blockchain is therefore defined as a decentralised database or ledger in which the state of information is maintained by the participants collectively, characterised by a certain degree of immutability, transparency, and immutability.Footnote 25
2. Smart contracts
Some blockchains support programming.Footnote 26 If the programming language used by the system is sufficiently flexible, it becomes possible to run software on the platform.Footnote 27 The term “smart contracts’ specifically refers to these software applications.Footnote 28 This illustrates why, despite their unfortunate name, smart contracts do not necessarily constitute legal agreements.Footnote 29
Smart contract platforms introduce two additional key characteristics. First, smart contracts execute automatically; as a result of the platform’s transparent nature, these software applications can monitor the state of information. As soon as a predefined condition encoded in the software, the smart contract will execute. Second, the enforcement of the conditions outlined in the smart contract is also automated. The smart contract exists on the same database that maintains the parties’ assets, allowing the smart contract to automatically intervene in that patrimony. Hence, the two additional key characteristics enjoyed by smart contract platforms are their automatic execution and automatic enforcement.
3. Tokens
Smart contracts enable programmers to create applications on the platform. Examples of such applications can be found in supply chains, the Internet of things, and financial services.Footnote 30 Similarly, smart contracts can be programmed to create tokenisation platforms. These platforms allow for the creation of tokens and facilitate their trade. Such tokens are files that function as units representing a record.Footnote 31 One notable use case of this technology that has gained attention in recent years is the NFT trading platform. In this application, the token represents a connection to a specific piece of (digital) art.Footnote 32 In such cases the value of the token comes from a source outside the platform, making it exogenous. In contrast, a simple cryptocurrency like Bitcoin is endogenous, as its value derives from the database entry or the data string itself. Low and Megumi note that tokens are essentially entries within a database that associate particular assets with a particular public address and require a corresponding private key to transact, using asymmetric cryptography.Footnote 33
4. Legal challenges
Exogenous tokens function as a connector, linking the on-chain state of information to off-chain reality. Parties might transfer rights in assets by way of tokens.Footnote 34 In such scenarios, the tokens are on-chain representations of the rights referenced by the tokens.Footnote 35 A tokenisation platform aims to enable parties to trade the tokens, thereby transacting the rights as referenced by the tokens. For a platform to function effectively in this regard, several legal issues must be addressed. These challenges include, for example, the transfer of the token and the effectiveness of that transfer concerning the referenced assets.Footnote 36
III. Token container model
Several legal systems on the European continent have enacted legislative initiatives that provide a legal framework to address such challenges. This section will describe these frameworks and describe how they conceptualise the token as a container of rights or assets. By outlining this ‘token container model’, this section will illustrate how these legal initiatives align with the technical possibilities provided by this novel technology. However, as will be outlined in sections IV and V, while this model could serve as inspiration for other legal systems, its potential for outright implementation is limited.
1. The Belarusian presidential decree on the development of the digital economy
Presidential Decree No. 8 On the Development of the Digital Economy by the Republic of Belarus (‘the Decree’) reflects the treatment of tokens as containers for private law.Footnote 37 Regarding tokens, the Decree aims to create an environment that facilitates the introduction of blockchain-based technologies into the economy of the Republic of Belarus.Footnote 38 Annexed to the Decree is a list of definitions that defines a ‘digital sign (token) as a record in a register of transaction blocks (blockchain), (or) other distributed information system, which certifies that the owner of a digital sign (token) has rights to objects of civil rights and (or) which constitutes cryptocurrency.’ This definition explicitly states that the token can represent an object of property.Footnote 39 Savelyev highlights that this definition should be viewed in light of the objects recognised as objects of civil rights by the Civil Code of Belarus.Footnote 40 Article 128 of the Civil Code of Belarus specifies that objects of property include ‘tangible things, including money and securities, other property, including property rights, work and services, undisclosed information, exclusive rights to the results of intellectual activity and means of individualisation of participants in civil circulation, goods, works or services; non-material (intangible) benefits.’Footnote 41
2. Liechtenstein token act
The Principality of Liechtenstein published the Law on Tokens and Token Technology Service Providers on 2 December 2019 (hereinafter ‘TVTG’).Footnote 42 This Law states that its overall object and purpose is to “establish a legal framework for all transaction systems based on Trustworthy Technology and in particular governs the basis in terms of civil law with regard to Tokens and the representation of rights through Tokens and their transfer.”Footnote 43 Tokens are subsequently defined as “a piece of information on a Trusted Technology system which can represent claims or rights of memberships against a person, right to property or other absolute or relative rights.”Footnote 44 Lastly, the TVTG states that a token, when Liechtenstein law is applicable, is considered an asset located in Liechtenstein.Footnote 45 The overarching purpose of the Act is “to resolve key legal questions regarding the token economy.”Footnote 46
The second section of the TVTG addresses an important private law question: how can rights referenced by a token be transferred to another person? Liechtenstein has responded to this question by introducing what is explicitly referred to as the “Token Container Model.” According to this model, the token functions as a shell that, rather than creating new rights, can act as a container that represents rights.Footnote 47 When the token that represents a specific right is transferred, that underlying right is transferred in tandem with the token.Footnote 48 The proprietary aspects are governed autonomously through the Act.Footnote 49 This means that the TVTG serves as a lex specialis for tokens in the area of private law, particularly regarding property law and relevant aspects of contract law.
3. Risks of the token container model
Private law seeks to regulate relationships among private parties and has evolved over the centuries in response to various social, political, and economic changes. The purpose and evolution of private law systems, as well as the mutual differences amongst the national European private law systems, present challenges for regimes based on the token container model. Primarily because the laws of property and contract are deeply embedded in their respective national private law systems. The relationship and interactions between a lex specialis regime based on the token container model and other relevant areas of private law – such as inheritance law, insolvency law, and consumer protection law – remain unclear. The tokenisation of rights, for example, makes it possible to introduce a right or a legal concept in an area of private that it is not prepared to deal with it. The token container model therefore creates real risks of ‘legal irritants’. Unlike legal irritants created in the context of the integration of national laws, which Teubner warned against in 1998,Footnote 50 these legal irritants are purely internal and are caused by tokenisation technology enabled by the token container model.
Furthermore, the subject matter that the token container model aims to regulate is defined by its inherently transboundary nature, as tokens can move freely, disregarding natural and legal boundaries. This creates a serious risk of conflict of laws problems.Footnote 51
In conclusion, the token container model establishes a new legal regime specifically designed for tokens. However, how this system will integrate with broader private (international) law frameworks remains to be determined.
IV. Transfers of absolute rights
The initiatives discussed above illustrate how some European jurisdictions have devoted explicit attention to the private law implications of tokenisation. It should be noted that jurisdictions other than the Republic of Belarus and the Principality of Liechtenstein have enacted legislation on tokens and tokenisation.Footnote 52 However, many of these legislative initiatives have concentrated on areas other than private law. The Maltese approach, for example, is comprehensive and puts a strong emphasis on the service aspects of the financial industry, the prevention of market abuse, licensing and audit requirements, and the powers of regulators.Footnote 53 This reflects the more regulatory approach adopted by many jurisdictions, which indicates a different focus and devotes less attention to the private law aspects of tokenisation.
Generally, the regimes in Belarus and Liechtenstein are exceptions as most European jurisdictions lack specific frameworks addressing the proprietary aspects of tokenisation. Furthermore, for the initiatives in Liechtenstein and Belarus, it remains to be seen how those special regimes on tokenisation will develop and take root in the overarching proprietary and contractual legal frameworks and whether any legal irritants might surface.
This raises a more fundamental question: could the token container model be utilised in these legal systems as a conceptual framework to facilitate such transfers? It should be reiterated here that this paper examines tokens through the lens of Principle 3(1) of the Unidroit Principles of Digital Assets and Private law and therefore views tokens as transferable objects falling within the sphere of property law.Footnote 54 Therefore, rather than attempting to classify tokens or attempting to determine the requirements imposed on the transfer of the token itself, this paper analyses whether national rules on transfers are compatible with the use of tokens as an instrument to transfer rights.
Fox conducts a relevant thought experiment and explores how traditional private law aligns with these transfers of tokenised assets. He emphasises that, in attempting to fit these novel constructs in existing private law frameworks, the expectations of the people who deal in tokenised assets need to match the legal reality that underpins their transactions: the token representing the right and the asset to which the right relates must march in step with each other.Footnote 55 Building on Fox’s thought experiment, the remainder of this article examines the different European transfer systems using the analytical framework proposed by Van Vliet.Footnote 56 In doing so, subsection IV.1 elaborates on the transfers of absolute rights in movables. Subsection IV.2 examines transfer systems as they relate to immovable objects. Subsections IV.3 and IV.4 subsequently analyse whether such rules are compatible with token-based transfers of the underlying object. Subsection IV.4, unlike Konashevych’s work, which concentrated on the registration of property rights in immovables, will concentrate on the substantive rules governing transfers of rights in immovables.Footnote 57
1. Movables
The analytical framework proposed by van Vliet effectively uses a two-axis framework to categorise the different transfer systems for movable objects.Footnote 58 The first axis differentiates between systems that require only a legal agreement to effectuate the transfer and systems that require both a legal agreement as well as delivery in order to effectuate the transfer.Footnote 59 The former are referred to as consensual systems whilst the latter are referred to as tradition (or delivery) systems. Under a consensual system, simply reaching an agreement among the parties (solo consensu) is sufficient to transfer the object. A consensual system therefore does not require the delivery of the movable object to the transferee in order to effectuate the transfer of the right over the object.
A tradition system, contrary to a consensual system, requires both a legal agreement as well as the delivery of the object in order to effectuate a transfer thereof. Hence, the transfer is only completed after the transferee has obtained possession of the object in question. Hence, a transfer of a movable object in a tradition system requires both a legal agreement (or ‘title’) as well as a delivery of that object to effectuate the transfer.Footnote 60 Tradition systems can be further classified along the axis: separating system with a causal nature from systems with an abstract nature.
Causal transfer systems connect the validity of the transfer to the validity of the underlying legal agreement. Therefore, if the underlying legal agreement is avoided, the right over the movable object reverts back to the transferor with retroactive effect.
This is quite apparent in consensual transfer systems: since only a legal agreement is required in order to transfer a movable object, the right in that object reverts to the transferor retroactively upon avoidance of that legal agreement.Footnote 61 Abstract systems, on the other hand, disconnect the validity of the underlying legal agreement from the validity of the transfer. Therefore, if the underlying legal agreement is avoided, the right over the movable object does not retroactively revert to the transferor. Instead, the transferor obtains a personal right to have the object retransferred to him.Footnote 62
Three different transfer systems can be identified within the main European civil law traditions: consensual systems, causal tradition systems, and abstract tradition systems.Footnote 63 The French transfer system, first of all, is an example of a consensual system.Footnote 64 Only a legal agreement is required to transfer a movable object. Additionally, this consensual character introduces a causal nature; in other words, the validity of the legal agreement that transfers ownership is necessary to transfer the right over the object.Footnote 65 As a result, if the contract is void, the transfer of the property reverts to the transferor with retroactive effect.Footnote 66 Second, the Dutch system serves as an example of a causal tradition system. This system requires both a legal agreement as well as a delivery of the movable object to effectuate the transfer. Its causal nature dictates that, as far as the effectiveness of the transfer is concerned, the two are connected. Therefore, if the legal agreement is void, the transfer itself is avoided as well. The object therefore reverts to the transferor with retroactive effect upon avoidance of the legal agreement.Footnote 67 Third, the German system on transfers of movables is an example of an abstract tradition system. According to the German rules on transfers, the validity of the transfer of a movable object is determined independently of the legal agreement that underlies it.Footnote 68 This means that, unlike in a system that is causal in nature, a transfer according to the German rules on transfer remains valid even if the underlying legal agreement is avoided.
Hence, a key practical difference between causal and abstract systems lies in the claim that the transferor has vis-à-vis the transferee following the avoidance of the title. In a causal tradition system, ownership reverts to the transferor automatically after the legal agreement is avoided. Hence, the transferee loses the right of ownership over the movable object whilst still being in possession. This provides the transferor with the right to revindicate: a strong claim based on his right of ownership. An action of revindication entails a claim to have the possession of the object restored to the person making the claim on the basis of him being the owner. In an abstract system ownership does not automatically revert to the transferor. This means that the option to revindicate is not available to him. Instead, the transferor now has a personal right vis-à-vis the transferee to have possession of the movable restored to him. This is a personal action that the transferor can only effectuate against the transferee. Consequently, if the transferee has transferred the object to a third party in good faith, the transferor’s only recourse is to seek damages.Footnote 69
2. Immovables
The rules on transfers of immovables generally reflect the rules on transfer of movables.Footnote 70 However, the intrinsic characteristics of immovable objects shape how the relevant legal requirements are applied.
In a consensual system, such as the French system, transfers can, strictly speaking, be effectuated by legal agreement alone.Footnote 71 However, the personal nature of such legal acts dictates that the effect of such transfers is limited to the parties alone.Footnote 72 Registration with the land registry is necessary to generate third-party effects. A failure to do so therefore results in a transfer that is ineffective vis-à-vis third parties and therefore does not enjoy erga omnes effects.Footnote 73 To register the sale with the land registry, a notarial act is required. The subsequent registration in the land registry has declaratory, rather than constitutive, effect.Footnote 74
Tradition systems require both a valid title and the delivery of the immovable object to transfer the right in that object from the transferor to the transferee. In voluntary transfers, the title is typically the sales agreement. Additionally, the delivery of the immovable object is done by way of a subsequent legal act. The German rules on transfer require, for example, a dingliche Einigung (or a “real agreement”) to satisfy the requirement for a valid title.Footnote 75 This real agreement is personal in nature and is therefore, generally speaking, not subject to specific form requirements.Footnote 76 However, when the transfer in question concerns the transfer of ownership in an immovable object, a special variation of the real agreement is required: the Auflassung (or ‘conveyance agreement’).Footnote 77 This requirement is subject to formal requirements, including the stipulation that it must be drawn up by a notary. Such notarial recordation, the notarielle Beurkundung, is necessary before the registration of the transfer by the notary in the land registry. This registration of the Auflassung serves as the delivery of the immovable, thereby satisfying the second requirement of the German tradition system.Footnote 78
3. Using tokens to transfer rights in movables
The transfer of a property right requires a valid title. In a consensual system, where a legal agreement can serve as a title and by itself is sufficient to transfer rights over a movable object, a tokenisation platform could effectively facilitate such transfers. In such cases, the solo consensu principle dictates that the transfer takes place the moment agreement is reached. Whether such a tokenisation platform could effectively transfer the right in question is dependent on whether the underlying legal agreement can be incorporated on that platform and within that token in a manner that is recognised by the legal system.
This is different in tradition systems as these systems require a delivery of the movable object in addition to the legal agreement. These systems therefore consist of both a contractual and a proprietary element. The delivery is the proprietary element that, in addition to the title, effectuates the transfer. However, the token merely references an asset. While a legal system might consider a token to be a digital asset in and of itself: such legal qualifications do not change the fundamental nature of the token. A token is, at its very foundation, a reference to an asset. At this point it should be reiterated that such qualification questions are outside the scope of this current research. This paper takes principle 3(1) of the Unidroit Principles on Digital Assets and Private Law as its starting point and defines tokens as transferable objects that fall within the scope of the law on property. By doing so, this paper can confront another fundamental question: whether the general rules on transfers are compatible with the use of tokens to transfer rights in movable or immovable objects, and claims. Therefore, while the token might be a digital asset, and perhaps even an object of property laws, as far as the tokenisation of assets is concerned; the token is simply a digital representation of the asset and cannot be equated to the asset itself. Therefore, if a token references a tangible movable object, that token is a digital representation of that movable object. Control over the token can therefore not be equated to possession of the movable object. It follows from this the that transfer of the token from one party to another cannot constitute delivery of the underlying movable object. Hence, the token cannot be equated to the object itself, control over the token cannot be equated to possession of the referenced object, which means that the transfer of the token from one party to another does not generate any change in the physical control over the object that is associated with the token. This means that a transfer of a token from one wallet to another cannot constitute delivery of the object. In other words, the token merely represents a right in an asset; its transfer is, in principle, unable to provide any actual control over the object and therefore has no constitutive effect. As far as tradition systems are concerned, this means that even if the legal system recognises the token or its contents as a valid title, or the transfer of a token as proof of a valid title, the transfer of the token alone cannot constitute the delivery of the referenced object itself. Therefore, in tradition systems, transferring a token that references a right over a movable object cannot result in a valid legal transfer: the proprietary element that tradition systems require in addition to the contractual element cannot be satisfied through a token transfer.
However, tradition systems recognise alternative methods of delivery to effectuate a transfer without transferring possession of the object.Footnote 79 These include, for example, the tradio constitutum possessorium and the traditio brevi manu for Dutch law. Kopalit, Verheul and Verstijlen demonstrate that both these instruments, the tradio constitutum possessorium more so than the tradio brevi manu, have inherent characteristics that are incompatible with the tokenisation of rights in movable objects.Footnote 80 The instrument of Besitzkonstitut fulfils a similar role in German law.Footnote 81
While the most common way to deliver a movable object to the transferee is to provide them with physical possession of it, physical custody of the object is not a requirement of possession. Methods to provide possession of an object to the transferee without providing him with the physical custody offer methods to transfer movable objects using tokens. One could provide the transferee possession of the object without giving him actual physical custody over the object by enabling him to exercise actual and exclusive control over the object and ensure that any control by the transferor over the object ceases.Footnote 82 If a token can be designed so that its holder has actual and exclusive control over the movable object to which it is connected, then the transfer of that token might satisfy the delivery requirement in a tradition system. One could consider, for instance, a token that functions much like a key to the movable object itself or the location where the movable object is stored. This can only occur if three conditions are met. First, the legal system in question recognises the token, its contents, or the smart contract that transfers it as a valid title. Second, the transferor must relinquish actual and exclusive control over the referenced asset upon transferring the token. Third, the transferee obtains exclusive and actual control over the object upon receiving the token. Only then can a tradition system recognise this as a legal transfer of the rights associated with the asset referenced by the token.
a. Retroactivity and the risk of discrepancy in causal systems
The retroactive effect of avoiding the legal agreement in causal systems introduces an additional obstacle. This affects both tradition systems of the causal variation and consensual systems. In these systems, a transfer is considered retroactively null and void if the underlying title is avoided. Practically, this means the transfer is treated as though it never occurred and the right in question never passed. This differs from abstract systems, where the transfer is not deemed retroactively null and void. Instead, the transferor obtains a personal right vis-à-vis the transferee to have possession of the object restored to him.
A tokenisation platform does not allow for such retroactivity. If a legal agreement is avoided in a causal system, the law recognises the transferor as the person entitled to the asset. However, the tokenisation platform is unable to access information regarding the avoidance of the title. This creates a discrepancy between the rights recognised by the law and those recognised on the platform. The platform, unable to access information on the invalidity of the legal agreement, will list the transferee as entitled to the asset. After all, the transferee holds the token that represents the asset. However, the law designates the transferor as entitled to the asset. Thus, a discrepancy exists. This discrepancy cannot arise in abstract systems as the transfer is not retroactively avoided. Consequently, unlike abstract systems, causal systems face an additional obstacle: their retroactive nature creates situations in which the law automatically treats a transfer as if it never took place, whilst the tokenisation platform is unable to take account of that information.
4. Using tokens to transfer rights in immovables
Fundamental obstacles exist with regard to token-based transfers of property rights in immovables. Legal systems impose strict requirements on the method and form of the sale and the delivery of immovable objects. As discussed in section IV.2, legal systems impose three requirements for transferring rights in immovable objects with third-party effect. First, the legal agreement must adhere to specific form requirements. Second, the transfer must be authenticated by a competent authority. Third, it must be registered in a public registry.
All three additional requirements for transferring rights in immovable objects create further obstacles to transferring these rights using tokens. First, a token alone cannot satisfy the specific form requirements imposed on the legal agreement. Second, legal systems impose strict requirements on authentication by competent authorities. Again, a token or a transfer thereof is unable to satisfy those requirements and therefore unable to be authenticated by a competent authority. Third, given that these token transfers cannot be authenticated, they cannot be registered in a public registry. Moreover, even if they were authenticated, the specific requirements imposed by legal systems on registration make it impossible for such token transfers to be registered. Therefore, transferring property rights in immovable objects using tokens is impossible under current legal regimes.Footnote 83
V. Transfers of personal rights: assignment of claims
This final section addresses the use of tokens to transfer claims rather than property rights.Footnote 84 Where the previous section discussed the token-based transfers of rights in movables and immovables, this final section will analyse token-based transfers of personal rights by exploring the assignment of claims.
Claims are amalgamates of contractual and proprietary aspects. Fundamentally, a claim represents a right to performance vis-à-vis another specific person.Footnote 85 These rights are personal in nature, meaning that a claimant with a right to enforce a debt can only do so against his debtor. The value of the claim is equal to the amount of money owed to the claimant. A claimant might wish to transfer his claim to another person. He could, for example, use the claim for security purposes. This might require the transfer of the claim. In such cases, while the claim itself is personal in nature, the transfer of the claim treats the claim as a proprietary object. Therefore, the transfer of a claim takes a contractual relationship and treats it as an object of property law.Footnote 86
Although significant differences exist between the rules on transfers of absolute rights and those governing the assignment of claims, the principles underpinning the rules on transfers of absolute rights are reflected in the rules on the assignment of claims. Generally speaking, an assignment of a claim from one party to another has inter partes effect as soon as an agreement is reached. The French Code Civil, for example, specifies that a contract for the assignment of a claim must be made in writing.Footnote 87 Before the 2016 reform of the French law of obligations, a subsequent action was required to ensure the assignment of the claim was effective vis-à-vis the debitor cessus. This could be done in one of two ways: the debtor might be given a notification or the assignment could be authenticated.Footnote 88 Following the reform, the distinction between the effects of the assignment between the parties and its effects vis-à-vis third parties has been maintained, albeit in a more nuanced manner. The effects of an assignment vis-à-vis third parties are now linked to the effects between the parties.Footnote 89 The claim is enforceable against the debtor from the moment of assignment. In case a dispute arises, the assignee has to establish that the debtor had consented to the assignment, had been notified of the assignment, or had taken note of the assignment.Footnote 90
The causal elements in the French rules on the assignment of claims are also present in the Dutch system. This system recognises two variations of assignment: a public variation and a silent variation. Both variations of assignment require a title and delivery. The title requirement typically requires a legal agreement, whilst the delivery is done through a deed. This deed could either be a notarial deed (authentieke akte) or a registered private deed (onderhandse akte).Footnote 91 Notification of the debtor is a constitutive requirement for public assignments.Footnote 92 Therefore, the public assignment is only complete following the notification of the debtor. It is also at this point that the debtor must pay the assignee rather than the assignor. If the parties opt for a silent assignment instead, the assignee becomes the debtor’s creditor once the deed – whether authenticated by a notary or registered private deed – is recorded. However, the assignee cannot collect the debt as long as the debtor has not been notified of the assignment. This means that the debtor, who is unaware of the assignment of his debt, is allowed to pay off the debt to the assignor as long as the assignment remains silent.Footnote 93 The assignor ceases to be the debtor’s creditor only after the debtor has been notified of the assignment. Therefore, before notification, the debtor can pay off his debt to the assignor; afterwards, he must pay the assignee.Footnote 94
The Burgerliches Gesetzbuch treats the transfer of a claim as purely a contractual matter.Footnote 95 The central requirement imposed by the German system on the assignment of claims is therefore the existence of a legal agreement between the assignor and the assignee. The BGB imposes no formal requirements concerning the form of this agreement.Footnote 96 Moreover, the German system does not require any registration thereof or notification to the debtor of the assignment.Footnote 97 As a matter of law, the German system on the assignment of claims requires no subsequent legal act in addition to the legal agreement.Footnote 98 As long as the debtor has not been notified of the assignment, he can pay the assignor. While strictly speaking, the assignor is no longer the creditor, the German system treats the assignment without notification as an authorisation for the assignor to collect on the new creditor’s behalf.Footnote 99
1. Using tokens to transfer claims: tokenised claims
Causal systems, such as those in France and the Netherlands, present challenges to assigning claims using tokens. The French system, for example, requires the legal agreement to be in writing.Footnote 100 Any platform under French law that uses tokens to transfer claims must ensure this requirement is met in a manner recognised by the Code Civil. To enable the use of tokens to actually transfer claims, this requirement can be met in satisfied one of two ways: either the legal agreement exists on-chain, or it remains off-chain in a way that allows the platform to accurately interface with it. This approach ensures that all necessary information for the assignment is readily available. To transfer a claim, details such as the validity of the agreement, the identities of the parties, and notification to the debtor might be required. If a platform does not meet this requirement, it cannot be used to assign claims directly; instead, it can only serve as an administration for claims that have been assigned.
It should be noted though that, with the 2016 reform, the French system appears to be more accommodating to token-based assignments of claims. As discussed in the previous section, since the reform, the assignment of a claim has been enforceable against the debtor from the moment of the assignment. This introduced the principle that, in the event of a dispute, the burden of proof for establishing that the debtor was notified or consented to the assignment rests with the assignee. This principle aligns better with token-based transfers. The transparent nature of the underlying technology, combined with the automatic execution and automatic enforcement of the software, enables a platform to be programmed so that the debtor can be notified the moment the assigning parties decide to assign his claim.
In Dutch law, public assignment and silent assignment offer similar opportunities. However, as was discussed above, the Dutch system imposes more formal requirements compared to French law. A distinctive requirement for silent assignments under Dutch law is the obligation to register a private deed with the tax authorities. Although this deed does not require a specific form, it must be made in writing and signed by the parties. Given these requirements, it is unlikely that a token could qualify as a private deed. Furthermore, the tax authorities require a physical deed. Therefore, a token, a smart contract or an agreement embedded in a token or expressed through a token transfer, cannot be registered as a private deed. Consequently, the formal requirements under Dutch law present additional obstacles to the transfer of tokenised claims.
The German system, which regulates the assignment of claims through the law of contract rather than the law of property, creates opportunities for tokenisation platforms that enable the transfers of claims. Since the assignment of claims is governed by contract law, the principle of freedom of contract is central. This places greater emphasis on the autonomy of the parties compared to the formal requirements of the abstract tradition system.
VI. Conclusion
The paper set out to examine whether European transfer systems are compatible with the use of tokens to transfer rights in objects. It has used the token container model as a conceptual framework by which token-based transfers of rights could be introduced into existing systems of law. Whilst the token container model aligns well with the opportunities offered by the technology, fundamental legal obstacles exist that prevent European transfer systems from accommodating token-based transfers.
This paper has identified three fundamental obstacles that bar the use of tokens for transferring rights in movable and immovable objects, as well as claims, within European transfer systems. First, the delivery requirement in certain transfer systems provides obstacles to the accommodation of token-based transfers. In order to transfer a right in a movable object to the transferee, the actual movable object must be delivered. This requirement cannot be satisfied by providing the transferee with a token rather than with the actual object of the transfer. Hence, the delivery requirement in tradition systems provides an obstacle to their compatibility with the use of tokens to transfer rights in movables. However, legal instruments that recognise the conferral of actual and exclusive control over a movable object on the transferee—rather than providing physical possession—may serve to satisfy the delivery requirement in tradition systems. In order to do so, the token must be programmed to grant its holder actual and exclusive control over the object. This means that the person who transferred the token to another person loses actual control over the object, whilst the other person gains actual control. It should be noted though that, even if such a software solution could be conceived, only tradition systems of an abstract nature could potentially accommodate such token transfers. After all, tradition systems of a causal nature are affected by the second obstacle: the compatibility of transfer systems with the use of tokens to transfer rights is limited by the manner in which they address the potential avoidance of the underlying legal agreement. A token-based transfer in a consensual system, provided the token-transfer is reflective of the consensus between the parties, might be effective in transferring the underlying movable object. However, causal systems, including both consensual and causal tradition systems, prevent the token from providing an accurate reflection of the right in the object. This is caused by the possibility of retroactive avoidance: if the legal agreement that effectuated the transfer is avoided, the right in question automatically reverts to the transferor. It seems impossible to reflect such a legal construct on a tokenisation platform. Third, the formal requirements imposed by certain transfer systems limit their compatibility with the use of tokens to transfer rights. This is particularly evident in the context of transfers of rights in immovables and, to a lesser extent, in the context of the rules on the assignment of claims. With regard to the latter, systems on assignment of claims that have a proprietary character, such as those in France and the Netherlands, introduce certain formal requirements that token-based transfers are unable to meet. These formal requirements bar the use of tokens to transfer claims, at least under the general rules on the assignment of claims. However, other systems—such as the German system—approach the assignment of claims from a contractual rather than a proprietary perspective. In those systems no such formal requirements exist. These systems are therefore more compatible with token-based assignments of claims.
The proprietary transfer systems within the scope of this research, in their current form, provide little, if any, room for the use of tokens to transfer rights. However, certain specific implementations of the technology may require tokens with a particular design and particular content. Such specific technology applications, and the context in which they are used, might make those particular tokens eligible for certain legal qualifications that introduce specific rules regarding their transfer. To illustrate, one could imagine a tokenised claim bearing particular resemblance to, for example, promissory notes or commercial papers. Depending on the design and implementation of the technology, as well as the context in which it is used, specific rules on transfer that apply to such instruments could provide a different level of compatibility for tokens. These might include the rules on transfers of transferable securities, but might also include the rules on the transfer of bearer instruments. Other examples of more specific qualifications that might make available more tailored rules on transfers exist as well. One could imagine movable objects in carriage being tokenised. Such tokens could bear a certain resemblance to bills of lading. This is another example of a situation in which a token might enjoy a specific qualification that provides specific rules on transfer. Future research is needed to map the types of scenarios in which tokens might enjoy a qualification that provides tailored rules on transfer and to analyse whether those rules on transfer are compatible with the use of tokens to transfer rights.
Acknowledgments
A previous version of this paper was presented at the “IALS Annual Symposium: Property Law, Climate Change, Tech Disruption” at the Australian National University and “The Blockchain and Digital Assets Conference” at HEC Paris. This paper was finalised in connection with the project “Land Registry’s Current Challenges: Blockchain and Data Protection” (PID2020-113995GB-I00), financed by the State Research Agency (Spanish Ministry of Science and Innovation), and presented at the Universidad de La Laguna. The author would like to thank the anonymous reviewers for their valuable comments as well as Jason Grant Allen, Sjef van Erp, Ilse Marije Lambers, Joseph Lee, Christina Lemke, Anne McNaughton, and Tim van Zuijlen. Any remaining errors are the responsibility of the author.
Competing interests
The author has no conflicts of interest to declare.
Open access
