[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	The Law Commission
You are here: BAILII >> Databases >> The Law Commission >> Digital Assets [2023] EWLC 412 (30 July 2024) URL: http://www.bailii.org/ew/other/EWLC/2023/LC412.html Cite as: [2023] EWLC 412

[New search] [Printable PDF version] [Help]

Law Commission

Reforming the law

Digital assets: Final report

Law

Commission

Reforming the law

Law Com No 412

Digital Assets: Final report

Presented to Parliament pursuant to section 3(2) of the Law Commissions Act 1965

Ordered by the House of Commons to be printed on 27 June 2023

HC 1486

© Crown copyright 2023

This publication is licensed under the terms of the Open Government Licence v3.0 except where otherwise stated. To view this licence, visit nationalarchives.gov.uk/doc/open-government-licence/version/3.

Where we have identified any third party copyright information you will need to obtain permission from the copyright holders concerned.

This publication is available at www.gov.uk/official-documents.

Any enquiries regarding this publication should be sent to us at [email protected].

978-1-5286-4173-9

E02924167 06/23

Printed on paper containing 40% recycled fibre content minimum

Printed in the UK by HH Associates Ltd. on behalf of the Controller of His Majesty’s Stationery Office

The Law Commission

The Law Commission was set up by the Law Commissions Act 1965 for the purpose of promoting the reform of the law.

The Law Commissioners are:

The Right Honourable Lord Justice Green, Chairman

Professor Sarah Green

Professor Nicholas Hopkins

Professor Penney Lewis

Nicholas Paines KC

The Chief Executives of the Law Commission are Stephanie Hack and Joanna Otterburn.

The Law Commission is located at 1st Floor, Tower, 52 Queen Anne's Gate, London SW1H 9AG.

The terms of this report were agreed on 17 May 2023. All website links were correct at 23 June 2023.

The text of this report is available on the Law Commission's website at

https://www.lawcom.gov.uk/project/digital-assets/.

Contents

Glossary

List of abbreviations

• CHAPTER 1:    INTRODUCTION

Digital assets and the law of personal property

About this project

Other work on digital assets

The team working on this paper

Acknowledgements and thanks

• CHAPTER 2:    SUMMARY OF OUR RECOMMENDATIONS AND

CONCLUSIONS

Our tripartite approach to law reform in this report

The principles underlying our approach

Summary of our recommendations and conclusions

CHAPTER 3: A “THIRD” CATEGORY OF THING TO WHICH PERSONAL

PROPERTY RIGHTS CAN RELATE

Overview

Are certain digital assets things in possession?

Are certain digital assets things in action?

A third category thing — analogies with both things in possession and things in action

Avoiding unnecessary boundary issues

Terminology: data objects, digital objects and third category things

Our recommendation: minimalist statutory confirmation

• CHAPTER 4:    OUR THIRD CATEGORY RECOMMENDATION AND

CONCLUSIONS IN PRACTICE

Introduction

Our proposed criteria

Different digital assets — applying technology neutral principles in a way that is responsive to specific technology

• CHAPTER 5:    CONTROL

Introduction

Factual and legal control

Factual control

Common law development and technical expert group

The legal consequences of control

• CHAPTER 6:    TRANSFERS

Introduction

Background to concepts used in this report

A transfer operation that effects a state change

Transfers by “a change of control”

Derivative transfers of title

A common law special defence of good faith purchaser for value without notice

• CHAPTER 7:    INTERMEDIATED HOLDING ARRANGEMENTS

Introduction

Defining and distinguishing intermediated holding arrangements

Legal frameworks for intermediated holding arrangements

• CHAPTER 8:    COLLATERAL ARRANGEMENTS

Introduction

Crypto-token collateral arrangement structuring options under the current law

Application of the FCARs regime to crypto-tokens, cryptoassets, and mere record/register tokens

Developing a bespoke legal framework for crypto-token collateral arrangements

• CHAPTER 9:    CAUSES OF ACTION AND ASSOCIATED REMEDIES

Introduction

Breach of contract

Vitiating factors

Following and tracing

Breach of trust, equitable wrongs, and constructive trusts

Burning tokens

Proprietary restitutionary claims at law

Glossary

List of abbreviations

Digital assets

To the Right Honourable Alex Chalk MP, Lord Chancellor and Secretary of State for Justice

• Chapter 1:    Introduction

• 1.1    Digital assets are fundamental to modern society and the contemporary economy. They are used for an expanding variety of purposes — including as valuable things in themselves, as a means of payment, or to represent or be linked to other things or rights — and in growing volumes. Electronic signatures, cryptography, distributed ledgers, smart contracts and associated technology have broadened the ways in which digital assets can be created, accessed, used and transferred.

• 1.2    Such technological development is set only to continue. As technology advances and humans spend increasing amounts of time online, our relationships with digital assets will become ever more important.

• 1.3    The term digital asset is extremely broad. It captures a huge variety of things including digital files, digital records, email accounts, domain names, in-game digital assets, digital carbon credits, crypto-tokens and NFTs. The technology used to create or manifest those digital assets is not the same. Nor are the characteristics or features of those digital assets.

• 1.4    This report considers how principles of private law, specifically personal property law, apply to digital assets and concludes that certain types of digital assets are capable of being things to which personal property rights can relate.

DIGITAL ASSETS AND THE LAW OF PERSONAL PROPERTY

• 1.5   Personal property rights are vital to modern social, economic and legal systems.¹

• 1.6   Property rights are important for many reasons. The concept of property rights is

widely used in statutes and cases and property rights feature in most commercial transactions relating to things of value. Property rights are important for the proper characterisation of numerous modern and complex legal relationships, including intermediated holding arrangements, collateral arrangements and structures involving trusts. Property rights are also important in cases of bankruptcy or insolvency, when objects of property rights are interfered with or unlawfully taken, and for the legal rules concerning succession on death.² Property rights are useful because, in principle, they are recognised against the whole world, whereas other — personal — rights are recognised only against someone who has assumed a relevant legal duty.

• 1.7    We demonstrate in this report that the law of England and Wales has proven itself sufficiently resilient and flexible to recognise some digital assets as capable of being things to which personal property rights can relate. We conclude that the law in this respect is now relatively certain and that most areas of residual legal uncertainty are highly nuanced and complex. That complexity remains, in part, because both the digital asset market and the technology in question is evolving and will continue to do so. We identify the remaining areas of residual legal uncertainty and recommend law reform to reduce that uncertainty, but in a way that acknowledges the different features of those different digital assets.

• 1.8    The law reform that we recommend in this report aims to ensure that the legal system, as part of a wider social framework, can reinforce the overall strength of digital asset ecosystems (which also rely on social elements). Our recommendations also aim to ensure that the private law of England and Wales remains a dynamic, globally competitive and flexible tool for market participants in the digital asset space.

ABOUT THIS PROJECT

Background

• 1.9    In March 2020, the Ministry of Justice asked the Law Commission to review the law on crypto-tokens and other digital assets and to consider whether the law of England and Wales required reform to ensure that it can accommodate such assets. Our full terms of reference are set out at Appendix 1.

Call for evidence, interim update paper and consultation paper

• 1.10    We published a call for evidence on digital assets on 30 April 2021.³ The purpose of our call for evidence was to give stakeholders and market participants an opportunity to input to us ahead of our formal consultation paper. It sought views about, and evidence on, the ways in which digital assets were being used, treated and dealt with by market participants. We received 37 responses.⁴

• 1.11    We published an interim update paper on our digital assets project on 24 November 2021.⁵ This included an update on the scope and detail of our work, reflecting the responses to our call for evidence.

• 1.12    We published our consultation paper on digital assets on 28 July 2022.⁶ It included a detailed consideration of private law issues relating to digital assets. Our consultation paper described our provisional proposals for law reform and asked 47 specific questions relating to those provisional proposals.

• 1.13    We received 81 ⁷responses to our consultation paper from practising lawyers, academics, technologists, industry bodies, digital asset and crypto-related businesses and other commercial entities. Many responses were highly sophisticated, precise and persuasive. We considered the responses in detail and have developed our analysis to reflect many of the points raised by consultees. We think that it is important that the responses to our consultation paper are made publicly available so that market participants, the courts and legislators (both in England and Wales and globally) can consider the arguments for and against our ultimate recommendations and conclusions in this report. In this report, we refer both to consultee responses that support our ultimate conclusions and those that do not. We consider that our consultation paper, the responses and this report provide a detailed and comprehensive discussion of the complex and highly nuanced residual areas of legal uncertainty that we attempt to resolve through our recommendations and conclusions. The responses to our consultation paper are available on our website.⁸

• 1.14    The wider context in which we produced this report is also important. The issues we discuss inevitably and intentionally dovetail with the work of other public and private bodies and initiatives. This includes the judiciary itself, including senior and specialist judges who have, over many years, considered the issues we discuss in this report. It also includes the UK Jurisdiction Taskforce (“UKJT”),⁹ international law reform initiatives such as the UNIDROIT Digital Assets and Private Law Working Group (“UNIDROIT Working Group”), ¹⁰the American Law Institute and the Uniform Law Commission’s Uniform Commercial Code and Emerging Technologies Committee (“UCC Committee”), ¹¹the Expert Reference Group on Digital Assets in Scots Private Law (“ERG”) ¹²and multiple regulatory initiatives. ¹³We conducted this project in close alignment with some of that work, with team members either participating in, or acting as observers to, the work of the UKJT, the UNIDROIT Working Group, the UCC Committee and the ERG. We have also had regular and in-depth discussions with senior and specialist judges on many of the issues considered in this report. Members of the judiciary said that they would welcome further a detailed consideration of the highly complex and nuanced remaining areas of residual legal uncertainty, particularly where complex technological concepts are involved. The Master of the Rolls has, in particular, welcomed our analyses and we are equally grateful for the ongoing work of the UKJT (which he chairs) and their input into our process.

• 1.15    Our report is therefore the culmination of two years of consultation with market participants and experts in the area. Their views and their markets have developed significantly during that time. Our recommendations and conclusions aim to ensure that the highly iterative and innovative process of evolution of the digital assets markets can continue under the law of England and Wales.

Territorial extent

• 1.16    As the Law Commission of England and Wales, we can make law reform recommendations only for England and Wales. This paper considers matters of private law, which are devolved in Scotland and transferred in Northern Ireland. There are differences between the law of personal property in England and Wales and its equivalent in Scots law. This paper does not identify those differences and does not address Scots law or the law of Northern Ireland.

OTHER WORK ON DIGITAL ASSETS

• 1.17    Recently, several UK initiatives have analysed the implications of the development of digital assets, particularly cryptoassets.¹⁴ The UK’s approach to digital assets has been characterised by careful, iterative reform based on detailed consultation with market participants. Some such projects include:

• (1)    three publications by the UKJT:

• (a)    a legal statement on cryptoassets and smart contracts;¹⁵

• (b)    a set of digital dispute resolution rules;¹⁶ and

• (c)    a legal statement on the issuance and transfer of digital securities;¹⁷

• (2)    HM Revenue and Customs’ Cryptoasset Manual, ¹⁸its call for evidence on the taxation of decentralised finance ¹⁹and its consultation on extending the investment manager exemption to include cryptoassets;²⁰

• (3)    the Bank of England and HM Treasury’s joint consultation on central bank digital currencies;²¹

• (4)    the Financial Conduct Authority’s guidance on cryptoassets;²²

• (5)    HM Treasury’s consultation on cryptoassets and stablecoins, ²³and its consultation on a future financial services regulatory regime for cryptoassets;²⁴ and

• (6)    the Bank of England’s analysis of cryptoassets and financial stability.²⁵

Related past and current Law Commission work

• 1.18    In addition to this work, we have completed or are working on four other digital asset related projects.

Past work

• 1.19    Smart legal contracts: We undertook a detailed analysis of the current law as it applies to smart legal contracts, highlighting any uncertainties or gaps, and identifying such further work as may be required now or in the future. We published our advice to Government in November 2021, concluding that the current legal framework in England and Wales is clearly able to facilitate and support the use of smart legal contracts.²⁶

• 1.20    Electronic trade documents: We undertook a project to make recommendations to enable the legal recognition, in electronic form, of certain trade documents such as bills of lading and bills of exchange. We published our final report with draft legislation in March 2022. ²⁷The Electronic Trade Documents Bill, based on our recommendations, was introduced into Parliament in October 2022 and is currently in the House of Commons’ committee stage.²⁸

Ongoing work

• 1.21    Digital assets: which law, which court?: We have agreed with the Government that we will undertake a project looking at the private international law questions associated with emerging technology, including digital assets and electronic trade documents, and consider whether reform is required. We are conducting initial research and scoping out this work.²⁹

• 1.22    Decentralised autonomous organisations (“DAOs”): In simple terms, a decentralised autonomous organisation (“DAO”) is a novel type of technology-mediated structure or organisation of participants made up of several composite elements. The Government asked the Law Commission to explore and describe the treatment of DAOs under the law of England and Wales and identify options for how DAOs should be treated in law in the future in a way which would clarify their status and facilitate their uptake. We published a call for evidence in November 2022. ³⁰We are analysing the responses to the call for evidence and preparing a scoping paper.

THE TEAM WORKING ON THIS PAPER

• 1.23    The following members of the commercial and common law team worked on this report: Laura Burgoyne (team manager), Matthew Kimber (team lawyer), Amila Kulasinghe (team lawyer), Daniella Lupini (team lawyer), Luke Broadway (research assistant), Joanna Connolly (research assistant) and Diana Stoean (research assistant). Additional support was provided by team lawyers Nathan Tamblyn and Peter Hunn and research assistant Thomas Wingfield.

ACKNOWLEDGEMENTS AND THANKS

• 1.24    In Appendix 2 to this report, we include a list of consultees who responded to our call for evidence and those who responded to our consultation paper, together with a list of stakeholders with whom we have met during the project. We are grateful to and would like to thank all those who took the time to respond to our call for evidence and/or our consultation paper or who have otherwise met with us or responded to other requests for assistance or information in support of this work.

• 1.25    We are also grateful for the feedback and comments from our advisory panel of experts, who commented on drafts of our work and shared their expertise with us. The members of the advisory panel are listed in Appendix 2.

• Chapter 2:    Summary of our recommendations and conclusions

OUR TRIPARTITE APPROACH TO LAW REFORM IN THIS REPORT

• 2.1    In this report, we make very few recommendations for law reform. That is for two reasons. First, because we conclude that the common law of England and Wales is, in general, sufficiently flexible, and already able, to accommodate digital assets. Second, because we want our recommendations to be highly practicable and achievable, to impact the law of England and Wales in a positive way and to be as direct and as implementable as possible.

• 2.2    In this respect, we agree with the view of Sir Geoffrey Vos (speaking extra-judicially) that:³¹

We should try to avoid the creation of a new legal and regulatory regime that will discourage the use of new technologies rather than provide the foundation for them to flourish.

• 2.3    We take a tripartite approach to law reform.

• (1)    We conclude that the common law of England and Wales is, in general, sufficiently flexible, and already able, to accommodate digital assets and therefore that any law reform should be through further common law development where possible.

• (2)    We recommend targeted statutory law reform only to confirm and support the existing common law position, or where common law development is not realistically possible.

• (3)    We recommend making arrangements for the provision of further guidance from industry experts which would support both the common law and statute.

• 2.4    We explain this approach below and summarise our conclusions and recommendations from paragraph 2.44.

Prioritising common law development

• 2.5    First, we use this report in large part to champion the common law of England and Wales and to draw its successes in the digital asset (including the crypto-token and cryptoasset) markets to the attention of market participants. Our analysis is intended to form the foundation on which further common law development can be based. We conclude that the law in this area is now relatively certain and that most areas of residual legal uncertainty are highly nuanced and complex. That complexity remains, in part, because both the digital asset markets and the technology that supports them are evolving and will continue to do so. We discuss these remaining areas of residual legal uncertainty and draw conclusions as to the most appropriate way for the common law to develop.

Targeted statutory law reform

• 2.6    Second, we make two recommendations for statutory law reform. We conclude that, although some digital assets are not easy to place within traditional categories of things to which personal property rights can relate, this does not prevent them from being capable of attracting personal property rights, and that this is clearly the position at common law. Nonetheless, some consultees, including senior and specialist judges, said to us that it would be helpful to express this position in legislation. We recommend such legislation and conclude that it will confirm and support the existing common law position.

• 2.7    In addition, we conclude that there is one area where the common law cannot give market participants sufficient legal certainty: the development of a new regime for collateral arrangements involving crypto-tokens and cryptoassets. We acknowledge that this issue does not merely involve legal questions; it also involves policy-based judgments beyond the scope of our report. We recommend that, as a matter of priority, the Government sets up a multi-disciplinary project to formulate and put in place a bespoke statutory legal framework that better and more clearly facilitates the entering into, operation and enforcement of (certain) crypto-token and (certain) cryptoasset collateral arrangements.

Support from industry-specific technical experts

• 2.8    Third, we acknowledge that increasingly advanced technology is likely to lead to a proliferation of digital assets over time, in terms of number, use-case, design and technological functionality. Many of those digital assets are likely to be complex, composable,³² multi-faceted and use different technology, which in turn will give rise to diverse products and services that the law will have to accommodate. We conclude that common law development is better able to keep up with this change than statutory law reform. However, it is an enormous task for the judiciary to remain alive to such technological development. We recommend that the Government creates or nominates a panel of industry-specific technical experts, ³³legal practitioners, academics and judges to provide non-binding guidance on the complex and evolving factual and legal issues relating to control involving third category things such as digital objects (and other issues relating to digital asset systems and markets more broadly). We conclude that such detailed and technology specific guidance will facilitate clear, logical and consistent applications of legal rules and reasoning over time.

THE PRINCIPLES UNDERLYING OUR APPROACH

• 2.9    In this section, we set out the principles and considerations behind our recommendations and conclusions as follows.

• (1)    Championing and supporting the inherent flexibility of the common law and making clear that, in general, it is sufficiently flexible, and already able, to accommodate digital assets.

• (2)   Statutory reform only to confirm the existing common law position or where the

common law cannot develop the legal certainty the market requires.

• (3)    Applying technology-neutral principles in a way that is responsive to specific technology.

• (4)    The need for more input from experts and market participants on technical issues.

• (5)    Consistency with other legal and regulatory regimes where possible.

The common law of England and Wales is sufficiently resilient and flexible for digital assets

• 2.10    We conclude that the common law of England and Wales has already proven itself sufficiently resilient in the face of new technology and flexible enough to answer legal questions concerning digital assets. It is already developing a sophisticated legal regime that recognises and protects the nuanced features of those digital assets, providing the market with a good balance of certainty and flexibility. The jurisdiction of England and Wales is well placed to provide a coherent, globally relevant legal regime for existing and new types of digital asset.

• 2.11    This conclusion is based on:

• (1)    the research we undertook for our call for evidence, our consultation paper and this report;

• (2)   our consultation and discussions with a wide range of market participants;³⁴

• (3)   our discussions with members of the judiciary of England and Wales at an open

roundtable and our subsequent discussions with senior and specialist judges;

• (4)    our ongoing participation in other domestic and international law reform initiatives; ³⁵and

• (5)    the highly sophisticated, comprehensive and persuasive responses to our consultation paper.³⁶

The interaction between common law development and statutory intervention

• 2.12    Many consultees said that the obvious benefit of leaving the law of personal property to develop incrementally through the common law is the flexibility to deal with rapidly developing technology giving rise to different factual scenarios. The courts of England and Wales have already recognised that new forms of digital assets are capable of being things to which personal property rights can relate at law and have accommodated those assets within the law of personal property without the need for statutory intervention. Many consultees also said that statutory intervention at this stage could undermine the high level of existing legal certainty, lead to undue complexity and create a significant risk of boundary issues. That is, statute might not be capable of distinguishing between different implementations of similar technology in the way the common law can.

• 2.13    We agree. We conclude that it is not necessary, appropriate or helpful for the law of England and Wales to adopt statutory definitions of digital things for the purposes of answering the question as to whether such things are capable of being objects of personal property rights. ³⁷We think that this logic applies equally to defining hard boundaries of a category of thing to which personal property rights can relate, distinct from things in possession and things in action.³⁸

• 2.14    The flexibility of the common law is one of the greatest strengths of the law of England and Wales, as the UKJT makes clear:³⁹

English law (like other common law systems) does not necessarily require statutory intervention in order to support new asset classes or financial structures. As we have seen recently with cryptoassets, and as has been demonstrated over past decades and centuries with numerous once-novel asset classes, the common law has inherent flexibility that allows it to adapt to accommodate commercial need.

• 2.15    Formulating and implementing legislation can take a long time. This process itself could have the effect of preventing or stunting common law development, or delay experimentation by market participants while they anticipate new legal rules. When we considered recommending statutory law reform, we also considered the negative impacts of a legislative process that could take many years.

• 2.16    For those reasons, we conclude that statutory law reform in this area is only helpful where it either confirms and supports the existing common law position or where it is targeted at problems that cannot be solved more effectively by common law development. Our recommendations for statutory law reform therefore aim either to confirm the existing common law position so that it can continue to develop or to provide market participants with legal tools that do not yet exist in England and Wales, but which already exist in other jurisdictions.

Legislation to confirm and support the existing common law position so that the common law can continue to develop

• 2.17    Our first recommendation for statutory intervention seeks merely to confirm and support what we consider to be the existing position at law. That is, that being neither a thing in possession nor a thing in action does not prevent a digital asset from being capable of being a thing to which personal property rights can relate. We discuss our detailed reasoning for this in Chapter 3 (Third thing), but in summary, the use of confirmatory legislation will constitute a clear and unequivocal statement from Parliament on the point which will cement legal certainty under the law of England and Wales. It will also indicate to the judiciary that Parliament has decided that the development of parameters which describe when a digital asset is capable of being an object of personal property rights (and if that digital asset does in fact attract personal property rights) is a matter for the common law. Such a statutory confirmation will allow court time that might otherwise be spent on questions of categorisation of things to which personal property rights can relate to be used instead to focus on substantive issues. It will provide a strong platform from which common law development can proceed.

Focused, technical statutory intervention

• 2.18    In Chapter 8 (Collateral arrangements), we conclude that it is not possible for the common law alone to develop a legal framework that best facilitates the entry into, operation and enforcement of collateral arrangements for crypto-tokens and cryptoassets. We conclude that such a regime would be beneficial for the law of England and Wales and would provide market participants with important legal tools that do not exist today.

• 2.19    Designing such a regime is a highly technical, and market-specific process which has complex policy implications that require a wide-ranging cross-functional investigation and rigorous cost benefit analysis. We recommend that, as a matter of priority, the Government sets up a multi-disciplinary project to formulate and put in place a bespoke statutory legal framework that better and more clearly facilitates the entering into, operation and enforcement of (certain) crypto-token and (certain) cryptoasset collateral arrangements.

Applying technology-neutral principles in a way that is responsive to specific technology

• 2.20    While our work seeks to support and facilitate the use and development of digital assets (including crypto-tokens and cryptoassets), it remains neutral as to the advantages and disadvantages of any single digital asset, crypto-token, protocol, system, network or technological feature. However, our work is not in its entirety “technology neutral”, because it necessarily discusses existing technology used in relation to digital assets and crypto-tokens, most specifically cryptography. This is necessary because different technological arrangements give rise to different factpatterns, with which the law must deal.

• 2.21    We conclude that the flexibility of the common law allows it to apply legal principles in a technology neutral way (that is, without favouring any one example of technology), while recognising and accommodating the distinct functionality of specific technologies.

• 2.22    Consultees agreed that this was a principal advantage of the common law. Many consultees said that, to analyse the proper application of the law, each digital asset needs to be considered on a case-by-case basis by reference to the particular features of the digital asset in question and, where applicable, the system which manifests it.

• 2.23    Each system might be constructed, or have developed, in different ways and the functional characteristics of that system will also differ depending on the ways in which different networks of participants participate in that system. ⁴⁰The law must retain the flexibility to accommodate and respond to those technological features, particularly where they constitute material facts of any given case, and therefore affect the legal principles that need to be applied. This is particularly the case where distinct objects of personal property rights are recognised at law as existing by virtue of the active operation of particular software by a network of participants.⁴¹

• 2.24    One of the principal conclusions in this report is that recognising the idiosyncrasies of distinct digital assets and digital asset systems will allow the law of England and Wales to develop legal principles and rules best suited to those particular digital assets. It can do so without being fettered by the wholesale application of imperfect analogies with other things with which the law has long been familiar.

• 2.25    The distinctive advantage of the common law in this context is already apparent at the Court of Appeal level. In Tulip Trading v Van der Laan (“Tulip Trading”), Lord Justice

Birss was careful to acknowledge the differences between different forks of the original Bitcoin network, each of which manifest different crypto-tokens:⁴²

There are four bitcoin networks in issue in this case: BSV, BTC, BCH and BCH ABC... Each later network started life as a copy of the blockchain of a pre-existing network (which is after all public) but by then applying different software thereafter.

• 2.26    A similar approach is also evident in proposed UK regulation. HM Treasury recently closed a consultation and call for evidence on a future financial services regulatory regime for cryptoassets.⁴³ In that consultation, HM Treasury said that one of its core principles was recognising where technology either gives rise to, or mitigates, specific risks that might not be found (or might take a different form) in markets that do not rely on such technology.⁴⁴

• 2.27    We decided on our approach, in part, to align with the approach taken by HM Treasury to regulatory law reform in this area: that maintaining the flexibility and agility of the law of England and Wales is important. HM Treasury said to us:

As part of the Financial Services and Markets Bill, the UK is clarifying powers to bring stablecoin and cryptoasset activities into the existing financial services framework in an agile way via secondary legislation. Detailed firm-facing rules will be delegated to the UK’s regulators.

This means that the UK will be able to update regulation in an agile way to support innovation and address risks as the sector rapidly evolves, rather than hard-coding detailed rules into primary legislation.

• 2.28    We acknowledge the argument that legislation to change the operation of private law has the potential to provide greater legal clarity and predictability for market participants than common law principles. However, legislation is also more likely than the common law to create boundary issues in its application. Legislation might lack the agility required to reconcile the application of technology neutral principles with specific technological facts. The common law concerning digital assets has developed well to date. We conclude that there is significant risk that legislative intervention in this area (other than the targeted, confirmatory legislation we recommend)⁴⁵ could undermine existing legal certainty and could reopen settled points of law.

Greater input from specialist market participants on technical issues

• 2.29    The digital asset market is still evolving rapidly and the technology in question will evolve further. Both the technology and market practice are highly responsive to the ongoing social utility of particular products and use-cases. Put simply, the digital assets market — or at least elements of it — remains unpredictable.⁴⁶ This means that digital assets and the markets that involve those assets are currently more susceptible to change than other assets and markets. If the law is to respond to and reflect those developments, it must draw on the expertise of those who are most familiar with the technology.

• 2.30    The UK has to date been successful in both canvassing and listening to views from a range of market participants on private and regulatory law reform in respect of digital assets and has relied heavily on such input for legal and policy analysis and development. ⁴⁷We conclude that the jurisdiction of England and Wales should continue to allow and encourage market participants and the industry to contribute to the process of legal development. This will help to ensure that the law accurately reflects the underlying technology and the ways in which it is used in practice.

Consistency with other legal and regulatory regimes

• 2.31    The law of England and Wales has a global reach as the legal system of choice for many commercial parties. ⁴⁸Our recommendations and conclusions aim to ensure that the law of England and Wales remains a dynamic, globally competitive and flexible tool for market participants in the digital asset space. Reducing or removing residual areas of legal uncertainty should help market participants better to understand and quantify the risks involved in their activities. The Government has said that a similar approach also is likely to be reflected at the regulatory level, focusing on growth, innovation, competition and risk disclosures.⁴⁹

Consistency

• 2.32    Given that digital asset markets facilitate participation at a global level and reduce geographic or jurisdictional frictions, our recommendations and conclusions aim to be consistent with other international law reform initiatives where possible. Most specifically, we intend that our recommendations and conclusions are broadly consistent with the recommendations of the UNIDROIT Working Group on Digital Assets (“UNIDROIT Working Group”) which recently published a set of international principles designed to facilitate transactions in digital assets.⁵⁰ However, while intended to be consistent as a matter of practical effect, our recommendations do not take the same form as the UNIDROIT Working Group principles.⁵¹

• 2.33    Our report also considers the work of the American Law Institute and the Uniform Law Commission’s Uniform Commercial Code and Emerging Technologies Committee (“UCC Committee”) in the United States. The UCC Committee recommended changes to the United States Uniform Commercial Code. ⁵²The recommendations include a new provision that, where implemented, would govern the transfer of personal property rights in relation to certain intangible digital assets that have been or may be created using new technologies.⁵³ Again, we intend that our recommendations are broadly consistent with the UCC Committee’s recommendations as a matter of practical effect, albeit not as a matter of form. In addition, our recommendations aim to provide those using the law of England and Wales with similar legal tools to those available in states that implement the UCC Committee’s recommendations.

Differentiation

• 2.34    We conclude that it is important that the law of England and Wales provides certainty within a clear, flexible legal environment for internationally mobile transactions and markets involving digital assets. In its report, Economic value of English law, Oxera said:⁵⁴

[Internationally mobile transactions and markets] have a tendency to ‘tip’, which means that once the use of a governing law reaches a critical mass in a given market, most market participants tend to use that law because of the benefit from using standardised contract and procedures.

• 2.35    We agree with Oxera that promoting the use of the law of England and Wales in fintech fits particularly well with the Government’s overall strategy to attract more technological innovation in the financial services sector and to “cement the UK’s position as the world’s pre-eminent financial centre”.⁵⁵

• 2.36    More recently, the Government announced its intentions “to make the UK a global hub for cryptoasset technology”, such that the “UK financial services industry is always at the forefront of technology and innovation.”⁵⁶

• 2.37    We also note that today no clear and consistent private legal and regulatory environment exists within which digital asset market participants can innovate, experiment and grow.

• 2.38    In the United States, the legal environment for digital assets has developed by a combination of sometimes contradictory, state-by-state statutory reform, ⁵⁷common law precedent and policy-led regulatory enforcement decisions and settlement negotiations.⁵⁸

• 2.39    The legal environment in certain jurisdictions including Hong Kong, Singapore, China and Japan has sometimes experienced sudden or unexpected reversals in direction, attitude, or policy with regard to digital assets.⁵⁹

• 2.40    The legal environment in other European jurisdictions — many of which are civil lawbased — is largely characterised by complex, comprehensive codification of rules relating to digital assets.⁶⁰ This approach has been continued in the regulatory sphere by the introduction in the European Union of the Markets in Crypto-assets Regulation (“MiCA”), which creates a complex regulatory regime specifically for digital assets.⁶¹

property rights under Hong Kong law and is capable of being held on trust: Re Gatecoin Limited [2023] HKCFI 914.

China, which in 2021 introduced widespread bans on crypto-token trading is reportedly considering the option of softening this strict stance. See Library of Congress, “China: Central Bank Issues New Regulatory Document on Cryptocurrency Trading” (2021); People’s Bank of China, “Circular on Further Preventing and Disposing of Speculative Risks in Virtual Currency Trading” (2021), available at: https://www.gov.cn/zhengce/zhengceku/2021-10/08/content_5641404.htm; and A Cuthbertson, “Bitcoin price poised for ‘orthogonal’ shakeup as China softens stance” (2023) Independent, available at: https://www.independent.co.uk/tech/bitcoin-price-latest-china-crypto-b2288046.html.

Singapore has recently put forward plans to reduce and mediate retail access to crypto-token markets despite the jurisdiction’s earlier ambitions to facilitate a more hands-off - albeit cautionary - approach to crypto-token regulation. See Monetary Authority of Singapore, “Proposed Regulatory Measures for Digital Payment Token Services” (2022); and Monetary Authority of Singapore, “Keynote Speech by Mr Ong Chong Tee, Deputy Managing Director (Financial Supervision), Monetary Authority of Singapore, at Asia Securities Industry & Financial Markets Association (ASIFMA) Annual Conference 2018 on 1 November 2018” (2018).

In Japan in 2015, the Tokyo District Court held that bitcoin was not a thing capable of ownership within the Japanese Civil Code. Since that judgment, an amendment to the Japanese Payment Services Act added the concept of “Virtual Currency”. Lee and Van de Looverbosch suggest that this intervention has brought crypto-tokens within the sphere of property law, as a form of “property value”: J Lee and M Van de Looverbosch, Property and Data: A Confused Relationship (2021). See also J Lee and A Darbellay, A Data Governance in AI, FinTech and RegTech: Law and Regulation in the Financial Sector (2022), available at: https://ssrn.com/abstract=3995492 and paras 4.58-4.61 of our consultation paper.

• ⁶⁰    In 2019, France enacted a comprehensive legislative framework for digital assets entitled “Plan d’Action pour la Croissance et la Transformation des Entreprises”, available at: https://www.gouvernement.fr/action/pacte-le-plan-d-action-pour-la-croissance-et-la-transformation-des-entreprises.

In 2020, Switzerland introduced the Federal Act on the Adaptation of Federal Law to Developments in Distributed Ledger Technology, which broadly amended a number of existing statutory regimes to better incorporate digital assets, available at: https://www.newsd.admin.ch/newsd/message/attachments/60601.pdf.

In 2020, Liechtenstein introduced the Token and Trusted Technology Service Provider Act 2019-301, which provides for the tokenisation of assets and rights. This legislation creates a new legal object - a token - and a specific, separate regime for the regulation and use of those tokens.

In 2020, Germany introduced the Gesetz zur Umsetzung der Anderungsrichtlinie zur Vierten EU-Geldwascherichtlinie which, among other things, introduced a new category of “crypto assets” into the definition of financial instruments for the purposes of German banking legislation. See also para 14.46 of our consultation paper.

In 2018, Malta introduced the Virtual Financial Assets Act which, alongside the Innovative Technology Arrangements and Services Act (2018) and the Malta Digital Innovation Authority Act (2018), comprehensively legislates on the issuance and use of virtual financial assets.

• 2.41    While codification projects in those jurisdictions have provided market participants with a high degree of legal certainty, and have been largely well received, we remain of the view that the UK’s approach provides an excellent balance of flexibility and legal certainty.

• 2.42    The UK — and the law of England and Wales — therefore has an opportunity to provide a clear and consistent private legal and regulatory environment for digital assets and digital asset related activities. That legal and regulatory environment might in turn help to attract more technological innovation in the financial services sector.

• 2.43    Our recommendations and conclusions in this report aim to help achieve this goal, although they are explicitly limited to private law issues.⁶²

SUMMARY OF OUR RECOMMENDATIONS AND CONCLUSIONS

• 2.44    This report comprises seven further chapters and two appendices. We summarise our recommendations and conclusions in each chapter below.

Chapter 3: A “third” category of thing to which personal property rights can relate

• 2.45    In Chapter 3 (Third thing), we conclude that some digital assets are neither things in possession nor things in action, but that nonetheless the law of England and Wales treats them as capable of being things to which personal property rights can relate. Some consultees, including senior and specialist judges, said that it would be helpful to express this position in legislation. They said that this would confirm the existing position at common law, facilitate the law’s continued development on the point and lay to rest any lingering authority suggesting that there can be no “third” category of this nature. We recommend such legislation and conclude that it will confirm and support the existing common law position.

• 2.46    Although we recommend legislation to confirm the common law position, we conclude that it is not necessary or appropriate to define in statute the hard boundaries of this third category of things. We conclude that the common law is the better vehicle for determining those things that properly can (and should) be objects of personal property rights and which fall within the third category: third category things. These might not necessarily be digital things. ⁶³We refer to digital things within the third category as digital objects.

Detailed overview of Chapter 3 (Third thing)⁶⁴

• 2.47    The law of England and Wales traditionally recognised two distinct categories of personal property rights: rights relating to things in possession (tangible things) and rights relating to things in action (legal rights or claims enforceable by action).⁶⁵

• 2.48    The legal concept of possession is traditionally limited to tangible things.⁶⁶ We recognise that it is therefore bound up with tangibility and has traditionally relied on the physical boundaries of a thing to help define the contours of legal duties in relation to that thing. ⁶⁷Digital assets are not tangible in the traditional sense. ⁶⁸We conclude that, given their vastly different functionality to tangible things, it is not necessary or appropriate for digital assets to be categorised as things in possession.

• 2.49    There exist certain digital assets that cannot be things in action in the narrow sense.⁶⁹ Indeed, some crypto-tokens and other intangible objects which the law recognises can be things to which personal property rights can relate are explicitly designed not to consist of a legal right or claim against a legal person. They can be, and are designed to be, used and enjoyed regardless of their legal recognition or enforceability. ⁷⁰We conclude that it is not necessary, conceptually coherent or helpful for such things to be categorised as things in action, or for things in action to be treated as a wider, residual category of things that captures any object of personal property rights that is not a thing in possession.

• 2.50    This conclusion reflects the current state of the common law, which over the last ten years has clearly moved toward the explicit recognition of a third category of thing to which personal property rights can relate (and which includes many digital assets). The law has now moved on from the (arguably hollow) ⁷¹debate on categorisation of digital assets to answering substantive questions as to the legal principles applicable to those digital assets. It is entirely appropriate for the common law of England and Wales to take this approach. Drawing analogies between (1) conventionally tangible things and digital assets; and (2) things in action and digital assets is helpful to a point but those analogies ultimately break down. This is particularly true in respect of those digital assets that rely on novel and idiosyncratic technology, such as publicly available open-source code, distributed consensus mechanisms and public key cryptography.

• 2.51    We conclude that it is not necessary or helpful for legislation to define the boundaries of a third category of thing to which personal property rights can relate. Whether or not a thing falls within the third category (and whether it is capable of attracting personal property rights under the law of England and Wales) is a complex and dynamic question, which is ill-suited to static definition in statute. As many consultees said, defining the parameters of digital assets, including identifying exactly what the “asset” consists of, can be extremely challenging and the answer will be different for different types of assets. Once the asset or thing is identified, the law can determine whether that thing can (and should) attract personal property rights. Specifying certain digital assets as capable of attracting personal property rights (or, alternatively, as incapable of attracting personal property rights) in statute simultaneously risks drawing parameters that are too narrow or too wide, and also risks opening up difficult definitional issues that could reduce, rather than increase, certainty in the law.

• 2.52    The third category of things includes a certain type of asset — namely crypto-tokens — that clearly satisfy the criteria we described in Chapter 5 of our consultation paper (see paragraph 4.5 below). Our proposed criteria accurately describe a certain “core” type of asset — namely crypto-tokens manifested by distributed, public, permissionless systems — that are things to which personal property rights can relate at law and which are neither things in possession nor things in action. But we conclude that it is not appropriate for those criteria to delimit absolutely the contours of such a category. Instead, it is better that the common law continues to (1) identify and describe exactly what the thing in question consists of; ⁷²and (2) consider whether that thing can (and should) be capable of being an object of personal property rights. This will allow the law of England and Wales to continue its nuanced approach to recognising that things such as milk quotas, export quotas, waste management licences and different types of carbon emissions allowances ⁷³are capable of being objects of personal property rights. As such, we now treat our criteria (as modified by this report) as indicia.

• 2.53    We recommend legislation confirming that the fact that a thing is neither a thing in possession nor a thing in action does not prevent it from being a thing to which personal property rights can relate. ⁷⁴Such things include crypto-tokens like bitcoin or ether which are not rights or claims in themselves and which can be used and enjoyed independently of whether any rights or claims in relation to them are enforceable by action. Over the last ten years the common law has clearly moved toward the explicit recognition of this position. However, some consultees, including senior and specialist judges, said to us that it would be helpful to express this position in legislation.⁷⁵ Although it would not change the common law position, ⁷⁶we conclude that such a statutory confirmation will provide greater legal certainty and will allow the law to develop from a strong and clear conceptual foundation. A statutory confirmation of the position might reduce the time spent by the courts on questions of categorisation of objects of personal property rights, and instead allow them to focus on the substantive issues before them. It gives explicit effect to:⁷⁷

The powerful case for reconsidering the dichotomy between [things] in possession and [things] in action and recognising a third category of intangible property... in a way that would take account of recent technological developments.

• 2.54    A statutory confirmation will also alleviate any concern of judges that recognising a third category is not an appropriate development for the common law to make.⁷⁸ In short, this recommendation lays to rest any lingering authority of Colonial Bank v

Whinney ⁷⁹which suggested that there could be no third category thing and confines to history the “red herring” ⁸⁰of a circular and hollow debate on categorisation.

• Chapter 4:    Our third category recommendation and conclusions in practice

• 2.55    In Chapter 4 (Third thing in practice), we consider in detail consultees’ concerns with defining hard boundaries for a third category of thing to which personal property rights can relate. Given that our recommendation in Chapter 3 (Third thing) amounts to a confirmation and restatement of the existing common law position that such a third category exists, we do not consider that it will cause any additional legal uncertainty. We demonstrate this by reference to a variety of digital assets, including cryptotokens, private, permissioned blockchain systems, voluntary carbon credits (“VCCs”), in-game digital assets and digital files. We also discuss consultees’ responses to the provisional criteria we proposed in our consultation paper for the third category. We make consequential modifications and clarifications to those criteria.

• 2.56    We do not make any recommendations for law reform in Chapter 4 (Third thing in practice).

Detailed overview of Chapter 4 (Third thing in practice)⁸¹

• 2.57    We discuss consultee responses to the provisional proposal in our consultation paper that a thing should be capable of falling within our proposed third category of things to which personal property rights can relate if:

• (1)    it is composed of data represented in an electronic medium, including in the form of computer code, electronic, digital or analogue signals;

• (2)    it exists independently of persons and exists independently of the legal system; and

• (3)    it is rivalrous.⁸²

• 2.58    Based on consultee responses, we conclude that “composed of data” need not be a criterion in itself, because the criterion: (1) overly focuses the conceptualisation of the thing in question on data; and (2) potentially creates an unnecessary hard boundary for the third category.

• 2.59    We clarify the application and interpretation of our second criterion — that a thing must exist independently of persons and exist independently of the legal system — and respond to some concerns raised by consultees about this criterion.

• 2.60    We re-iterate and reconfirm our analysis of the criterion that a thing must be rivalrous. Specifically, we clarify that whether a thing is rivalrous is binary and we distinguish our criterion that a thing must be rivalrous from the concepts of exclusivity of control and excludability.

• 2.61    We conclude that our criteria (as modified in this report and which we now treat as indicia) accurately describe a certain “core” type of asset that are things to which personal property rights can relate at law and which are neither things in possession nor things in action. The most obvious example of this type of digital asset is a cryptotoken, so we discuss crypto-tokens by way of illustrative example. ⁸³Our indicia — specifically, the concept that a thing must be “rivalrous” as endorsed by the Court of Appeal in Tulip Trading ⁸⁴— usefully distinguish this type of digital asset from other digital things such as digital files ⁸⁵that are not (as currently designed) capable of attracting personal property rights as a matter of law.

• 2.62    We consider consultees’ concerns about boundary issues in more detail. We discuss those concerns by reference to private, permissioned blockchain systems, voluntary carbon credits (“VCCs”), in-game digital assets and digital files by way of example. We conclude that pre-existing boundary issues will remain and that those boundary issues cannot be solved by (and indeed, would likely be exacerbated by) statutory law reform. We conclude that the common law is the most appropriate tool for dealing with difficult boundary issues relating to digital assets that are based on varied technologies and for determining whether such digital assets can (and should) attract personal property rights on particular sets of facts.

• Chapter 5:    Control

• 2.63    In Chapter 5 (Control), we describe (but do not define) the factual concept that best captures the ability to (1) exclude or to permit access to a third category thing; and (2) put the third category thing to the uses of which it is capable. We call this factual concept “control”. We go on to discuss the legal significance of the concept of control over third category things. We conclude that both the factual concept of control and the legal consequences of control work differently for, and are highly complex in relation to, third category things — particularly digital objects.

• 2.64    We recommend that the Government creates or nominates a panel of industry-specific technical experts, legal practitioners, academics and judges to provide non-binding guidance on the complex and evolving factual and legal issues relating to control involving third category things such as digital objects (and other issues relating to digital asset systems and markets more broadly).

Detailed overview of Chapter 5 (Control)⁸⁶

• 2.65    We consider two related but distinct ways in which jurisprudence concerning a concept of control might develop under the law of England and Wales.

• 2.66    First, we conclude that common law jurisprudence will be enhanced and made easier to understand for market participants by focusing on better descriptions and real-world examples of factual control. Factual control in this context is a highly technology specific concept, in large part determined by the way in which the particular technology in question facilitates the imposition or creation of varying degrees of technical encumbrances in respect of the digital object in question. For example, in many decentralised systems, differing levels and types of control will be distributed across different participants. ⁸⁷Both the High Court and the Court of Appeal recognised this point in Tulip Trading.⁸⁸

• 2.67    The broad concept of factual control might require significant refinement or malleability if it is properly to encompass the variety of possible relationships involving third category things including digital objects. We do not use the concept of factual control as a definitional characteristic of third category things.⁸⁹ Nor do we recommend that the concept of factual control should be defined in legislation. ⁹⁰Instead, we acknowledge that the common law has already begun to develop jurisprudence concerning the concept of control as it applies to third category things and conclude that it should continue to do so.

• 2.68    To assist with this process, we recommend that the Government creates or nominates a panel of industry-specific technical experts, ⁹¹legal practitioners, academics and judges to provide non-binding guidance on the complex and evolving factual and legal issues relating to control involving third category things such as digital objects (and other issues relating to digital asset systems and markets more broadly).⁹² We envisage that the technical expert group would provide practical, considered and technology-specific guidance on a (semi-regular) and iterative basis over time. As technology changes and methods and mechanisms for control change, the technical expert group can update its guidance. Gradually, it will build up a body of useful factual examples without having to wait for or rely on specific factual scenarios being raised or disputed in court. We conclude that this approach strikes the most appropriate balance between creating legal certainty for the market and maintaining the dynamism and flexibility that characterises the law of England and Wales.

• 2.69    Second, the legal consequences of control are necessarily complex and varied. We do not think that the concept of control alone is sufficiently nuanced, refined, or sensitive to market specificities to determine adequately and definitively the consequences of complex legal arrangements. Instead, we see control as a composite part of more complex legal principles and mechanisms (such as legal transfers, intermediated holding arrangements, collateral arrangements and actions and remedies in respect of third category things including digital objects). There are also a vast number of technically distinct digital assets, some of which function more like “digital bearer instruments” ⁹³and some of which do not. Control works differently for those different digital assets, by virtue of the inherent features and functions of the technology itself. The application of control and its legal consequences will therefore be different for those different digital assets. Specifically, control works differently for things in possession, things in action and third category things (and, potentially, differently for different third category things). We conclude that the law should recognise and accept this reality. We discuss consultees’ differences in views on this point, which were most acute in relation to whether concepts of relativity of title could apply to third category things. We do not make any recommendations for law reform but do conclude on what we consider to be the best way for the common law to develop in this area.

• 2.70    The work of the technical expert group will help to enhance common law jurisprudence by making it easier for courts to make decisions that are sensitive and responsive to specific technology. This will be particularly important for navigating the difficult boundary issues between (pre-existing) rights-based systems and emerging and novel systems which give rise to “bearer-like” objects that can attract personal property rights, and in cases where the two overlap or are linked.⁹⁴ The corollary of not defining the hard contours of a third category of thing to which personal property rights can relate is that the boundaries are less clear. However, we think that clarity can instead be achieved by applying general legal principles to the specific factual circumstances that arise as a result of specific technology.⁹⁵

• Chapter 6:    Transfers

• 2.71    In Chapter 6 (Transfers), we consider how legal transfers of crypto-tokens operate based on, among other things, how a crypto-token transfers as a matter of fact, and the different perspectives of consultees on this issue. While some third category objects things will transfer in similar ways to crypto-tokens, others will not. As we conclude elsewhere in this report, the law will need to recognise the idiosyncratic technological attributes of the third category object thing in question when considering legal issues relating to that third category thing. Because of that, and because cryptotoken systems raise some specific questions as to transfers that are currently relevant to market participants, we generally discuss crypto-tokens (and not third category things) in Chapter 6 (Transfers).

• 2.72    We conclude that it is possible to effect a legal transfer of a crypto-token offchain, by a “change of control” (along with the requisite intention). An example might include the physical transfer of control through the transfer of hardware, or a transfer on a Layer 2 system.

• 2.73    We also conclude that a special defence of good faith purchaser for value without notice applicable to crypto-tokens can be recognised and developed by the courts through incremental development of the common law. We conclude that this reasoning can also be extended to other third category things.

• 2.74    We do not make any recommendations for law reform in Chapter 6 but do conclude on what we consider to be the best way for the common law to develop in this area.

Detailed overview of Chapter 6 (Transfers)⁹⁶

• 2.75    In Chapter 6 (Transfers), we do three things, by close reference to consultee responses.

• 2.76    First, we discuss our observation in our consultation paper that “a transfer operation within a crypto-token system will typically involve the replacing, modifying, destroying, cancelling, or eliminating of a pre-transfer crypto-token and the resulting and corresponding causal creation of a new, modified or causally-related crypto-token.”⁹⁷ We discuss two opposing views put forward by consultees as to the legal characterisation of such a transfer operation that effects a state change. First, that such a transfer extinguishes a pre-transfer object of personal property rights and creates a “new”, post-transfer object of personal property rights (the “extinction/creation analysis”). Second, that such a transfer involves the persistence of an object of personal property rights through the transfer (the “persistent thing analysis”). We conclude that there is no single correct answer. However, consultees offered a number of reasons as to why the most practically accurate analysis is that a notional quantity unit (or the ability/power in respect of the notional quantity unit) can persist through a transaction.

• 2.77    Second, we consider the view of the majority of consultees that legal transfers of a crypto-token can, in certain circumstances, be effected by a “change of control” and should not be limited to a transfer operation that effects a state change. We agree with this view.

• 2.78    Third, we acknowledge that our recommendation for targeted, confirmatory legislation combined with common law development of the parameters of a third category of thing to which personal property rights can relate excludes the possibility of a statutory definition of such third category things (or some subset thereof). We acknowledge that this likely prevents a general statutory “innocent acquisition rule” in respect of such objects of personal property rights, because a statutory innocent acquisition rule would almost certainly need to define the objects of personal property rights in question that benefitted from the rule. We also recognise that the majority of consultees made strong arguments in favour of the recognition and development of a common law special defence of good faith purchaser for value without notice applicable to crypto-tokens (and third category things more broadly). We agree with the arguments made by consultees. We conclude that a special defence of good faith purchaser for value without notice applicable to crypto-tokens can be recognised and developed by the courts through incremental development of the common law. We conclude that this reasoning can also be extended to other third category things.

• Chapter 7:    Intermediated holding arrangements

• 2.79    In Chapter 7 (Intermediated holding arrangements), we consider how intermediated holding arrangements in respect of crypto-tokens can be structured under the law of England and Wales. We consider crypto-tokens by way of example given the importance of intermediated holding arrangements to crypto-token markets. We compare different types of intermediated holding arrangements and the potential legal consequences of such arrangements.

• 2.80    We do not make any recommendations for law reform in Chapter 7 (Intermediated holding arrangements) but do conclude on what we consider to be the best way for the common law to develop in this area.

Detailed overview of Chapter 7 (Intermediated holding arrangements)⁹⁸

• 2.81    First, we discuss consultee responses to the terminology in our consultation paper that we used to describe crypto-token specific intermediated holding arrangements, particularly our use of the term “custody”. In light of consultee responses, we now draw a distinction between “custodial intermediated holding arrangements”, “noncustodial intermediated holding arrangements” and “non-holding arrangements” based on the legal consequences of such arrangements. In particular, we highlight the risks that users of intermediated holding arrangements could be exposed to on the onset of insolvency proceedings of a holding intermediary.

• 2.82    Second, we consider the application of contract and trust law to crypto-token intermediated holding arrangements. We conclude that trusts can support a broad range of custodial intermediated holding arrangements, including where the underlying crypto-token entitlements are held on a consolidated unallocated basis for the benefit of multiple users. We confirm our preferred conceptual approach to the establishment of a such a trust arrangement under the law of England and Wales. We conclude that a presumption of trust for intermediated holding arrangements involving crypto-tokens is not necessary or appropriate.

• 2.83    We consider the potential impact of statutory formalities on the operation of trustbased crypto-token intermediated holding arrangements. We conclude that the existing common law is sufficiently certain in this area and that statutory law reform in respect of section 53(1)(c) of the Law of Property Act 1925 is not necessary at this time. We leave open the possibility that it might be necessary or warranted in future as the market evolves.

• 2.84    We consider but do not recommend a general pro rata shortfall allocation rule in respect of commingled unallocated holdings of crypto-tokens or crypto-token entitlements held on trust by a holding intermediary that enters insolvency proceedings. Instead, we conclude that a more extensive, in-depth assessment of the merits of potential insolvency law reform applicable to specific custodial holding intermediaries is more appropriate.

• 2.85    We discuss the possibility of the common law developing alternative and supplementary legal structures for custodial intermediated holding arrangements that do not rely on trusts. We conclude that this could take the form of holding intermediaries being recognised as acquiring a limited control-based proprietary interest in held crypto-token entitlements that is subject to a superior title retained by users, but do not make any recommendations to this effect. We also discuss the application of other private law principles including agency and fiduciary duties.

• Chapter 8:    Collateral arrangements

• 2.86    In Chapter 8 (Collateral arrangements), we consider how collateral arrangements in respect of crypto-tokens and cryptoassets can be structured under the law of England and Wales. Again, we specifically consider crypto-tokens and cryptoassets given their prominence in the digital asset markets.

• 2.87    We make two recommendations in Chapter 8 (Collateral arrangements), which we describe in more detail below.

Detailed overview of Chapter 8 (Collateral arrangements)⁹⁹

• 2.88    First, we discuss how title transfer and non-possessory security-based arrangements can be used to structure crypto-token and cryptoasset collateral arrangements without the need for law reform. Second, we explain that possessory security-based arrangements do not apply to crypto-tokens and cryptoassets.

• 2.89    We then discuss how the recognition of a control-based proprietary interest to facilitate both the holding of and the grant of security over crypto-tokens and cryptoassets might be a beneficial development within the common law. We conclude that the common law could develop to recognise a control-based security interest in respect of crypto-tokens and cryptoassets (possibly by analogy with pledge).

However, the development of such a security interest would likely not be a complete solution given that such a security interest would likely be reliant on static, comprehensive notions of control.

• 2.90    We then consider the applicability of the Financial Collateral Arrangements (No 2) Regulations 2003 (“FCARs”) to crypto-tokens and other collateral that might use and/or be linked to public, permissionless crypto-token systems or private, permissioned blockchain systems (including Central Bank Digital Currencies (“CBDCs”), stablecoins, equity and debt securities and credit claims). We broadly subdivide our analysis into an analysis of crypto-tokens, cryptoassets and mere record/register tokens.

• 2.91    We conclude that many crypto-tokens are likely to fall outside of the scope of the FCARs regime. However, for collateral arrangements in respect of cryptoassets (including CBDCs, stablecoins, equity and debt securities and credit claims) or mere record/register tokens we think the answer is possibly different. For at least some of those things, there is a better argument that they fall within the scope of the FCARs regime. We recommend law reform to clarify this position, although we do not ultimately conclude on what the complete scope of the FCARs regime should be, given that question necessarily involves policy considerations.

• 2.92    As part of this discussion, we discuss the tokenisation of securities. We recommend that the laws governing the tokenisation of equity and other registered corporate securities by UK companies are reviewed. The aim of this review would be to confirm, and where appropriate extend, the range of technological facilities (including potentially to public, permissionless ledgers) and operational arrangements through which the valid creation, transfer and use of such tokenised equity and other registered corporate securities would be legally possible. This would require further legislative change.

• 2.93    We conclude that, although the law of England and Wales does provide options for granting security in respect of crypto-tokens and cryptoassets, those options are not adequate. As such, we recommend that, as a matter of priority, the Government sets up a multi-disciplinary project to formulate and put in place a bespoke statutory legal framework that better and more clearly facilitates the entering into, operation and enforcement of (certain) crypto-token and (certain) cryptoasset collateral arrangements.

• 2.94    Chapter 8 (Collateral arrangements) therefore departs from much of the rest of this report in two ways. First, it concludes that there are existing problems with the law of England and Wales that cannot be solved or improved by further common law development. That is largely because we conclude that an effective collateral regime for (certain) crypto-token and (certain) cryptoasset collateral requires statutory intervention if it is to provide the level of certainty required by the market. Second, it recommends law reform, but acknowledges that identifying the optimal approach cannot solely be determined by reference to applicable private law principles or the mere technical feasibility of statutory commercial law reform (although both will be of central importance). Any such reform has complex policy implications that require a wide-ranging cross-functional investigation and rigorous cost benefit analysis. It will also require coordination with and the appropriate allocation of policy objectives between other policy initiatives impacting the operation and development of cryptotoken and cryptoasset markets and the conduct of market participants, such as changes in the regulatory environment.

• 2.95    Although this recommendation and the work required to implement it are significant, we conclude that there is a very high degree of support and demand for such law reform among consultees, markets participants and industry bodies.

• Chapter 9:    Causes of action and associated remedies

• 2.96    In Chapter 9 (Remedies), we consider causes of action and associated remedies in the context of third category things. We conclude that much of the current law concerning causes of action and remedies can be applied to third category things without law reform. Often, the law does not distinguish between causes of action and remedies that apply to things in possession, to things in action or to third category things and we identify where that is currently the case. In those cases there is no need for bespoke rules or for law reform. Instead, what is required is that the courts continue to recognise the nuances or idiosyncrasies of third category things (including their distinct functionality and technical characteristics) and apply existing legal principles to such things as appropriate.

Detailed overview of Chapter 9 (Remedies)¹⁰⁰

• 2.97    First, we consider the application of various causes of action that arise in relation to contracts, with particular focus on the legal characterisation of an obligation to “pay” non-monetary units such as crypto-tokens. We also discuss the application of various vitiating factors to contracts involving third category things. We conclude that the vitiating factors of mistake, misrepresentation, duress and undue influence apply similarly to contracts involving third category things as they do to contracts involving things in possession and things in action. We also conclude that the legal principles relating to void contracts can apply to third category things, in the same way as they do to other objects of personal property rights, without law reform.

• 2.98    Second, we consider how the evidentiary processes of following and tracing might apply to third category things and discuss how the factual nature of third category things, particularly crypto-tokens, might complicate legal analysis in relation to those evidentiary processes.

• 2.99    Third, we consider the application of principles relating to breach of trust, equitable wrongs and constructive trusts. We conclude that, as regards breach of trust and fiduciary duty, the principles of equity are sufficiently flexible to be applied in situations involving third category things. In relation to constructive trusts, we conclude that the common law is perfectly able to evolve in a logical and clear way and we do not recommend law reform.

• 2.100    Fourth, we consider three key common law causes of action and how they apply to factual scenarios involving third category things: proprietary restitution; restitution for unjust enrichment; and conversion. We conclude that claims in proprietary restitution and restitution for unjust enrichment likely will be available in the context of third category things, whereas a claim in conversion will not be available. This is because conversion only applies to things in possession.

• 2.101    However, despite the broad availability of claims in proprietary restitution and restitution for unjust enrichment, we conclude that such claims are unlikely to succeed in at least one specific factual scenario: where a claimant’s crypto-token is burned by a defendant without consent. Burning involves irreversibly sending a crypto-token to an inaccessible “burn address”, the result being that the token is removed from circulation.

• 2.102    Given the unavailability of a claim for proprietary restitution, restitution for unjust enrichment, or conversion following a defendant’s burning of a claimant’s crypto-token without consent, we conclude that there is a lacuna in the law relating specifically to certain things that fall within the third category (specifically, crypto-tokens). We do not consider that common law development of the principles of proprietary restitution or unjust enrichment would be the most appropriate means by which to fill this lacuna. Instead, we conclude it would be constructive for the courts to develop specific and discrete principles of tortious liability by analogy with, or which draw on some elements of, the tort of conversion to deal with wrongful interferences with third category things such as digital objects. However, this conclusion also acknowledges that the lacuna currently existing within the law is small and arises in situations where a claim based on unjust enrichment or proprietary restitution cannot be made out.

• 2.103    Finally, we consider some procedural aspects of the law, specifically the law relating to injunctions, enforcement, and monetary awards.

Chapter 3: A “third” category of thing to which personal property rights can relate

OVERVIEW

• 3.1    “Property” can be divided into real property (interests in land) and personal property (interests in other things). The law of England and Wales traditionally recognised two distinct categories of personal property rights: rights relating to things in possession (tangible things), and rights relating to things in action (legal rights or claims enforceable by action).¹⁰¹

• 3.2    In the 1885 case of Colonial Bank v Whinney, Lord Justice Fry said:¹⁰²

All personal things are either in possession or in action. The law knows no tertium quid ¹⁰³between the two.

• 3.3    That statement is no longer entirely correct (to the extent that it ever was). ¹⁰⁴Rather, the common law of England and Wales has over the last ten years clearly moved toward the explicit recognition of a “third”¹⁰⁵ category of things to which personal property rights can relate. Over time, the courts of England and Wales have deliberately and iteratively redefined the scope of things that the common law treats as capable of attracting personal property rights even where those things do not fall easily within either of the two traditionally recognised categories. Initially, this development was in response to emergent forms of intangible things such as milk

quotas; ¹⁰⁶more recently, it has been in response to crypto-tokens.¹⁰⁷

• 3.4    One overarching purpose of our consultation paper was to explore whether consultees agreed with this analysis, whether law reform was necessary or useful to clarify or confirm the common law position and, if so, how to implement such law reform. A strong majority of consultees agreed that either a third category of things to which personal property rights can relate has already developed in England and Wales at common law, or that, to the extent it has not, one should be recognised as existing.¹⁰⁸ This was a clear theme throughout consultee responses, even in those that provided mixed or qualified views.

• 3.5    However, there was also clear and consistent recognition that it is difficult to define the boundaries of such a category. Much of the consistency in responses as to the existence and desirability of such a third category was lost as consultees suggested a variety of different options, support, criticisms and alternative approaches to defining the contours of that category.¹⁰⁹

• 3.6    In this chapter, we discuss these issues in more detail. We conclude that some digital assets are neither things in possession nor things in action but are nonetheless capable of being things to which personal property rights can relate and that this is the existing position at law.

• 3.7    Some consultees, including senior and specialist judges, said to us that it would be helpful to express this position in legislation. They said that this would confirm the existing law, facilitate the law’s future development and lay to rest any lingering doubt that there can be a third category of things to which personal property rights can relate. We recommend such legislation.

• 3.8    However, we conclude that it is not necessary or appropriate for legislation to define the boundaries of such a third category. Whether or not a thing amounts to a third category thing to which personal property rights can relate under the law of England and Wales is a complex and dynamic question, which is ill-suited to static definition in statute. As we said in our consultation paper (in the context of crypto-tokens): ¹¹⁰

Once the thing that constitutes a crypto-token has been determined, the law can then apply existing concepts to determine whether property rights can relate to that thing.

• 3.9    Courts have not only recognised the existence of certain (third category) things but they have recognised that those things are capable of being objects of personal property rights at law. ¹¹¹In doing so, the courts have made a value judgment (either implicitly or explicitly). On balance, we agree with this value judgment, for the following reasons.

• 3.10    First, treating certain digital assets (including crypto-tokens) as things to which personal property rights can relate is a practical and effective means of meeting the legitimate expectations of those parties who interact with such digital assets.¹¹²

• 3.11    Second, this conclusion maintains consistency with the existing common law position,¹¹³ and other international case law, statute and law reform initiatives.¹¹⁴ Multiple courts worldwide, many different statutes and several law reform initiatives already treat certain digital assets (including crypto-tokens) as things to which personal property rights can relate. Each made (implicit or explicit) value judgments to come to that conclusion, but ultimately arrived at the same (or a similar) position. We do not in this report attempt to second-guess or unwind that consistent and diverse reasoning.

• 3.12    Third, this conclusion reflects the factual reality of how certain technology functions. Some digital assets (specifically, crypto-tokens) replicate some of the features of a property relationship between a person and an object. The functionality of cryptotokens allows a person to control access to the crypto-token and gives that person the ability to exclude others from it.¹¹⁵ We conclude that the law can recognise this factual state of affairs and strengthen this technical functionality by aligning it with the social (and legal) construct of personal property rights. In this way, digital things that are recognised by the law as being capable of attracting personal property rights would have some of the functional qualities of an object of personal property rights by technical design, underpinned by a social layer of legal recognition. ¹¹⁶That more accurately reflects commercial realities and the way in which market participants treat such digital things.

• 3.13    Fourth, as we identify in this report, the law of personal property is perfectly able to deal with such things/assets and there is a high degree of existing legal certainty as to how principles of personal property law should apply to such things. As many consultees said, most areas of residual legal uncertainty are highly nuanced and complex. Even so, we make very few recommendations for law reform in this report because of how well the common law has been able to accommodate new digital things. We conclude that it will continue to do so.

• 3.14    Fifth, consultees overwhelmingly supported the proposition that certain digital assets, including crypto-tokens, can and should be capable of being things to which personal property rights can relate.¹¹⁷ This was the position not only in response to our consultation paper, but also in response to our call for evidence.¹¹⁸

• 3.15    Sixth, the idea that some digital assets such as crypto-tokens are capable of being an object/thing to which personal property rights can relate has been discussed and considered in detail for many years and is already widespread in legal and academic commentary.¹¹⁹

• 3.16    We conclude that the common law remains best placed to describe the parameters of third category things that are capable of being objects of personal property rights.¹²⁰ We describe those things that fall within such a third category as “third category things” and digital things that fall within the third category as “digital objects”.

ARE CERTAIN DIGITAL ASSETS THINGS IN POSSESSION?

Things in action and things in possession

• 3.17    The first broad category of personal property rights are rights relating to things in possession. Under the current law, a thing in possession is any object which the law considers amenable to possession. ¹²¹This includes assets which are “tangible, moveable and visible and of which possession can be taken”. ¹²²Since things in possession are capable of being physically possessed, rights in them can be asserted by use and enjoyment as well as by the exclusion of others from them.¹²³ An example of this is a bag of gold: possession of a bag of gold gives its possessor a property right which is enforceable against the world.¹²⁴

• 3.18    The second broad category of personal property rights are rights relating to things in action. They are often described in a narrow sense: “rights in things in action, denied physical enjoyment, are asserted by taking legal action or proceedings”. ¹²⁵Common examples of things in action are debts, rights to sue for breach of contract, and shares in a company. But some commentators have argued that the category of things in action is much broader. ¹²⁶Their argument is that the category of things in action can be treated as a residual category of things that captures any object of personal property rights that is not a thing in possession.

• 3.19    One distinctive feature of things in possession ¹²⁷is that they are things which exist regardless of whether anyone lays claim to them, and regardless of whether any legal system recognises or is available to enforce such claims. In contrast, things in action have no independent form and exist only insofar as they are recognised by a legal system. This means that the presence of a thing in action in the world is dependent on there being both a party against whom the thing in action (the right) can be enforced and a legal system willing to recognise and enforce that right.

Our recommendations on electronic trade documents

• 3.20    In our consultation paper, we said that the category of things in possession is limited to tangible things. ¹²⁸That statement has now been qualified by clause 3 of the Electronic Trade Documents Bill 2023, which allows a person to “possess, indorse and part with possession of an electronic trade document,” ¹²⁹provided that the document satisfies certain criteria.

• 3.21    Even so, possession is not defined in legislation. ¹³⁰It is a common law concept. We recognise that tangibility remains an important characteristic of things in possession: some physical control over a tangible thing is generally required to engage the factual and legal concepts of possession.¹³¹

• 3.22    But in our report on electronic trade documents we said that, while the concept of tangibility helps accurately to describe those things amenable to possession, it is not — nor should it be — a necessary criterion for the law’s recognition of amenability to possession.¹³² So, in the limited context of electronic trade documents, we recommended that it should be possible for electronic versions of trade documents to be treated as possessable things, provided that they meet certain criteria. We also identified elements of the concept of possession which we thought could be extrapolated to electronic trade documents, notwithstanding that they are things which are treated by the law as being intangible.¹³³

• 3.23    As we explain in detail in our report on electronic trade documents, policy and practical considerations in that context led us to recommend that electronic trade documents should be treated as possessable things.¹³⁴ In summary, the concept of possession has a fundamental role in the functionality of paper trade documents such as bills of lading and bills of exchange, both at common law and in domestic statutes. ¹³⁵Possession of a trade document is crucial in establishing which party has certain rights and entitlements, how the documents are held and how they are used in collateral arrangements. Treating electronic trade documents as possessable things allows them to be plugged into an existing legal framework in respect of a limited category of documents that already enjoy a special status in law. In effect, this allows functionally equivalent documents to be treated by law in exactly the same way regardless of their medium.

• 3.24    The Electronic Trade Documents Bill explicitly engages the legal concept of possession for that limited group of electronic trade documents that fall within its scope.¹³⁶ Electronic trade documents are not tangible in the conventional sense.¹³⁷ Under the Bill, the concept of possession is not limited in its application to tangible things.¹³⁸ But we recognise that the legal concept of possession is nonetheless bound up with tangibility. It has traditionally relied on the physical boundaries of a thing to help define the contours of legal duties in relation to that thing.¹³⁹

Our approach to third category things and possession

• 3.25    Some consultees questioned the distinction between electronic trade documents (which are to be treated as possessable things as a matter of statute) and third category things which we do not recommend are categorised as things in possession.¹⁴⁰

• 3.26    As we discussed in detail in chapter 11 of our consultation paper, we do not think that the arguments for using possession as the operative concept in respect of electronic trade documents are as persuasive in respect of other forms of third category things. One reason is that other third category things, in general, do not seek to replicate the legal functionality of a specific form of tangible thing in the same way that electronic trade documents attempt to replicate exactly the legal functionality of paper trade documents. ¹⁴¹Indeed, many third category things were designed to avoid replicating some of those features. Most obviously, crypto-tokens were designed to facilitate communication of value on a global and trust-minimised basis, without the need for physical exchanges of tangible things.¹⁴²

• 3.27    Our approach does not require third category things to be treated as possessable things, which confirms the current legal position. We think that drawing analogies between tangible things and third category things is helpful to a point but, inevitably, those analogies are not perfect. This is particularly true in respect of those third category things that rely on novel or idiosyncratic technology, such as open-source code, public, distributed ledgers and public key cryptography.

• 3.28    Our preferred approach is for the law to develop (by a combination of common law and statute, informed by non-binding guidance) specific legal rules and principles which apply to third category things and are responsive to the specific technology that they use. We conclude that the law should not attempt directly to apply legal principles applicable to things in possession or things in action to third category things. For this reason and for the reasons we discuss in Chapter 5 (Control), we conclude that, instead of directly applying the concept of possession, the law of England and Wales can develop jurisprudence around a concept of control which is better suited to the functions of third category things and the technology they use. We think that control will form an important constituent part of the legal rules and principles that apply to complex interactions with third category things. In particular, we conclude that control should be considered as a core constituent element within a range of higher-level organising or framing principles such as legal concepts of transfer, intermediated holding arrangements, collateral arrangements and causes of action and associated remedies. ¹⁴³This is consistent with international developments which also focus on control and not possession. ¹⁴⁴Moreover, control might be a more suitable concept for assets that are global in nature, given that, very broadly speaking, many civil law systems require a thing to have a tangible or physical form for it to be recognised as an appropriate object of certain property rights such as “ownership”.¹⁴⁵

• 3.29    Things in possession, things in action and third category things are different. They function in different ways. Control in the digital context, as applicable to third category things, is highly composable and nuanced, and remains open to technological evolution. Possession of conventional tangible objects and control over things in action, however, are perhaps more limited. Different markets and market practices have arisen in relation to tangible things, to things in action and to third category things.

• 3.30    For these reasons — and to the extent that third category things remain an important part of the global economy — we conclude that the law should develop legal principles specific to third category things. Those principles are likely to diverge, either significantly or in small ways, from the existing legal principle of possession and the associated legal constructs applicable to possessable things, such as bailment, pledge, lien, and the tort of conversion.¹⁴⁶ This report concludes that it is entirely appropriate for the law to do so and reinforces the legal foundations for that process.

ARE CERTAIN DIGITAL ASSETS THINGS IN ACTION?

• 3.31    Broadly, there are two arguments that might be made in favour of characterising certain digital assets as things in action. The first argument is that the digital asset in question exists as a right or claim in action. Or, alternatively, that rights or claims in relation to the digital asset in question are rights or claims enforceable only by action (and that they cannot be asserted merely by use and enjoyment as well as by the exclusion of others from them as is possible with things in possession).¹⁴⁷ The second argument is that the category of things in action can be treated as a wide, residual category of things that captures any object of personal property rights that is not a thing in possession. Below, we explain why we do not agree with either argument.

Crypto-tokens as things in action?

• 3.32    Almost all consultees agreed that there exist certain intangible things that are not things in action in the traditional, narrow sense, but that those intangible things should nevertheless be things to which personal property rights can relate. Specifically in respect of crypto-tokens, almost all consultees said that crypto-tokens cannot be conceived of as rights or claims in themselves and that they can be used and enjoyed independently of whether any rights or claims in relation to them are enforceable by action.¹⁴⁸

• 3.33    However, one consultee, Professor Low, made two contrary arguments. First, that crypto-tokens could be narrowly conceived as “the right to the unique data strings on a particular distributed ledger, or put slightly differently, the right to have unspent transaction output (UTXO) locked to a public address with a particular ledger”.¹⁴⁹ Second, that the category of things in action is in any case not limited to things which consist of rights or claims enforceable only by action.¹⁵⁰

• 3.34    On the first argument, we disagree that this is an analysis that can properly be applied to crypto-tokens. In this report, we describe crypto-tokens as composite things: notional quantity units manifested by the combination of the active operation of software by a network of participants and network-instantiated data. ¹⁵¹Although those notional quantity units themselves can be considered “abstract” and can have “value”, that explanation is not complete because the software and data together manifest a “thing” that is necessarily independent of any rights that may be claimed over that thing. For a crypto-token, the thing itself (the token) is not therefore co-extensive with the right in relation to it in the way that a thing in action is. Crypto-tokens would continue to exist even if the law were to fail to recognise them as objects of personal property rights and even were a law to prohibit their existence.¹⁵² Their useful characteristics and the ability of people to use, enjoy and interact with them (and exclude others from them) would also continue to exist: the functionality of the cryptotoken system would remain unaffected. For example, in Japan, the proprietary status of crypto-tokens was considered after the bankruptcy of the Mt. Gox crypto exchange in 2014. In August 2015, the Tokyo District Court held that bitcoin was not a thing that was capable of ownership within Article 85 of the Japanese Civil Code.¹⁵³ That finding prevented the potential argument that customers of the exchange had a proprietary right or interest in the bitcoin held by the exchange (as opposed to merely a personal claim against the exchange). Regardless of that finding, the bitcoin in question (and its functionality and utility) continued to exist and distributions of the remaining Mt. Gox held bitcoin to creditors of Mt. Gox are scheduled for 2023.¹⁵⁴ We conclude, therefore, that the better view is that a crypto-token is a thing in itself to which personal property rights can relate.¹⁵⁵ Indeed, that was also the conclusion that the Japanese legislature came to after initially rejecting the analysis that bitcoin could be an object of personal property rights.¹⁵⁶

Things in action as a residual category

• 3.35    On the second argument, we consider that it would be possible for the courts to recognise the category of things in action as a wider, residual category of things encompassing everything that is not a thing in possession. It would also then be possible to develop, as Professor Low suggested:¹⁵⁷

a sub-classificatory system distinguishing certain kinds of [things] in action (such as contractual rights) from others (such as intellectual property rights).

• 3.36    However, it is not entirely clear how this would be practical or helpful for the development of the law. Nor is it obvious why, if such an approach were desirable, it has not been the position taken by the common law for the last ten years, despite the fact that the courts have considered the issue directly on multiple occasions.¹⁵⁸ Moreover, we conclude that such an approach risks creating additional legal uncertainty, particularly if certain third category things such as crypto-tokens were inadvertently conceptualised as “rights” to which personal property rights could relate (or were co-extensive with). It also risks diluting or confusing the defining features of things in action (in the narrow sense), which at the moment can (for the most part) be clearly identified, and would mean that different things in action would be subject to different treatment. We conclude that, if different things are to be treated differently, it is cleaner to recognise a separate third category rather than multiple “sub-categories” of a broad residual category. Our recommended approach is more direct and reflects the views of a strong majority of consultees that the more conceptually coherent position is to recognise a third category of things to which personal property rights can relate. Our approach is also consistent with and does not disrupt a clear — and ever increasing — line of case law (which we discuss in detail at paragraph 3.38 below).

• 3.37    For that reason, we conclude that the better argument is reflected in the submission made by Linklaters LLP:¹⁵⁹

It is clear from the consultation paper and the UKJT Legal Statement that there is already a high degree of legal certainty that there exist certain types of intangible property that are not things in action (in the strict sense). Given this, we strongly agree that the law of England and Wales should recognise a third category of property (in our view, it already does). We agree that it is not conceptually coherent or helpful for crypto-tokens and other intangible assets that do not consist of a legal claim by the property holder against another legal person to be treated as things in action (or at least not things in action to which the legal principles applicable to things in action in the strict sense apply). Treating them as such implicitly limits the flexibility of the common law to develop legal principles that are best suited to such intangible assets, including in relation to transfers, security and relative title. Put another way, if such intangible property were to be included in a category of things in action, it would be necessary to bifurcate the legal principles applicable to things in action between those that consist of a legal claim by the property holder against another legal person and those that do not, so that the common law develops appropriate legal principles for this new sub-category. This seems tantamount to acknowledging a third category of property. Similar considerations apply to suggestions that intangible assets that do not consist of a legal claim by the property holder against another legal person should, instead, be characterised as things in

possession, which is a suggestion made by other commentators that consider that such property should be subject to possessory-style legal concepts.

The position at common law

• 3.38    Our conclusion is consistent with the existing position at common law which has over the last ten years clearly moved toward the explicit recognition of a third category of thing to which personal property rights can relate.

• 3.39    Courts have consistently concluded that certain things (often digital assets) are capable of being objects of personal property rights, even where the thing in question does not neatly fit within either of the traditionally recognised categories of thing to which personal property rights can relate. ¹⁶⁰The courts have done so, either expressly or impliedly, in respect of milk quotas, ¹⁶¹European Union carbon emission allowances (“EUAs”), ¹⁶²export quotas, ¹⁶³waste management licences, ¹⁶⁴and a wide variety of crypto-tokens, including NFTs.

• 3.40    Indeed, since the publication of our consultation paper, the Court of Appeal said that “a cryptoasset such as bitcoin is property” under the law of England and Wales.¹⁶⁵ This is also affirmed (or necessarily implicit) in at least 23 other cases decided at first

instance, ¹⁶⁶although most were decided in connection with interim relief.¹⁶⁷

• 3.41    Since the judgment in AA v Persons Unknown¹⁶⁸ was handed down in 2019, courts in at least 14 of those 24 cases, including the Court of Appeal, ¹⁶⁹have cited that judgment in support of the proposition that the digital asset in question is a thing which is capable of being an object of personal property rights. In AA v Persons Unknown, Mr Justice Bryan said that it would be “fallacious” to proceed on the basis that the law of England and Wales recognises no form of property other than things in possession and things in action. He explicitly recognised the difficulty in the classification of crypto-tokens (which, on their face are things which are neither things in action nor things in possession). Citing the full reasoning of the UKJT Legal Statement on the point, he held that a crypto-token could be an object of personal property rights even if it was not a thing in action in the narrow sense.¹⁷⁰

• 3.42    Taken together, the case law demonstrates that the courts of England and Wales now recognise crypto-tokens as distinct things which are capable of being objects of personal property rights. Further, through the consistent application of AA v Persons Unknown (as opposed to any contrary approach), ¹⁷¹courts have deliberately proceeded in a manner that carves out a third common law-based category of thing to which personal property rights can relate.

• 3.43    Courts in other jurisdictions have reached the same (or a similar) conclusion. Courts across the common law world now consistently proceed on the basis that cryptotokens are capable of being objects of personal property rights and are therefore susceptible to the various consequences that follow. This includes recognition that crypto-tokens can be subject to an interlocutory proprietary injunction, are capable of being held on trust and fall within certain broad statutory definitions of “property”. See, for example, cases in Australia,¹⁷² Canada, ¹⁷³Hong Kong, ¹⁷⁴New Zealand,¹⁷⁵

Singapore,¹⁷⁶ and the United States,¹⁷⁷ including in substantive proceedings. ¹⁷⁸In those cases, courts have also referred to or relied on the analysis in AA v Persons Unknown,¹⁷⁹ the UKJT Legal Statement,¹⁸⁰ and our consultation paper.¹⁸¹

• 3.44    Moreover, as we discussed in more detail our consultation paper, ¹⁸²we consider that it is possible that over time, civil law systems will also move toward a broader recognition of some type of third category of thing to which personal property rights can relate.

• 3.45    As Professor Allen, Henry Wells and Marco Mauer suggest in the context of civil law systems:¹⁸³

We may have reached the point [where] ‘native’ digital objects demand recognition in their own right, however disruptive this may be for the dogmatic structure of inherited legal categories.

• 3.46    Examples of this can already be seen in some civil law-based systems, including Japan,¹⁸⁴ Liechtenstein, ¹⁸⁵and Switzerland,¹⁸⁶ each of which we discuss in more detail in our consultation paper.

• 3.47    Our conclusions are also consistent with international law reform developments, including those that are intended to be applicable in civil law jurisdictions. The UNIDROIT Working Group recently published a set of international principles, ¹⁸⁷which set out a proprietary framework applicable to digital assets.¹⁸⁸ The UNIDROIT Working Group Principles apply to “electronic records”, of which digital assets are a sub-set.¹⁸⁹ In effect, the Principles apply proprietary concepts to a category of things distinct from things in possession and things in action.

• 3.48    In England and Wales, across the common law world, and in other jurisdictions, there is now a persuasive, clear, and well-reasoned body of case law that concludes that certain digital assets are capable of being objects of personal property rights. Much of the reasoning in that case law relies on analysis that supports or is consistent with our approach that recognises there is a third category of thing to which personal property rights can relate. As such, our conclusion and recommendation that the law of England and Wales either already does recognise, or should explicitly recognise, such a category is intended to be confirmatory only. We conclude that the common law has already moved on from the question as to the proprietary status of certain digital assets, and how best to categorise them. The more difficult and important residual areas of legal uncertainty relate to the legal principles applicable to those things that fall within the third category.

A THIRD CATEGORY THING — ANALOGIES WITH BOTH THINGS IN POSSESSION AND THINGS IN ACTION

• 3.49    We think it wrong implicitly (or explicitly) to limit the flexibility of the common law to develop legal principles specific to third category things. Rather, we conclude that the common law should remain free to develop, where appropriate, legal principles specific to third category things, including but not limited to, crypto-tokens, cryptoassets,¹⁹⁰ “digital bearer securities” ¹⁹¹and to the specific technology that manifests them.

• 3.50    Some of the fundamental, defining features and purposes of these types of thing are that they function differently to both things in possession and to things in action. As Timothy Chan and Professor Low argue:¹⁹²

it is crucial that courts faced with cryptoasset disputes avoid the simplistic analogy between the tangible and intangible.

• 3.51    We agree. We think the better approach is for the law to develop by analogy with principles applicable to things in possession or things in action where appropriate, while also recognising that those analogies will be imperfect. The law should instead focus on the attributes or characteristics of the thing with which it is concerned in a particular case. It should not attempt rigidly to apply to third category things legal principles that were formulated by reference to other things that are capable of being objects of personal property rights.

• 3.52    As a specific example, some crypto-tokens were explicitly designed to remove the need for trusted intermediaries and instead to facilitate peer-to-peer communication of value. They function in some ways like “digital bearer instruments”, ¹⁹³and are in some ways tangible object-like in nature. Crucially, they can be self-held (“self-custodied”)¹⁹⁴ by a person and they exist independently of the legal system. In other words, they are not rights in themselves but are instead things to which rights can relate. They can be used and enjoyed independently of whether any rights or claims exist in relation to them and any rights in them can be asserted by use and enjoyment as well as by the exclusion of others from them.¹⁹⁵

• 3.53    This is all achieved through software. ¹⁹⁶Such third category things are, in general, not tangible in the conventional sense and so analogies with tangible things are imperfect. For example, transfers of tangible things (by delivery) work in a very different way to transfers of third category things.¹⁹⁷

• 3.54    Equally, analogies with things in action are imperfect and misunderstand the nature and functionality of digital objects such as crypto-tokens. Things in action, such as debts, have no existence independent of persons or legal systems and, as such, are not involuntarily alienable. The use or enjoyment of a thing in action is dependent only on the enforceability of the right or claim of which it is constituted. That is not true of digital objects. This is the crucial distinction that needs to be made for proprietary classification purposes. The susceptibility of something to involuntary alienation is relevant to a proprietary classification because it helps to distinguish the different functional characteristics of the thing in question. A debt, for example, cannot be effectively alienated from a person without a legal process (usually one which requires that person’s consent). A crypto-token, on the other hand, like a car or a watch, can as a matter of fact be alienated from a person without a legal process and without their consent. A crypto-token can be used, enjoyed and interacted with without the need to rely on the enforceability of any rights or claims in relation to the token. The functional characteristics of a digital object and a thing in action are therefore fundamentally different. To apply analogies between things in action and digital objects would therefore be a legally erroneous step.

• 3.55    This is an important point because it helps to identify a way of dealing with legal and policy questions that have practical implications for the law of England and Wales. For example, the UKJT Legal Statement on Digital Securities started by asking how new digital objects are different:¹⁹⁸

The starting point of our analysis is therefore, as it was in the First Legal Statement, to identify what, if anything, might genuinely be novel and distinctive from a legal perspective about the use of blockchain and DLT in such structures.

• 3.56    Similarly, HM Treasury grounds its proposed regulatory framework on the view that crypto-tokens can function differently to both things in possession and to things in action, and might give rise to different risks:¹⁹⁹

Over recent years cryptoassets and the activities underpinning their use (or ‘crypto’) has evolved into an extensive, complex, and rapidly evolving ecosystem. It features a myriad of different activities and business models, each generating different types of opportunities and risks for the actors involved.

• 3.57    We conclude that precise, technology-specific analysis of where particular digital assets, networks or systems remove or ameliorate some risks and introduce other risks is the most appropriate method for developing a comprehensive, fair and competitive legal and regulatory framework for the benefit of all participants. Our view is that such analysis is not possible where legal principles applicable to tangible things, or legal principles applicable to things in action, are applied directly to third category things without modification.

• 3.58    Sophisticated markets have already developed for third category things such as crypto-tokens, NFTs and VCCs, and it is likely that those markets will grow and develop over time. That said, it is not a foregone conclusion that third category things will continue to be used in markets. ²⁰⁰To the extent they are not, the third category of things to which personal property rights can relate will remain ancillary and of limited importance. But in the meantime, there will likely be a period of uncertainty as the value proposition and competitiveness of third category things are tested against other existing markets, structures and arrangements which achieve similar functionality. Private law should not stand still in this period.

AVOIDING UNNECESSARY BOUNDARY ISSUES

• 3.59    In our consultation paper, we provisionally proposed that a thing which satisfies certain criteria should be recognised as falling within our proposed third category of things to which personal property rights can relate. These criteria were that a thing (which we called a “data object”) ²⁰¹must:

• (1)    be composed of data represented in an electronic medium, including in the form of computer code, electronic, digital or analogue signals;

• (2)    exist independently of persons and exist independently of the legal system; and

• (3)    be rivalrous.

  • 3.60    In Chapter 4 (Third thing in practice), we discuss consultee responses to our proposed criteria and make consequential modifications and clarifications to them. We also explain why we no longer recommend the creation of hard boundaries for such a third category, and now treat our criteria as indicia.

  • 3.61    Given that we do not recommend the creation of bright line limits to such a third category, no new boundary issues should arise as a consequence of our recommendation. However, there are bound to be things that might remain difficult to categorise. Courts will also have to determine whether new things can (and should) be capable of being objects of personal property rights.

TERMINOLOGY: DATA OBJECTS, DIGITAL OBJECTS AND THIRD CATEGORY THINGS

Data objects and digital objects

• 3.62    In our consultation paper, we used the term “data objects” to describe the things we proposed should fall within a third category of thing to which personal property rights can relate.

• 3.63    Consultees made helpful comments on the term “data object” which highlighted two broad points. First, that the use of the term “data” implied an overly data-centric approach to the conceptualisation of the asset in question. ²⁰²Second, that the term “data object” already has a specific use in programming and it would be sensible to avoid confusion between the two. Taking consultees’ responses on board, we no longer use the term “data object”.

• 3.64    Instead, in this report, we use the term digital asset in a very broad, descriptive sense. We use the term “third category thing” for those things that fall within the third category. We now use the term “digital object” to refer to a digital asset within the third category (examples include a crypto-token or an NFT). Where appropriate or where greater specificity is required, we use more precise, asset and technology-specific terms, such as crypto-token, cryptoasset, NFT or the name of the crypto-token in question.

“Third” category

• 3.65    We use the term “third category” to describe a category of thing distinct from both things in possession and things in action. In adopting this terminology, we acknowledge the argument that other distinct categories of thing to which personal property rights can relate might already exist at law (including patents and statutorily created intellectual property rights).²⁰³ However, we nonetheless adopt the term third category as shorthand: in part, as a direct reference to Lord Justice Fry’s judgment in Colonial Bank v Whinney and the longstanding practice among lawyers and judges of referring to the things in possession/things in action dichotomy;²⁰⁴ in part, as an intuitive, convenient and readily understandable term, which almost all consultees were comfortable with; and in part, because we consider that this distinct third category of thing that is better suited to encompassing new assets (including new digital assets) will become increasingly important to the modern world.

OUR RECOMMENDATION: MINIMALIST STATUTORY CONFIRMATION

• 3.66    In an increasingly online and digital world, we expect that other intangible things and assets whose parameters are difficult to predict and define in the abstract are bound to develop.²⁰⁵

• 3.67    We conclude that the parameters of those things²⁰⁶ that the law treats as third category things capable of attracting personal property rights should continue to be defined by the common law. As such we conclude that there is no need to define the boundaries of such a third category in statute.

• 3.68    Instead, we recommend that the law explicitly recognise that a thing will not be deprived of legal status as an object of personal property rights merely by reason of the fact that it is neither a thing in action nor a thing in possession. Such things include crypto-tokens like bitcoin and ether. Those things do not exist as rights or claims in themselves (they instead exist independently). They also can be used and enjoyed independently of whether any rights or claims in relation to them are enforceable by action.

• 3.69    We conclude that over the last ten years the common law has clearly moved toward the explicit recognition of a third category of thing to which personal property rights can relate. This conclusion is supported by over 24 cases.²⁰⁷ In that sense, our recommendation amounts only to a confirmation and restatement of the existing common law position.

• 3.70    However, some consultees, including senior and specialist judges, ²⁰⁸said to us that it would be helpful to express this position in legislation.

• 3.71    Although it would not change the common law position, ²⁰⁹we conclude that such a statutory confirmation will provide greater legal certainty and will allow the law to develop from a strong and clear conceptual foundation. Such confirmatory legislation will also indicate to the courts that Parliament has decided that the exact parameters which describe a digital asset, and the circumstances in which such a digital asset is capable of being an object of personal property rights, is a matter for common law.

There are centuries of case law considering the factors that make a thing an appropriate object of personal property rights, which the courts can continue to apply in this context so that the third category does not become inappropriately broad. We consider this to be the most effective and least interventionist recommendation that we can make to facilitate the law’s development on this point.

• 3.72    A statutory confirmation will emphatically recognise the reality of the modern world that intangible things which are not rights or claims enforceable only by action exist and that the law is capable of treating those things as objects of personal property rights. This in turn will allow the law of England and Wales to discuss crypto-token systems (and other systems that might manifest third category things) more directly in terms of powers and incentives/incentive mechanisms of participants, rather than in terms of claims/rights, corresponding duties and obligations.²¹⁰ It permits things in action to remain as a usefully distinct and descriptively accurate category.²¹¹

• 3.73    Moreover, our recommendation and the minimalist statutory confirmation we envisage are technology neutral in that they do not focus on any single or class of (digital) asset, crypto-token in particular, or any protocol, system, network or technological feature. Our recommendation allows the law to interrogate the idiosyncratic features of the asset in question when considering its proprietary status. It avoids drawing arbitrary boundaries or creating rigid definitional issues. It emphasises the success, and trusts in the continued ability, of the common law to develop sensitively and flexibly in the face of rapidly developing technology. It maintains the common law’s position at the heart of the law of personal property. It distinguishes the law of England and Wales as a flexible and open system that is alive to the particular characteristics and design features of specific technology.

• 3.74    A statutory confirmation will reduce the time spent by the courts on questions of categorisation of objects of personal property rights, and instead allow them to focus on the substantive issues before them. It gives explicit effect to:²¹²

[the] powerful case for reconsidering the dichotomy between [things] in possession and [things] in action and recognising a third category of intangible property ... in a way that would take account of recent technological developments.

• 3.75    A statutory confirmation will also alleviate any judicial concern that recognising a third category is not an appropriate development for the common law to make.²¹³ In short, this recommendation lays to rest the lingering authority of Colonial Bank v Whinney²¹⁴ which suggested that there could be no third category thing and confines to history the “red herring” ²¹⁵of a circular and hollow debate on categorisation. The law can instead focus on answering substantive questions as to the proper application of legal principles to third category things. We discuss some of those legal principles in the rest of this report.

Recommendation 1.

• 3.76    We recommend statutory confirmation that a thing will not be deprived of legal status as an object of personal property rights merely by reason of the fact that it is neither a thing in action nor a thing in possession.

Chapter 4: Our third category recommendation and conclusions in practice

INTRODUCTION

• 4.1   In Chapter 3 (Third thing), we conclude that the common law of England and Wales

now recognises certain digital assets, including crypto-tokens, as distinct things which are capable of being objects of personal property rights. We conclude that such things should be regarded as falling within a third category of things to which personal property rights can relate, rather than being categorised as things in action or things in possession.

• 4.2   In this chapter, we consider the types of things that fall within such a third category.

• 4.3   We set out the criteria for third category things that we proposed in our consultation

paper and discuss consultees’ comments in response to them. We make consequential modifications and clarifications to those criteria (which we now treat as indicia).

• 4.4    We then consider consultees’ concerns about boundary issues. We discuss cryptotokens, private, permissioned blockchain systems, voluntary carbon credits (“VCCs”), in-game digital assets and digital files by way of example. We explain why we conclude that it is not appropriate to recommend criteria definitively to delimit the boundaries of an exhaustive third category of things to which personal property rights can relate. Instead, we conclude that the parameters of this category of things should continue to be developed by the common law, guided by the indicia we set out.

OUR PROPOSED CRITERIA

• 4.5    In our consultation paper we provisionally proposed that a thing should be recognised as falling within our proposed third category of things to which personal property rights can relate ²¹⁶if:

• (1)    it is composed of data represented in an electronic medium, including in the form of computer code, electronic, digital or analogue signals;

• (2)    it exists independently of persons and exists independently of the legal system; and

• (3)    it is rivalrous.²¹⁷

Data

• 4.6   Our first provisional criterion was that a thing must be:

Composed of data represented in an electronic medium, including in the form of computer code, electronic, digital or analogue signals.

• 4.7    We said that this was a wide gateway criterion, unable on its own to distinguish “data objects” from information in a broader sense. That distinguishing role was instead performed by our third criterion that a thing must also be rivalrous.²¹⁸

• 4.8    Although many consultees agreed with our first criterion, many did not. ²¹⁹Those who did not agree said that “composed of data” resulted in an overly data-centric approach to conceptualising the digital asset in question and could potentially create a hard boundary for the third category.²²⁰ Those consultees acknowledged that data was important part of these digital assets but said that they are not mere data alone. As the UKJT Legal Statement put it, “it is not what the data tells you but what it allows you to do”.²²¹

• 4.9    We did not intend our criterion to focus enquiry on data alone because we conceptualise certain digital assets, such as crypto-tokens, as composite things, not mere data.²²² The intention was:

• (1)    to use this criterion to distinguish digital assets from tangible things; ²²³and

• (2)    to acknowledge that the instantiation of data within a crypto-token system is a core part of the object in question but that data alone is not the thing which is the object of personal property rights.²²⁴

• 4.10    Given what we see as valid criticism of this criterion, we conclude that the better argument is simply to acknowledge that such digital assets are not entirely analogous with conventional tangible things, thus distinguishing them from tangible things in possession.²²⁵ And we reiterate our conclusion that the informational and functional qualities of network-instantiated data provide the substance to the concept of rivalrousness.²²⁶ This means that rivalrousness can function as our principal criterion and that “composed of data” need not be a criterion in itself.²²⁷

• 4.11    We develop this point further by reference to crypto-tokens below, with the benefit of the persuasive Court of Appeal authority in Tulip Trading. ²²⁸That case has brought a high degree of certainty to the law of England and Wales: it recognises that cryptotokens can be things to which personal property rights can relate, that they can be rivalrous and that their characteristics are manifested by the active operation of software.²²⁹ In short, the judgment demonstrates the practical advantages of our conclusion that the common law of England and Wales remains the appropriate forum for development and resolution of the legal issues discussed in Chapter 3 (Third thing) and this chapter.

• 4.12    Consistent with the reasoning in Tulip Trading, our approach remains directly to characterise crypto-tokens as composite things (as outlined in our consultation paper).²³⁰

Crypto-tokens as composite things

• 4.13    By way of example, for crypto-tokens, we broadly view the thing, the asset or the object in the following way:

A crypto-token exists as a notional quantity unit manifested by the combination of the active operation of software by a network of participants and network-instantiated data.

• 4.14    This conceptualisation has been expressed in very similar terms by expert market participants, legal academics and the Court of Appeal:

• (1)    Linklaters LLP: “For crypto-tokens, we view the asset as the power to effect a state change within the crypto-token system, as instantiated in that system. This power is not composed of data, although it is instantiated in the system through the combination of certain data (including cryptographic keys) and the system protocol (including the embedded cryptography), which ensure that the ‘cryptotoken’ (an ideational construct) cannot be double spent.”²³¹

• (2)    Professor Fox: “A [crypto-token is] an ideational thing containing different

components. It is more complex than the £1 coin since it lacks any tangible basis and its most significant properties are matters of digital functionality rather than legal attribution. Like the coin, however, it comprises more than one component. It is grounded in, but not confined to, the technical features of its own digital design. Its outward manifestation is a string of data generated by transactions between participants on a distributed ledger system. But to see the asset as mere data would ignore its larger functionality, just as we would fail to appreciate the full economic or legal significance of a coin by treating it as a mere metal disc. If the law is to recognise digital assets as property for private law purposes, then it would benefit from analysing them as composite things. The asset is more than mere data. It is a set of transactional functionalities. The most important of these is the capacity of the person who holds the private key to effect new transactions which will be recognised as valid by the technical rules of the system. Analysed in this way, the asset can be viewed as a specific transactional power over unique data entries on the ledger.”²³²

• (3)   Clifford Chance LLP: “The thing that comprises the property could be

characterised as the notional thing which is recorded in the distributed ledger, but is not the data (or data strings or data structures) which form the ledger.”²³³

• (4)    The Court of Appeal: “The unusual factual feature of the present case is that literally all there is, is software. A physical coin has properties which exist outside the minds of people who use it and in that sense is tangible. Bitcoin is similar. It also has properties which exist outside the minds of individuals, but those properties only exist inside computers as a consequence of the bitcoin software. There is nothing else.”²³⁴

• (5)    See also our consultation paper at paras 1.8, 10.16, 10.26, 14.3 and Appendices 3 and 4.

• 4.15    So, a crypto-token exists as a notional quantity unit manifested by the combination of the active operation of software by a network of participants and network-instantiated data.

• 4.16    It is this combination of the active operation of software by a network of participants and network-instantiated data which gives rise to certain functionalities of cryptotokens that manifest characteristics which make them distinct from other digital assets. ²³⁵As we said in our consultation paper: ²³⁶

While [a crypto-token’s] form relies on its technical instantiation as a data structure, its function is derived not merely from the abstract existence of the technical system in which it persists, but fundamentally by the active operation of that system by a network of users. A crypto-token is consequently an object that has both, and is a composite of, technical and social dimensions — crypto-tokens exist as instantiations in socio-technical systems.

• 4.17    One of the functions of the network-instantiated data is to track the “state” — the canonical and chronological order of transactional events as recorded within the transaction-based ledger or record — of the crypto-token system.²³⁷ However, the record is neither the crypto-token system nor the asset itself. The crypto-token system is a “transaction-based state machine” or “state transition system”.²³⁸ The active operation of the specific system protocol software by a network of participants gives a crypto-token intrinsic functionality by specifying, among other things, how consensus is reached to introduce a valid “state transition” into the canonical system state.²³⁹

• 4.18    A principal operative functionality of a crypto-token is the ability/power ²⁴⁰uniquely to perform an operation (or an action) ²⁴¹within the crypto-token system in respect of that notional quantity unit. Neither the notional quantity unit (as an ideational thing) nor this ability/power is composed of data itself. However, form and the functional qualities of the notional quantity unit, including the ability/power in respect of such notional quantity unit, only arise (and continue to exist) as a result of the crypto-token system.

This means that the functional qualities arise through the combination of network-instantiated data ²⁴²and the active operation of the system by participants in the network (that run the specific system protocol software).²⁴³ A core purpose of the crypto-token system is to ensure that the notional quantity units cannot be double spent — by specifying how consensus as to “state” (including state transitions) is reached — meaning that the notional quantity unit in question is rivalrous.²⁴⁴

• 4.19    So, as Professor Fox says: “If the law is to recognise digital assets as property for private law purposes, then it would benefit from analysing them as composite things”. ²⁴⁵A crypto-token is not a record or representation of something; it is a distinct and independent thing.

• 4.20    This description recognises the reality that a crypto-token is a digital object that has both, and is a composite of, technical and social dimensions. It is crypto-tokens as notional quantity units, arising from a composite of technical form, technical function and social participation/recognition, that the market and the legal system treat as a thing, and to which society has chosen to attach legal consequences.²⁴⁶

• 4.21    But there is an easier way of describing this: simply by referring to the notional quantity unit itself, or to “tokens”. ²⁴⁷Market participants and non-lawyers settled on this terminology long ago. In fact, colloquial use provides a perfectly adequate general-purpose abstraction of the legal concept we describe here. We think for example that it is both accurate and precise to say “1 bitcoin”,²⁴⁸ “10 ether”,²⁴⁹ “100 UNI” ²⁵⁰or “1 Bored Ape”. ²⁵¹In the vast majority of cases, there is little need to say more. Indeed, courts in general simply refer to a specific quantity of the particular notional quantity unit/crypto-token in question.²⁵²

Independent of persons and independent of the legal system

• 4.22    Our second criterion was that a thing must exist independently of persons and exist independently of the legal system.

• 4.23    The first part of this criterion was intended to ensure that our proposed third category admitted only those things that are properly identified as distinct things, existing independently from any particular person. ²⁵³The second part of this criterion aimed to prevent things in action (and statutorily created property rights, such as intellectual property rights) from satisfying the criteria for our proposed third category of thing to which personal property rights can relate.²⁵⁴

• 4.24    We asked consultees whether they agreed with this criterion. Consultee responses were evenly split on this point, although a slight majority agreed outright.²⁵⁵ Broadly, those in favour agreed that certain digital assets exist both independently of persons and independently of the legal system, as described in our consultation paper. Most of those consultee responses centred around crypto-tokens, distinguishing crypto-tokens from other existing things and other digital assets.

• 4.25    However, some consultees criticised this criterion. The principal criticisms were:

• (1)    Existence independent of persons should not exclude those things — such as crypto-tokens — that rely on a set of network participants for their continued existence.²⁵⁶

• (2)    Some third category things, such as crypto-tokens, are tangentially related to the legal system. For example, crypto-tokens might be controlled or operated via smart contracts and these smart contracts might contain arbitration mechanisms.²⁵⁷ Further, crypto-tokens have been recognised as things to which personal property rights can relate by court judgments and various statutes.²⁵⁸

• (3)    Digital assets recorded in a private, permissioned blockchain or DLT-based system may not be clearly independent of the legal system. As the Financial Law Committee of the City of London Law Society (“CLLS-FLC”) said:²⁵⁹

Such systems will be operated, managed and administered under contractual and/or statutory rules and protocols. They will or are likely to create private law rights and obligations as between participants and the operator/administrator in relation to the maintenance of the distributed ledger/structured record and settlement processes (for the holding and transfer of title to the digital assets recorded in the systems). The existence of such private rights and obligations is likely to obfuscate the analysis as to whether the digital assets themselves (issued, held and transferred by means of such a system) can properly be considered as existing “independently of the legal system”.

• 4.26    We acknowledge the potential strength of these criticisms and clarify the operation of this criterion to reflect the concerns of consultees in the following ways.

• (1)    The primary intention of this criterion was to distinguish third category things from things in action.²⁶⁰ We do not think, nor did we intend, that those cryptotokens which rely on a network of participants for their continued existence should be excluded by this criterion. We conclude that it remains accurate to describe those things as existing independently from any particular person. We explain our reasoning for distinguishing things in action in Chapter 3 (Third thing).

• (2)    A thing’s recognition by a legal system (including a legal system’s recognition of things to which personal property rights can relate) is not the same as that thing’s reliance on a legal system for its continued existence. Anything can be recognised by a legal system, but things in action can only come into being by virtue of, and can only function within, a legal system. Bags of gold, for example, exist independently of the legal system, and rights in relation to them might subsequently receive legal recognition. The same is true of crypto-tokens within crypto-token systems. Crypto-tokens are not rights in themselves and they exist independently of any rights or claims in that might also exist in relation to them. They can also be used and enjoyed independently of whether any rights or claims in relation to them are enforceable by action. In contrast, the same is not true of debts: their existence is not anterior to legal recognition; rather it is co-extensive with it. This means that they cannot function, be used or enjoyed without that legal recognition.²⁶¹

• (3)    We acknowledge that private, permissioned blockchain or DLT-based systems work differently from public, permissionless systems. We also acknowledge that even public, permissionless systems can be used in different ways by market participants — for example, merely as a method of recording certain “offchain” things using tokens. ²⁶²However, given that we no longer propose that our criteria define the hard boundaries of the third category of things to which personal property rights can relate, we do not see this as problematic. Legal rights (as opposed to digital objects) that are created within private, permissioned blockchain or DLT-based systems or multi-lateral contractual frameworks will be treated as things in action by the law. ²⁶³Those things in action will therefore be different from, and will attract different legal treatment to, third category things.²⁶⁴

• (4)   Some third category things have an even closer relationship with the legal

system than crypto-tokens. Specifically, European Union carbon emission allowances (“EUAs”) and carbon emission allowances (“CEAs”) rely on statutory provisions for their continued existence, yet have been categorised by the courts as intangible things that are not a thing in action in the narrow sense.²⁶⁵ In our consultation paper, we said that EUAs would not satisfy our proposed criteria because their existence relies on statutory provisions.²⁶⁶ However, we no longer use our criteria to define the hard boundaries of the third category of thing to which personal property rights can relate and as such, our recommendation and conclusions do not seek to change the existing common law position.

Rivalrous

• 4.27    Our third criterion was that a thing must be rivalrous.

• 4.28    A thing is rivalrous if the use or consumption of the thing by one person, or a specific group of persons, necessarily prejudices the use or consumption of that thing by one or more other persons.²⁶⁷

• 4.29    Many consultees said that whether a thing is rivalrous is at the heart of the proprietary nature of the thing in question. ²⁶⁸Consultees also said that, where a thing is rivalrous (even if that thing is intangible), it is distinguishable from pure information. Clifford Chance LLP said:²⁶⁹

Rivalrousness seems to be a suitable concept for inclusion in the definition of the third category. It acts as a useful separator between pure information and cryptoassets. It follows from the reasoning given in Ruscoe v Cryptopia,²⁷⁰ with which we agree, in which cryptocurrencies were distinguished from other digital assets such as databases or photographs where the information could be “infinitely duplicated”. The idea of cryptocurrencies is to create an item of tradeable value, not simply to impart information.

• 4.30    Other consultees made similar points, although they noted that the concept is complex. Linklaters LLP said:²⁷¹

While we agree that “rivalrousness” is central to the question of whether or not something constitutes property under common law principles, this is a complex and nuanced issue and not something that is suited to definition in statute.

• 4.31    Moreover, in Tulip Trading, Lord Justice Birss said that cryptoassets such as bitcoin are things to which personal property rights can relate and endorsed the concept:²⁷²

As a result it is meaningful to describe bitcoin not merely as something which is transferable but as “rivalrous” (see the Law Commission’s recent Digital Assets: Consultation Paper). For a transferable thing to be rivalrous, the holding of it by one person necessarily prevents another from holding that very thing at the same time. Because the holder cannot double spend their bitcoin, such that it is rivalrous, the cryptoasset can be said to be capable of assumption by a third party (see the definition of property in National Provincial Bank v Ainsworth). ²⁷³Thus, as Mr Justice Bryan held in AA v Persons Unknown ²⁷⁴citing Ainsworth, a cryptoasset such as bitcoin is property.

• 4.32    While the concept has therefore been adopted as a matter of common law at Court of Appeal level, ²⁷⁵some consultees did offer minor criticism. The idea that a digital asset can be rivalrous is fundamentally important both to the market itself through solving the “double spend” problem²⁷⁶ and to the question of whether the digital asset can be a thing to which personal property rights can relate. As such, we deal with consultees’ criticisms below.

• 4.33    Professor Low’s response discussed the relationship between rivalrousness and control. ²⁷⁷To give context to this discussion, we summarise here two distinct but related concepts that we discussed in our consultation paper.²⁷⁸

• (1)    Fragile rivalrousness: Crypto-tokens are rivalrous by design. However, while crypto-tokens can be rivalrous, that quality could be thought of as “fragile” because it is contingent on the existence of robust technical authentication and validation mechanisms which ensure that the same crypto-token cannot be consumed twice or associated with two public addresses. That quality is also contingent on the continued active operation of the system by a network of participants.

• (2)    Dynamic excludability: The factual ability of a person to exercise control over access to a crypto-token (its excludability) will, to some extent, be determined by the continued efficacy of the system rules and the active operation of the system itself. Therefore, the practical ability to exclude others from a rivalrous crypto-token is likely to exist as a graduated quality that manifests on a continuum. Indeed, the factual ability of a person to exercise control over the access to a crypto-token (its excludability) might be dynamic over time. ²⁷⁹For example, the technical mechanisms from which a rivalrous crypto-token obtains its excludability could be degraded to such an extent that the crypto-token is no longer practically excludable. A hypothetical example would be if the hashing mechanism used to derive public keys from private keys failed or was rendered less effective by other technological advancements. In such a scenario, while the crypto-tokens within the system might still be rivalrous in that they could not be double-spent, ²⁸⁰their factual excludability could be degraded to such an extent that it was no longer possible to exclude others from the use or enjoyment of those tokens.

• 4.34    Professor Low and Peter Watts KC said that the discussion of fragile rivalrousness in our consultation paper implied that rivalrousness is not an absolute concept. ²⁸¹We disagree. Whether a thing is rivalrous is a binary question and we attempted to make this clear in our consultation paper.²⁸²

• 4.35    We recognise however that it might have been confusing to discuss the concept of fragile rivalrousness in the same section as dynamic excludability. As we say above, while rivalrousness is absolute, excludability is clearly not. And as we discuss in our consultation paper, ²⁸³excludability is not absolute in relation to tangible things. The threshold for excludability is not always determined by reference only to the factual characteristics of a thing. It also involves the exercise of legal and social value judgments.

• 4.36    It is for this reason that we distinguish our criterion that a thing must be rivalrous from those criteria that combine rivalrousness and control, or excludability.²⁸⁴

• 4.37    The concepts of rivalrousness and excludability are often intertwined.²⁸⁵ Sometimes, the concepts of rivalrousness and excludability are instead described or defined by a concept of (exclusive) control. For example, both the UNIDROIT Working Group²⁸⁶ and the UCC Committee define a digital asset as an electronic record which is capable of being subject to control.²⁸⁷

• 4.38    We consider that the two approaches are very similar and, in practice, are likely to lead to functionally similar results. ²⁸⁸However, we think it is worth repeating why we ultimately chose to focus on whether a thing is rivalrous.

• 4.39    First, discussion in terms of rivalrousness tends to concentrate the enquiry directly on the characteristics of the thing itself. Discussion of possession or control instead focuses attention on the ability to use the thing which is exercised by the person who holds it. Framing the question in terms of rivalrousness is analytically more direct. This is particularly helpful when considering digital assets, given the complex, technologyspecific and highly composable ²⁸⁹way in which control works with respect to particular digital assets.

• 4.40    Second, while rivalrous resources are likely to be (factually) excludable, not all excludable resources are rivalrous. Nor are all controllable resources rivalrous. A criterion that required either some level of exclusivity of control or some level of excludability would therefore need some additional element to narrow the scope of objects that satisfy the criterion. This is the function of limbs (ii) and (iii) in the UNIDROIT Working Group and the UCC Committee’s definition of control.²⁹⁰ And even those limbs do not exclude from the UNIDROIT definition things that are generally considered as non-rivalrous, such as Excel or Word files. ²⁹¹We discuss how we think rivalrousness excludes Excel or Word files (even where password protected) in our consultation paper.²⁹²

• 4.41    Third, there may be differing degrees of simultaneous control or use that can be made of a thing. For example, while a large sofa is rivalrous, more than one person can still sit on it at once.²⁹³ This nuance is reasonably easy to recognise where the discussion in framed in terms of rivalrousness. However, it is more complex to express where the discussion is framed in terms of (exclusivity of) control or excludability. Differing degrees of simultaneous control or excludability might exist, which might give rise to further definitional difficulties.²⁹⁴ Control and exclusion are not absolute concepts, unlike rivalrousness. Adams and McCormick say:²⁹⁵

The feasibility, cost, efficiency and equity of exclusion is separable from rivalry in consumption.

• 4.42    Moreover, (exclusivity of) control or excludability in the specific context of digital objects must be considered by reference to the technical characteristics of the digital object/system in question. As we discuss in Chapter 5 (Control), these technical characteristics can be difficult to describe or define and are not uniform across different third category things including digital objects. That makes defining a third category thing by reference to (exclusivity of) control or excludability much more difficult.

Rivalrousness in the context of third category things

• 4.43    To reiterate, we see the concept of rivalrousness working in the context of third category things (specifically crypto-tokens) as follows.

• 4.44    The criterion requires that it is possible to specify a rivalrous thing that is different from mere information and different from the physical medium on which that information is recorded. Otherwise, there exists no thing to which a personal property right can relate.

• 4.45    We say above that a crypto-token is best conceptualised as a notional quantity unit manifested by the combination of the active operation of software by a network of participants and network-instantiated data. That means that a particular crypto-token has a form and function that together imbue it with a rivalrous quality.²⁹⁶ A core function of the crypto-token is that it manifests the ability/power uniquely to perform an operation ²⁹⁷within the crypto-token system in respect of a certain notional quantity unit. Considered in this way, crypto-tokens are rivalrous. The use or consumption of the crypto-token by one person, or a specific group of persons, necessarily prejudices its use or consumption by one or more other persons. The technology ensures this. In other words, the technology ensures that a crypto-token cannot be double-spent.²⁹⁸ And while the data that gives rise to the crypto-token can be copied,²⁹⁹ its function within the relevant crypto-token system, as regulated by the inherent rules of the protocol, cannot. This means that crypto-tokens can be created such that they are mathematically scarce.³⁰⁰

• 4.46    A crypto-token is designed to achieve this functionality by the active operation of software by a network of participants, a point explicitly recognised by the Court of Appeal in Tulip Trading:³⁰¹

The signing of the hashed transaction record with users’ private keys in the first place, and the incorporation of these records into a hashed chain of blocks produced by the proof of work, solves the double spending problem. This characteristic of bitcoin does not emerge as a matter of law or convention, it is a characteristic which arises as a matter of fact from the way the software works. As a result it is meaningful to describe bitcoin not merely as something which is transferable but as “rivalrous”.

• 4.47    “Forks”³⁰² in crypto-token networks (broadly where subsets of network participants choose to run a substantively similar or the same protocol) further support the concept that a crypto-token is rivalrous. Even though each of the individual data elements of the crypto-token can be copied — in the sense that the information can be reproduced on an equivalent medium — the copier does not get the same discrete instance of a crypto-token. Instead, what the copier gets is data in a different system. In addition to pure information and mathematics, crypto-tokens rely on their respective protocol rules, real physical infrastructure, the work of humans and/or machines, energy expenditure, ³⁰³network effects, liquidity, and integration in existing social, economic or financial infrastructure.³⁰⁴ Each of these elements would need to be replicated in full to “copy” the thing that is a crypto-token. But even an exact recreation of all of the elements of a particular crypto-token would result in the creation of a materially identical, yet distinct, network, populated by materially identical but distinct, rivalrous crypto-tokens.³⁰⁵

DIFFERENT DIGITAL ASSETS — APPLYING TECHNOLOGY NEUTRAL PRINCIPLES IN A WAY THAT IS RESPONSIVE TO SPECIFIC TECHNOLOGY

• 4.48    A principal concern of consultees was that delineating hard boundaries of a third category of thing to which personal property rights can relate risked being either overly exclusive or overly inclusive. ³⁰⁶Consultees were concerned that the digital assets with which they were most involved might be incorrectly categorised by the rigid application of hard boundaries, which could lead to the incorrect application of existing common law principles. This concern was most apparent in relation to private, permissioned blockchain systems, ³⁰⁷various carbon emission allowances³⁰⁸ and in-game digital assets.³⁰⁹

• 4.49    Our recommendation and conclusions therefore do not seek to delineate hard boundaries of a third category of thing to which personal property rights can relate. Instead, we use our criteria (which we now treat as indicia) to describe a certain “core” type of assets — namely crypto-tokens manifested by distributed, public, permissionless systems — that are things to which personal property rights can relate at law and which are neither things in possession nor things in action. Rather than being determinative exclusionary criteria, therefore, our criteria can function instead as indicia of such things from which analogical, incremental categorisation can proceed. ³¹⁰This is no different from what already happens in relation to the established categories of things to which personal property rights can relate. Indeed, this was one of the questions in dispute in Colonial Bank v Whinney, ³¹¹in which the court considered whether shares in an incorporated company were “things in action” within the meaning of the Bankruptcy Act 1883. Given that we do not recommend the creation of hard boundaries for a third category, no new boundary issues should arise as a consequence of our recommendation. However, there will of course be things that might remain difficult to categorise. Courts will also have to determine whether new things can (and should) be capable of being objects of personal property rights at law.

• 4.50    We think that this allows the common law to remain technology neutral by ensuring that its principles are not based on any particular technology. Instead, the common law can apply technology neutral principles in a way that is responsive to specific technological forms, both now and in the future. For example, the mere use of a blockchain will not in itself be determinative as to the existence or otherwise of a third category thing. The nature and composition of the system and network that incorporate the blockchain will be the more relevant question and will therefore require reference to the specific technology used by the system and network.

Private, permissioned blockchain systems and different uses of blockchain technology

• 4.51    Some consultees ³¹²said that blockchain technology can be used to structure instruments and relationships in different ways, and that some of those structures might not necessarily give rise to distinct third category things.

Private, permissioned blockchain or DLT-based systems that do not manifest third category things

• 4.52    A principal example given by consultees was that of private, permissioned blockchain or DLT-based systems. Consultees made two broad points. First, that such systems might simply function as electronic registers or records of other things. As we said in our consultation paper, and as we discuss in more detail below, we think it possible that third category things could themselves be used merely to register or record information. ³¹³However, consultees also argued — and we recognise — that private, permissioned systems might not give rise to third category things at all. In some systems, the register or record would not be a third category thing at all and would instead simply be a digital record or piece of information, like a database entry.

• 4.53    Second, consultees said that many private, permissioned blockchain or DLT-based systems function as multi-lateral contractual frameworks (in direct contrast to distributed, public, permissionless blockchain systems that manifest crypto-tokens such as bitcoin or ether, which are not based on multi-lateral contracts). Those complex multi-lateral contractual frameworks might give rise to only contractual rights between the participants that are properly characterised as things in action (and might not manifest any third category thing). But equally, it remains possible that contractual or statutory rights could be used within systems that are intended to manifest third category things. ³¹⁴Some consultees were concerned that our proposed criteria were potentially too rigid to allow for or to recognise these distinctions, which could lead to mischaracterisation of some private, permissioned blockchain or DLT-based systems, or the tokens that they (do or do not) manifest.³¹⁵

• 4.54    We agree, and we recognise consultees’ concern. We think this concern applies equally to other attempts to define the hard boundaries of new digital assets. By way of comparative example, we think “records” within private, permissioned blockchain or DLT-based systems would be captured by the definition of controllable “electronic record” in both the UCC Committee’s Article 12 and the UNIDROIT Working Group’s Principle 2.³¹⁶ This means that such records would be things to which personal property rights can relate, broadly akin to third category things.³¹⁷ Indeed, this is the explicit intention.³¹⁸ We have chosen not to follow that path, due to, in part, the concerns raised in consultee responses.³¹⁹ As we discuss in Chapter 3 (Third thing), we conclude that the best answer to this difficulty is to maintain the existing common law position, with all the flexibility and nuance that comes with it. In short, the design of systems will affect the characterisation of the rights and/or relationships between participants in those systems and the (proprietary) characterisation of the tokens that they (do or do not) manifest.

• 4.55    That said, we also agree with the UKJT that cases involving permissioned, centrally managed, blockchains or DLT-based systems, where all participants are contractually bound to a common rulebook, are generally unproblematic and are unlikely to give rise to novel legal issues. ³²⁰The law is already clear in relation to pure information not being an appropriate object of personal property rights and in relation to things in action.

Different use-cases for third category things

• 4.56    Broadly speaking, commercial practice has over time resulted in certain types of document being elevated beyond a mere record of obligations in writing to something more — an embodiment of the right to claim performance of the obligations recorded in the document. ³²¹These types of documents can be issued either as “bearer documents” or as “order documents”. This determines the method of transfer used to transfer the document. ³²²Critically, the transfer of such a document does not require the consent of any other party, nor does it require any actions to be taken other than those described above.³²³ In the case of both bearer documents and order documents, the right to claim performance of the relevant obligation simply “travels with the document”.³²⁴

• 4.57    Some digital objects which are manifested by a distributed, public and permissionless system and are capable of transmission without the need for centralised intermediaries function in some ways like “digital bearer instruments”.³²⁵ As we conclude in Chapter 3 (Third thing), those digital objects are treated by the common law of England and Wales as third category things to which personal property rights can relate.

• 4.58    Such digital objects can be used in a variety of ways. Many such digital objects (often crypto-tokens) are not linked to anything external to the (crypto-token) system in which they are manifested. In crypto-token systems, for example, the crypto-token itself — as a distinct thing ³²⁶— is often the asset of interest or value. However, a crypto-token — as a distinct thing — can also be “linked” or “stapled” to something external to the crypto-token system, or indeed another crypto-token.³²⁷ For example, a crypto-token might purport to link to an intangible thing in action (like an equity or debt security, a contractual debt or a commercial licence), or to a tangible thing (like goods or land), or to another crypto-token or interest therein. A crypto-token itself — as a distinct thing — can also be used merely as part of a register or record (albeit a register or record composed of things to which personal property rights can relate, like the beads on an abacus).³²⁸

• 4.59    In the context of commercial financial markets, the UKJT made the same point and said that public, permissionless and decentralised blockchain or DLT-based systems could facilitate the issue of digital bonds in a number of different forms:³²⁹

• (1)   an issuer could use crypto-tokens on a blockchain or DLT-based system (with

each crypto-token being a distinct thing), instead of physical bearer instruments, creating “digital bearer instruments”; or

• (2)   an issuer could issue a bond using a blockchain or DLT-based system as a

register or record of interests instead of a conventional database (with each crypto-token being a distinct thing, but used simply to record or register something system-external).

• 4.60    The unavoidable technological reality is that some digital assets — specifically cryptotokens that rely on public, permissionless crypto-token systems — function in some ways like digital bearer instruments. A principal and fundamental design feature of these digital assets is that they are object-like in nature: they can be self-held or “self-custodied” by a holder and they are things that exist independently of any legal rights that they might embody or be linked to.

• 4.61    Nonetheless, it is important not to overstate the impact of that technological reality or the impact of the law recognising that such digital assets are things to which personal property rights can relate. As we discuss above, many digital assets do not function as digital bearer instruments and many private, permissioned blockchain or DLT-based systems might not even manifest third category things at all. Where such systems create only legal rights and obligations as between participants and the operator/administrator, existing legal principles currently applicable to things in action will apply.³³⁰ And even where the distinct features of third category things justify and require different legal treatment, in this report we conclude that the common law is perfectly able to achieve this. There are only two qualifications to this:

• (1)    Our recommendation in Chapter 3 (Third thing) for a statutory confirmation that a thing will not be deprived of legal status as an object of personal property rights merely by reason of the fact that it is neither a thing in action nor a thing in possession.

• (2)    Our recommendation in Chapter 8 (Collateral arrangements) that, as a matter of priority, the Government sets up a multi-disciplinary project to formulate and put in place a bespoke statutory legal framework that better and more clearly facilitates the entering into, operation and enforcement of (certain) crypto-token and (certain) cryptoasset collateral arrangements.

Increased optionality in legal structuring using third category things

• 4.62    Some consultees said that the distinction between third category things and things in action allows for greater legal structuring optionality, especially where things in action are linked to third category things. ³³¹They said that greater legal structuring optionality might lead to greater complexity in terms of the legal principles applicable to a particular structure. The legal analysis might be complex where a third category thing is used to embody or is linked to a thing in action, or where combinations of different technology and legal structures (such as direct holding and intermediated holding arrangements) are used.³³²

• 4.63    We agree with these observations, but think that increased optionality in legal structuring using third category things is an inevitable (and intentional) consequence of the creation of new technology. The alternative is for the law to ignore the conscious technological decision-making and fundamental design features of third category things and instead treat them as directly akin to things in action, limiting the legal structuring options available. Such an approach would simply not reflect the commercial expectations of market participants.³³³

• 4.64    Nevertheless, we also consider that the risk is small that the law of England and Wales — or market participants — will not be able to deal with this increased optionality in legal structuring using third category things. Even the hugely diverse crypto-token markets are dominated by a very limited number of crypto-token systems with large market capitalisations, with few dominant smart contract standards having emerged on each. Many decentralised finance market applications rely on the relative stability and inflexibility of such systems and token standards for their ongoing functionality. Similarly, where crypto-token markets involve intermediated holding arrangements, they replicate much of the legal structuring of the intermediated securities markets (albeit on a largely unregulated basis). The novel legal structuring optionality is that such markets can also facilitate self-holding/“self-custody” ³³⁴options and decentralised trading options (and sometimes users can quickly switch between the two). We expect that over time those markets are likely to mature and develop their own best practices, which are likely to be distinct from the intermediated securities market in some ways but replicate its features in others. We expect this will in part be driven by the products in question and their user base, in part by legal structuring optionality and in part by regulatory intervention.

• 4.65    In addition, while we accept that there is currently some residual legal uncertainty in the market, in this report we suggest that this is largely confined to highly nuanced and specific areas. We also consider that the market will, in general, gravitate towards legal structuring of arrangements where existing legal certainty is high, and develop and test edge-case products over time. Edge-case products which test areas of legal uncertainty are the likely places for litigation, which should eventually result in further common law development, leading to growing pools of legal certainty. Much remaining uncertainty will therefore be transient and will eventually diminish through the operation of markets. In addition, the areas of uncertainty that cause actual, practical issues which might give rise to litigation or might be worth litigating are hard to predict. For the law to ignore the distinct features of third category things based on theoretical uncertainty as to the future development of markets involving those things would be to prevent or stifle competitive innovation.

• 4.66    Our recommendation recognises that new technology is likely to continue to develop mechanisms for how value can be communicated online. Our recommendation simply allows the courts to continue to recognise that there is a separate third category of things to which personal property rights can relate, the boundaries of which can be incrementally developed over time. It permits the law of England and Wales to take the logically coherent, and flexible, position that different legal principles might then apply to different types of thing.³³⁵ This is of course likely to lead to greater legal structuring optionality.

Carbon emissions allowances and voluntary carbon credits

• 4.67    In our consultation paper, we discussed CEAs and VCCs. We provisionally concluded that CEAs did not satisfy our proposed criteria and therefore that they fell outside our proposed third category of things to which personal property rights can relate. However, we concluded that the common law already recognises that CEAs are things to which personal property rights can relate.³³⁶

• 4.68    Similarly, we provisionally concluded that most VCCs (as currently structured) did not satisfy our proposed criteria and therefore that they fell outside our proposed third category. ³³⁷Even so, we said that the prevailing view in most jurisdictions (including under the law of England and Wales) is that VCCs are “a form of intangible property” ³³⁸— they are capable at law of being things to which personal property rights can relate.³³⁹

• 4.69    We also said that other intangible things such as milk quotas, export quotas and waste management licences are capable at law of being things to which personal property rights can relate.³⁴⁰

• 4.70    A number of consultees, including International Swaps and Derivatives Association (ISDA), Linklaters LLP, Deloitte, Professor Cutts, Dr Crawford and Kings College London ³⁴¹said they did not agree with delineating hard boundaries of a third category things to which personal property rights can relate, because doing so risked excluding from the category things such as VCCs. ISDA explicitly tied this comment to a question as to the status of VCCs. They said:³⁴²

Defining the third category of property as “data objects” raises (without necessarily addressing) questions as to whether EUAs and VCCs constitute property at all, and if so, whether a fourth category of property exists under English law.

• 4.71    We agree with these observations. As such, we conclude that our criteria are best used as indicia to guide the incremental population of the third category of thing to which personal property rights can relate, rather than as hard exclusionary boundaries that, over time, are likely to become less effective. EUAs of the sort that formed the basis of the claim in Armstrong³⁴³ provide an illustration of this point.

• 4.72    In Armstrong, the EUAs in question were removed by one party from the claimant’s account and transferred to another, thereby depriving the claimant of their use. This demonstrates that they are both rivalrous and exist independently of persons. They are rivalrous because their use (by transfer) necessarily resulted in the claimant no longer being able to use or benefit from them. They exist independently of persons because they were clearly, in a factual sense, involuntarily alienable. A true thing in action such as a debt is not involuntarily alienable because a person cannot be divested of it without a legal process (which usually requires their consent, implicit or otherwise, such as via assignment or novation). As such, EUAs do not function like things in action.³⁴⁴

• 4.73    It might however be argued that EUAs, while fully independent of persons, are not fully independent of the legal system because their function and therefore their value lies in their (statutorily created) legal status and effect. But there is a difference between things or assets which are “causally connected to legal networks, services, and relationships”, and things or assets which are only legal relationships, existing only insofar as they are rights or claims enforceable by action.³⁴⁵ The significance of that distinction often manifests as the first type of thing being involuntarily alienable as a matter of fact. So, while EUAs are causally connected to legal networks (specifically a statutory framework), they do not exist as only a legal relationship. Whether this is true of other types of carbon emissions allowance will be context dependent, according to the way in which any particular allowance or credit is set up, configured and administered. EUAs therefore provide an example of “grey area” things which are not things in action and which, on an application of our indicia, are clearly more analogous to third category things.

• 4.74    We conclude in this report that it is important that the law of England and Wales can be responsive to specific technology and interrogate the idiosyncratic features of the asset in question when considering its proprietary status. We think this is equally true of carbon emission allowances, including VCCs. While the VCC market matures, it is important that each VCC arrangement can be considered on a case-by-case basis, by reference to the specific technology and legal structuring arrangements in question. Nonetheless, for completeness and to help increase legal certainty within the market, we consider that it is possible to structure VCCs such that they are capable of being a third category thing in themselves (even where VCCs are not “tokenised”). In that respect, we agree with ISDA that:³⁴⁶

VCCs can be seen as representing exclusive access to a finite resource - namely, an independently verified certification that the holder either directly or indirectly has reduced or removed from the atmosphere one metric ton of carbon dioxide equivalent, in accordance with relevant carbon standards and registry rules. They therefore constitute an intangible asset that is distinct from any underlying register in which entitlements to such VCCs are recorded.

• 4.75    We think that the law of England and Wales therefore provides a high degree of legal certainty with respect to VCCs. At the same time, it retains the inherent flexibility of the common law to be responsive to particular technologies when considering the question of proprietary status. Together, we think that this should help to cement the law of England and Wales as a highly competitive law of choice for structuring arrangements involving VCCs.

In-game digital assets and domain names

• 4.76    In our consultation paper, we said that, in general, email accounts ³⁴⁷and in-game digital assets (as they are currently constructed) are unlikely to satisfy our proposed criteria.

• 4.77    We said that, in general, email accounts and in-game digital assets are structured such that the account holder, or the player, has (contractual) rights against the service provider of the email account or the in-game digital asset. Moreover, we said that when considering an in-game digital asset, it is extremely difficult to point to a standalone thing that is rivalrous and that could be the object of personal property rights. Instead, the in-game digital asset exists as the result of a combination of infrastructure, intellectual property, and servers which enable a network of players to play together in the same ecosystem.

• 4.78    We went on to discuss a number of emerging use-cases where in-game digital assets are already testing the boundaries of the law of personal property, and the boundaries of whether and how rights relating to those in-game digital assets are properly characterised. We said that legal structuring in this area remains at an early stage of development. However, we anticipated that the ability of market participants to use digital assets that are object-like in some of their functionalities — such as NFTs — in novel ways to help govern, or as part of wider, online relationships will become increasingly important. In particular, we said that such digital objects might facilitate greater participation by users worldwide, the more efficient allocation to users of online resources that are considered valuable by market participants and online arrangements that have the potential to “fractionate existing power structures”.³⁴⁸

• 4.79    Digital assets that rely on public, permissionless and decentralised systems are particularly interesting in this respect because they allow the creation of global, standardised, user-controlled and potentially cross-application methods of communicating/transmitting value and recording provenance. ³⁴⁹We said that the law of England and Wales is well-placed to facilitate this experimentation, innovation and iterative development.³⁵⁰

• 4.80    Notwithstanding this, some consultees expressed concern that in-game digital assets might not satisfy our proposed criteria. Here, we discuss some of the issues raised by consultees.

• 4.81    First, a number of consultees said that NFTs or other digital assets with object-like qualities can now be used within in-game environments in a variety of ways. Games could permit self-holders of NFTs certain benefits (such as visual representations of clothing or weaponry within in-game environments). ³⁵¹The platform or provider could itself hold the NFT (possibly under some form of intermediated holding arrangement) with the user being given a temporary right of use. ³⁵²In-game environments could integrate with crypto-token systems to offer bespoke trading environments for specific NFTs or crypto-tokens, which themselves might be interchangeable with in-game currency or items that were more centralised in nature. ³⁵³We agree that these structuring options are possible under the law of England and Wales.

• 4.82    Second, some consultees said that certain in-game digital assets might exist in environments that could potentially give rise to things to which personal property rights can relate. Many in-game digital assets are constituted by a multimedia arrangement where a processing engine interacts with user devices, user accounts, a data store, a user and — perhaps — an operator.³⁵⁴ Deloitte, Meta, Professor Cutts, Russell-Cooke LLP and The Cloud Legal Project³⁵⁵ all said that there are potential combinations of these arrangements which could give rise to a rivalrous thing that should be capable of being an object of personal property rights. They said that in those situations, although legal agreements might be used as part of the overall multimedia arrangement, the in-game digital asset could still be considered as an independently existing thing. On this point, Professor Cutts said that there is a difference between assets which are “causally connected to legal networks, services, and relationships”, ³⁵⁶and assets which are only legal relationships, existing only insofar as they are rights or claims enforceable by action. Some consultees made similar arguments in respect of domain names.³⁵⁷ The argument is that even though these digital assets depend on legal relationships, they exist as more than the legal relationships themselves.³⁵⁸

• 4.83    Our conclusions in this report do not preclude this argument. Instead, we conclude that it is important that the law of England and Wales can be technology responsive and interrogate the idiosyncratic features of the asset in question when considering its proprietary status. That is as true of novel and evolving in-game digital assets as it is of crypto-tokens, VCCs and domain names.

Digital files and digital records

• 4.84    In our consultation paper, we provisionally concluded that digital files did not satisfy our proposed criteria and we did not propose that the law should treat them as things to which personal property rights can relate. However, we noted that, in future, digital files might be designed in such a way that they did satisfy our proposed criteria. We also said that market participants should be able to structure their arrangements such that things that do not fall within our proposed third category could be linked in some way to things that do. In this way, we considered that our provisional proposals allowed the greatest degree of flexibility for market participants to structure their arrangements as they choose.

• 4.85    Most consultees agreed. ³⁵⁹However, some consultees said that digital files are now so advanced in their construction that they are rivalrous and that logical access controls mean that a person can exclude others from, and so make exclusive use of, a digital file. The Cloud Legal Project said:³⁶⁰

Digital files are excludable by design through access controls at the logical layer. This typically includes an identity and access management system, often in the form of user accounts with associated privileges and passwords.

• 4.86    The Cloud Legal Project also said that the UNIDROIT Working Group concluded that password protected Word and Excel files fall within the definition of controllable “electronic record” in the UNIDROIT Principles, meaning that they are, under those principles, things to which personal property rights can relate.³⁶¹

• 4.87    We consider that the provisional conclusion in our consultation paper that digital files are not, in general, things to which personal property rights can relate remains correct. But we acknowledge that — as with other types of intangible things — it might be possible now or in future to construct digital files such that they are appropriate objects of personal property rights.³⁶² Much will depend on the specific technology and design of the digital file in question. ³⁶³For example, it might be possible now or in future to structure digital files so that they are rivalrous in a way that, historically, digital files were not.³⁶⁴

• 4.88    For completeness, we note that this area has attracted renewed academic interest since the publication of our consultation paper as digital files continue to develop from a technological perspective.³⁶⁵

• Chapter 5: Control

INTRODUCTION

• 5.1    The factual ability either to exclude or to permit access to a thing is fundamental to the law of personal property. As Professor Gray suggests, property rights have more to do with control over access to a thing than with enjoyment of the thing.³⁶⁶

• 5.2    In this chapter, we describe (but do not define) the factual concept that best captures the ability to (1) exclude or permit access to a third category thing; and (2) put a third category thing to the uses of which it is capable. We call this factual concept “control”.³⁶⁷ We go on to discuss the legal significance of the concept of control over third category things.

• 5.3   There is scope within the third category for a variety of different types of things. We

consider that the concept of control is likely to be appropriate for the vast majority of these third category things. However, we do not think that third category things should be defined by the concept of control directly. The concept of control might not always map neatly or consistently onto those third category things: control is highly complex, composable ³⁶⁸and multi-faceted in the context of third category things; and different technology, products and services use control in different ways.

• 5.4    As such, our principal (and only) recommendation in this chapter is that the Government creates or nominates a panel of industry-specific technical experts,³⁶⁹ legal practitioners, academics and judges to provide non-binding guidance on the complex and evolving factual and legal issues relating to control involving third category things such as digital objects (and other issues relating to digital asset systems and markets more broadly).

FACTUAL AND LEGAL CONTROL

• 5.5    Most consultees agreed that that the law of England and Wales should develop jurisprudence concerning a concept of control, as applicable to third category things.³⁷⁰

• 5.6    Many consultees also said that there is a useful distinction to be drawn between factual control and the legal consequences of that factual control.³⁷¹

• 5.7    We discuss both factual control and its legal consequences in more detail below. Two core areas of inconsistency emerged from consultee responses. First, consultees did not give uniform descriptions of factual control itself, and those descriptions were often different depending on the specific technology to which the response referred. Second, consultees did not all have the same view as to what should be the legal consequences of factual control when considered in the context of transfers, intermediated holding arrangements, collateral arrangements and causes of action and associated remedies in relation to third category things.

FACTUAL CONTROL

• 5.8    We describe control as a factual matter. Factual control functions as an important constituent element of more complex legal mechanisms or structures.

• 5.9    This mirrors the position taken by the UNIDROIT Working Group:³⁷²

The ... requirements ... contemplate that ‘control’ assumes a role that is a functional equivalent to that of ‘possession’ of movables. However, ‘possession’ in this context is a purely factual matter and not a legal concept. Moreover, because a digital asset is intangible, this functional equivalence to possession involves only the dominion and power over a digital asset but does not involve the physical situs dimension applicable to possession of movables. Whether ‘control’ . exists is a matter of fact and does not depend on a legal conclusion. However . the presence of control gives rise to legal consequences.

• 5.10    In our consultation paper, we said that, broadly speaking, the person in control of a [digital] object ³⁷³at a particular moment in time³⁷⁴ is the person who is able sufficiently to:

• (1)    exclude others from the [digital] object;

• (2)    put the [digital] object to the uses of which it is capable; ³⁷⁵and

• (3)   identify themselves as the person with the abilities specified in (1) to (2) above.

• 5.11    We explained our reasoning for the language in these limbs in detail in paragraphs 11.90 to 11.101 of our consultation paper.

• 5.12    Based on consultee responses, we do not recommend a statutory definition of factual control. Instead, we conclude that the common law is best placed to develop jurisprudence concerning a concept of control, for the following reasons.

Factual control over digital objects is complex and technology specific

• 5.13    While consultees broadly agreed with our high-level description of factual control,³⁷⁶ they set out a number of ways in which factual control can be more complex than our description recognises. Many consultees gave examples of situations in which factual control is not well covered by this description, because of how it is mediated by specific technological arrangements. Many consultees focused their discussion on digital objects, including crypto-tokens and NFTs, by way of example.

• 5.14    The most obvious and most-cited area of complexity was the (common) situation in which multi-signature arrangements are used in respect of digital objects such as crypto-tokens.³⁷⁷ Consultees also said that participant activities within crypto-token systems and/or DeFi arrangements can add complexity to the concept of factual control and gave examples related to mining, validator staking and associated arrangements, other types of non-validator staking, ³⁷⁸and the operation and deployment of smart contracts.³⁷⁹ Further situations which could give rise to added complexity include the receipt of unexpected (or unknown) crypto-token airdrops,³⁸⁰ and crypto-token system or network downtime.³⁸¹

• 5.15    This illustrates that digital objects in the third category give rise to a fundamental challenge that the law will have to acknowledge. That is, technology now facilitates the creation of intangible³⁸² things of value which are programmable, are rivalrous and which can be transferred, stored or traded electronically on permissionless and public global systems.³⁸³ Because this facilitating technology is often open-source, it is more likely than not that these types of digital object will proliferate over time, in number, use-case, design and technological functionality. ³⁸⁴If that happens, the law will find it increasingly difficult to maintain and uniformly apply rigid definitions to different types of technology or to different digital objects. We conclude that this is equally true of a broad, all-inclusive definition of control.

Rivalrousness is preferable to control when defining the thing in question

• 5.16    Both the UNIDROIT Working Group and the UCC Committee define the digital assets to which their provisions relate by reference to the concept of control. They use the principle of a controllable “electronic record”.³⁸⁵ Therefore, it makes sense that those provisions also define the concept of control. In contrast, and for the reasons we discuss in Chapter 4 (Third thing in practice), we decided not to define the digital assets in question by reference to control. Instead, we conclude that, when defining the thing in question, it is more appropriate for the law of England and Wales to focus on the rivalrous nature of a thing, rather than its excludability or susceptibility to (exclusive) control. This means that a rigid definition of control is not required. Instead, we treat control as an important constituent part of a relationship that a person can have with the thing in question.

Consistency with the law of possession

• 5.17    Our proposals treat control as a (factual) relationship that a person can have with a third category thing — an analogous (but distinct) concept to possession. In our consultation paper, we acknowledged the layers of complexity to the concept of possession. While we do not necessarily see this complexity as desirable in the context of third category things, it might be inevitable. But although it is complex, the concept of possession has remained a remarkably flexible tool for the law of England and Wales, able to deal with myriad things in possession and parties’ relationships with them. The law would almost certainly not have been able to develop the necessary nuance had possession been defined in legislation.³⁸⁶

• 5.18    We want to keep this same flexibility for third category things, and ensure that the law does not ask more of third category things than the law asks of things in possession. We consider that defining a concept of control in statute would risk undermining this position.

COMMON LAW DEVELOPMENT AND TECHNICAL EXPERT GROUP

• 5.19    We conclude that the common law will work as the principal driving force in developing accurate and nuanced jurisprudence concerning control that can apply to digital objects for the following reasons.

• 5.20    We think that the courts will turn to the concept of control as a matter of default. The broad concept of factual control can apply to both things in possession³⁸⁷ and to things in action, ³⁸⁸so ought equally to be able to apply to third category things as a matter of default (given that possession will not apply). The courts will also be able to draw on,³⁸⁹ if necessary, analogous case law in other jurisdictions, and international law reform initiatives, such as the UNIDROIT Working Group’s Control Principle, to help them develop the concept of control under the law of England and Wales. Indeed, the UNIDROIT Working Group does not explicitly recommend that Member States adopt a statutory definition of control: it instead frames its Control Principle as a broad guiding principle.³⁹⁰

• 5.21    We prefer this approach because it is most consistent with existing law. The common law has developed the principles of possession (and factual control) over time, in response to market developments and legal challenges. It is certainly able to do so again in the context of digital objects. In essence, we conclude that questions of factual control are essentially akin to any complex finding of fact, which courts are more than capable of reaching.

• 5.22    This will also allow legislative reform to focus on highly nuanced and market-specific areas of law such as regulating aspects of intermediated holding arrangements ³⁹¹or developing a statutory regime for the provision of (certain) crypto-tokens and (certain) cryptoassets as collateral. ³⁹²Those areas are likely to need to use the concept of factual control as a constituent element within a range of higher-level organising or framing principles. For example, the more appropriate focus of regulation in respect of custodial intermediated holding arrangements might be “safeguarding” assets rather than simply “controlling” them. ³⁹³And, as we discuss in Chapter 8 (Collateral arrangements), a perfection requirement for security interests could incorporate factual control as a core constituent element, but could be defined in terms of a more flexible, higher order framing principle, such as “the provision of collateral”.³⁹⁴ We think that innovation and incremental development in those areas might be hampered by prescribing a concept of factual control that could not, in itself, recognise the nuance and market-specificity of those types of complex arrangements.

A technical expert group on factual and legal issues relating to control and other issues relating to digital asset systems and markets

• 5.23    We recognise that the courts will still encounter difficulties in making complex findings of fact where control works differently for different third category things and through different technological implementations. In anticipation of those difficulties, we recommend that the Government creates or nominates a panel of industry-specific technical experts, legal practitioners, academics and judges to provide non-binding guidance on the complex and evolving factual and legal issues relating to control involving third category things such as digital objects (and other issues relating to digital asset systems and markets more broadly).

• 5.24    There are already many parallels with this approach. For example, the UKJT Legal Statement, although non-binding, was integral to the development of the law of England and Wales (and many other common-law jurisdictions) in respect of cryptotokens. The UKJT continues to innovate in this area: in 2021 it published the UKJT digital dispute resolution rules and in 2023 it published its Legal Statement on Digital Securities. It has proven that technology-specific, detailed, practical and industry-led legal guidance is fundamental to the facilitation and development of legal principles relating to novel technology. And, in this vein, Government has recognised the importance of a high-level industry group, the Cryptoasset Engagement Group, to help guide policy development.³⁹⁵ We think that either body could potentially take on the role of the technical expert group, although we think it should be a separate sub-committee/panel with additional members drawn from industry who have technical and practical experience of the crypto-token markets if it is to provide detailed, ongoing guidance. Alternatively, the technical expert group could be formed as a separate group and could potentially draw from both the UKJT and the Cryptoasset Engagement Group, as well as wider industry participants.

• 5.25    Similarly, in our report on Electronic Execution of Documents, ³⁹⁶we concluded that electronic signatures were valid for the vast majority of business transactions and legal processes. Nevertheless, that report also recognised that many uncertainties existed which have hindered the use of electronic signatures and limited the confidence of professionals and individuals in their use. For those reasons, we recommended that a multi-disciplinary group of business, legal and technical experts should be convened to consider the practical and technical issues involved, and to identify potential solutions. In response, an Industry Working Group was set up in 2021 and its task was to produce best practice guidelines and make proposals for further reform and development. That group published an interim report in February 2022³⁹⁷ and a final report in February 2023. ³⁹⁸We think that a similar approach would be helpful in the context of control, and other issues relating to third category things more broadly. Such guidance, while remaining non-binding, should be useful for a court that is considering a case involving similar issues. For example, in AA v Persons Unknown, Mr Justice Bryan referred to the guidance of the UKJT and said:³⁹⁹

It follows that the legal statement is not in fact a statement of the law. Nevertheless, in my judgment, it is relevant to consider the analysis in that Legal Statement as to the proprietary status of crypto currencies because it is a detailed and careful consideration and, as I shall come on to, I consider that that analysis as to the proprietary status of crypto currencies is compelling and for the reasons identified therein should be adopted by this court.

• 5.26    Consultees overwhelmingly agreed with our provisional proposal that a technical expert group would be beneficial.⁴⁰⁰ Consultees said that similar technical expert groups have proven helpful in other areas of law. ⁴⁰¹Our discussions with senior and specialist judges also indicated that the guidance of the technical expert group would be helpful for the development of principles of private law in this context.⁴⁰²

Possible scope and output of the technical expert group

• 5.27    We conclude that common law jurisprudence will be enhanced and made easier to understand for market participants by focusing on better descriptions and real-world examples of factual control. The technical expert group could describe the different factual ways in which control is used in the market today (and, crucially, update those descriptions as the market and technology evolves). It could undertake this exercise on the specific understanding that its output will be helpful for the judiciary when considering legal issues relating to control. It would therefore need to balance technical accuracy with accessibility for legal professionals.

• 5.28    A good pre-existing example is the set of Appendices to the UKJT consultation paper, which considered highly technology-specific issues, including the differences between UTXO-based and Account-based systems.⁴⁰³ Similar work would need to be undertaken for new and evolving token standards and technical implementations, and for the principal ways in which both public, permissionless crypto-token systems and private, permissioned blockchain systems manifest, manipulate and use control. Much of this work is already underway and is, as experience in other areas has shown, eminently achievable.

A control example

• 5.29    By way of specific example, control in the context of transfers within crypto-token systems is different for UTXO-based systems like Bitcoin and Account-based systems like Ethereum. Ethereum, for example, has a different way of: (a) signing a transaction operation to a Bitcoin transaction operation; and (b) composing the transaction data structure. Bitcoin transactions are composed from UTXOs that reference a particular way of unlocking/spending a notional unit of account — bitcoin/satoshis. They are composed from select UTXOs across the transactions on the blockchain which have a reference to a public address (as opposed to Ethereum transactions where a value is directly associated in an account). The UTXOs used as inputs to a transaction are usually determined by the wallet software being used.

• 5.30    And even within Bitcoin, control can work differently. For example, transactions can also be composed from discrete UTXOs (“coin selection”) and UTXOs can be encumbered in different ways (for example, through hashlocks and timelocks). Hash Time Locked Contracts are used for atomic swaps (that is, swaps between blockchains) and in payment channels (such as on the Lightning Network).⁴⁰⁴ These architecture differences and different forms of technical encumbrance can significantly change the analysis of control relative to the conception of a transaction as simply authenticating a data structure using a private key.

Other examples of control

• 5.31    Control is also used in different ways by different market participants and service providers. It would be useful for the technical expert group to describe practical, real-world examples of how control works in different circumstances.

• 5.32    Principal areas which we consider could be covered in due course include:⁴⁰⁵

• (1)    The factual operation of control within different types of (public, permissionless) Layer 1 protocols, such as Bitcoin, Ethereum, Solana and Avalanche.⁴⁰⁶

• (2)    The factual operation of control within different types of (private, permissioned) blockchain systems, such as Hyperledger Fabric and Corda.⁴⁰⁷

• (3)    The factual operation of control in relation to the principal token standards (common interfaces for contracts)⁴⁰⁸ deployed to different protocols. Well known examples are the ERC-20 and Solana SPL token standards, and the ERC-721 and Tezos FA2 token standards.⁴⁰⁹

• (4)    How factual control operates and is used within some of the principal smart contract protocols that are used for DeFi activities. Describing how this control works will help to distinguish the functionality of those systems from the functionality of existing traditional finance systems. Broad areas of focus could be:

• (a)    systems that mimic or recreate some of the functionality or results of traditional financial services such as “loan” markets (examples include Maker and Aave);⁴¹⁰

• (b)    systems that facilitate peer-to-peer, or decentralised, swapping or trading of tokens using automated market maker protocols (examples include Uniswap and Curve); ⁴¹¹and

• (c)    various “lock and mint” arrangements which involve relinquishing control over one form of crypto-token (the “locked” token), and the related receipt of control of a different form of crypto-token, which is the “minted” token.⁴¹² These types of arrangements are commonly used for varied and important purposes within DeFi ecosystems. Examples include “liquid staking derivative” systems related to validator staking (such as Lido and Rocket Pool); token bridges and wrapping protocols (such as the Wormhole bridge protocol and wrapped bitcoin); and NFT fractionalisation programs.⁴¹³

• (5)    How factual control operates and is used for different types of staking mechanisms, arrangements and services, including (but not limited to) direct, individual validator staking, delegated validator staking (or “staking as a service” mechanisms), pooled validator staking arrangements and “liquid staking derivative” systems related to validator staking (to the extent not covered above).⁴¹⁴ Examples of providers of some such services include Allnodes, Coinbase and P2P Validator.⁴¹⁵

• (6)    How factual control operates and is used within Layer 2 protocols, rollups and other “scaling” solutions, including “modular” blockchain systems. ⁴¹⁶Examples include Optimism, Arbitrum, zkSync, the Lightning Network, Cosmos and Celestia.⁴¹⁷

• (7)    How factual control operates and is used within wallet provider software, and how security features such as “social recovery” might change the operation of control in those circumstances. Examples of such software include Metamask and Argent.⁴¹⁸

• (8)    How factual control can be changed or manipulated through “account abstraction” techniques. That is, the addition of arbitrary logic to smart contracts to control crypto-tokens or other assets and perform transactions or state transitions that might change the functional role of private keys through principles of “account abstraction”.⁴¹⁹

• (9)    How factual control operates and is used or secured through physical hardware wallets. Examples include private key storage on internet connected devices such as computers or mobile phones, paper wallets, Ledger and Trezor hardware wallets and Opendime USB sticks.⁴²⁰

• (10)    How factual control operates and is used within intermediated holding arrangements, including both custodial intermediated holding arrangements and non-custodial intermediated holding arrangements (as we use those terms in Chapter 7 (Intermediated holding arrangements)). This might include how the books and records of such holding intermediaries typically reflect and record crypto-tokens and crypto-token entitlements, how such holding intermediaries themselves secure tokens, the interrelationship between “hot”, “warm” and “cold” storage and how any “sweeping” or transfers between such storage occurs. Examples of holding intermediaries (often referred to as crypto-token exchanges) include Coinbase, Binance and Kraken.⁴²¹

• (11)    How other providers of software-as-a-service in non-holding relationships manipulate factual control. Such services might be provided by service providers who do not themselves hold tokens on an intermediated basis but do, for example, “safeguard” private keys or parts thereof. Examples include Copper and Gnosis Safe.⁴²²

• 5.33    These questions are not hypothetical. For example, the High Court of England and Wales recently had to consider some of the legal consequences of token bridges and wrapping protocols in the context of claims and applications for injunctive relief following a hack of the Wormhole bridge protocol.⁴²³ And in Piroozzadeh v Persons Unknown the court had to consider the operation of non-custodial intermediated holding arrangements provided by a crypto-token exchange.⁴²⁴

• 5.34    The purpose of the technical expert group considering these different types of control is to provide the legal market and the judiciary with a common toolkit and factual reference point from which to develop clear and consistent legal analysis. Rather than relying on inaccurate analogies (such as an analogy with a physical object in a warehouse, the door to which has multiple keys; or a traditional bank account; or a bag of gold), market participants and courts will benefit from an analysis of the actual technology in question.⁴²⁵ This will force claimants and defendants more accurately to disclose and present their cases, which should lead to clearer, more logical and more consistent applications of legal rules and reasoning over time. This is fundamental to our conclusion that the common law is the most appropriate mechanism for legal development with respect to the nuanced features of digital objects. A principal goal of the technical expert group could be to maintain a repository of material that considers different factual scenarios and ensures that technological analysis is in line with existing market practice and proposed future developments.

• 5.35    Our recommendation is based on our conclusion that the law of England and Wales must remain sufficiently flexible to recognise and respond to the idiosyncratic features of digital objects. Specifically, it must recognise that control over third category things is manifested and mediated by software and works differently to control over things in possession and things in action. To ignore the fundamental design-features of third category things (particularly crypto-tokens that exist by reference to public, permissionless and distributed crypto-token systems) risks the law evolving at odds with the expectations and intentions of market participants. Instead, it is better for the law to accept the reality of modern technology and to adapt to it in a way that is responsive to the peculiar features of different technology, with the help and guidance of experts.

Recommendation 2.

• 5.36    We recommend that the Government creates or nominates a panel of industryspecific technical experts, legal practitioners, academics and judges to provide nonbinding guidance on the complex and evolving factual and legal issues relating to control involving third category things such as digital objects (and other issues relating to digital asset systems and markets more broadly).

• 5.37    This panel would need to include those with expertise in the crypto-token markets, and not just those with expertise in traditional finance markets or intermediated securities markets.

THE LEGAL CONSEQUENCES OF CONTROL

• 5.38    In our consultation paper, we said that the legal consequences of control are complex and varied. Factual control operates as a composite part of more complex legal principles and mechanisms (such as transfers, intermediated holding arrangements, collateral arrangements and causes of action and associated remedies in respect of digital assets). We described control by reference to a subset of digital assets — those that satisfied our proposed criteria — “data objects”. However, as we discuss in Chapter 3 (Third thing), consultees said that the criteria for “data objects” could create unnecessary boundary issues. In part, this is also likely to have informed consultees’ approaches to the potential legal consequences of control.

• 5.39    Two broad arguments emerged from consultee responses. On the one hand, a first group of consultees said that control should have similar legal consequences for crypto-tokens (and other third category things) as possession does for tangible things, in various respects.⁴²⁶ But many of those consultees said that there is merit in treating control as a distinct concept from possession, because:⁴²⁷

It allows for subtle but important differences to develop without diluting or distorting the concept of possession as a physical construct or automatically importing all the “baggage” that comes with the concept of possession.

• 5.40    On the other hand, a second, smaller group of consultees preferred to treat control in respect of crypto-tokens (and other third category things) as having similar legal consequences to control over things in action.⁴²⁸ Any jurisprudential development of control ⁴²⁹to recognise the bearer-like features of third category things was seen by those consultees as developing a concept of control that shared many features of possession, which they considered undesirable.⁴³⁰

• 5.41    Broadly, the position of the first group of consultees is perhaps more reflective of legal issues relevant to the effective operation of markets in — or use-cases that draw on the particular characteristics and functionalities of — crypto-tokens manifested by public, permissionless crypto-token systems. ⁴³¹The position of the second group is perhaps more reflective of legal issues relevant to the effective operation of applications that utilise the capabilities of private, permissioned blockchain systems.⁴³² This dissonance is compounded further by the fact that the crypto-token and cryptoasset markets make use of two distinct structuring mechanisms: one based largely on traditional intermediated securities markets and one based on the technological capabilities of crypto-token systems (which partly reflects some of the features of the market for negotiable bearer instruments). ⁴³³Often these structuring mechanisms are used interchangeably. For example, many holding intermediaries offer both self-holding or “self-custody” holding arrangements for crypto-tokens and intermediated holding arrangements (either custodial or non-custodial) and users can quickly and easily transfer between the various options (often at no cost).

• 5.42    The view of the first group of consultees was more holistic in that it sought to accommodate both public and private system applications (while recognising their differences). The second group of consultees focused more on private, permissioned blockchain systems and existing intermediated securities structures and markets and sought to explain the legal operation of public, permissionless crypto-token systems and crypto-token markets exclusively by reference to those structures and markets.⁴³⁴

• 5.43    Consultees therefore had fundamentally different views, which broadly centred around their conceptualisation of the digital asset in question and the particular technology with which they were most concerned. As we discuss in Chapter 4 (Third thing in practice), those different views were also apparent in consultees’ concerns about creating hard boundaries of a third category of thing to which personal property rights can relate.

• 5.44    Despite this difference in approach, consultee responses were aligned on the legal consequences of control in most circumstances. Below, we describe the areas where consultees were aligned and then highlight the remaining differences. We return to those differences in later chapters on transfers, intermediated holding arrangements, collateral arrangements and causes of action and associated remedies. Many consultees focused their discussion on digital objects, including crypto-tokens and NFTs, by way of example.

The legal consequences of control — areas of alignment

Situations in which “ownership” and “control” are separated

• 5.45    Almost all⁴³⁵ consultees agreed that there are situations in which superior legal title (sometimes referred to as “ownership”)⁴³⁶ to a digital object can be separated from (factual) control over that digital object. Consultees gave a variety of practical examples, including void transfers, some types of intermediated holding arrangements and agency relationships.

Defects in transfers rendering the transfer void

• 5.46    Some consultees gave examples of defects in the transfer of a digital object that would render the transfer void at law. Examples included situations where a thief steals a digital object outright,⁴³⁷ where there has been a fundamental mistake in a transfer, ⁴³⁸or when the transferor lacked the mental capacity to make a valid legal transfer.⁴³⁹

• 5.47    In those scenarios, the transferor will, in general, retain legal title to the digital object, but the transferee will obtain control over the digital object. This has been recognised or referred to by the courts in cases including Tulip Trading (in relation to bitcoin),⁴⁴⁰

Osbourne (in relation to NFTs), ⁴⁴¹and Armstrong (in relation to EUAs). ⁴⁴²We discuss this in more detail in Chapter 9 (Remedies).⁴⁴³

• 5.48    In these scenarios, the level of factual control over a digital object that a thief or a transferee in a void transaction obtains is the exact same level of factual control that a transferee would obtain in a voidable transaction or in a valid transaction. It is the operation of the law of derivative transfers of title and other principles of private law, rather than differences in the level of factual control obtained, which determine the legal consequences. We return to this point in more detail at paragraph 5.71 below, because not all consultees gave uniform responses on what should be the legal consequences of the factual situation.

Intermediated holding arrangements and agency

• 5.49    Some consultees said that intermediated holding arrangements (and agency) are a good example of where superior legal title to a digital object might be separated from factual control over the digital object.⁴⁴⁴

• 5.50    We discuss various intermediated holding arrangements, non-holding arrangements where control is manipulated, and agency in detail in Chapter 7 (Intermediated holding arrangements). In complex intermediated holding arrangements involving digital objects, the level of control obtained by the holding intermediary will be an important factor but might not be determinative of the legal consequences of the arrangement.⁴⁴⁵ It is not the fact of control alone, but its implications within the law of personal property, contract, trust, agency and representation which determine the legal consequences of the arrangement.

• 5.51    This is an area in which the technical expert group could give practical guidance, referencing real-life technological implementations and intermediated holding arrangements (involving both public, permissionless crypto-token systems and private, permissioned blockchain systems).

Transfers by a change of control

• 5.52    In our consultation paper, we provisionally proposed that the law should be clarified to confirm that a transfer operation that effects a state change⁴⁴⁶ is a necessary (but not sufficient) condition for a legal transfer of a crypto-token. However, we also considered the possibility that a potentially wider condition could apply — namely a “change of control”.⁴⁴⁷

• 5.53    Consultees were broadly aligned in arguing that it was possible as a matter of law to effect a transfer of a digital object by a change of control as opposed to the potentially more limited transfer operation that effects a state change.⁴⁴⁸ Many consultees said that intention would be a necessary element in this change of control analysis.⁴⁴⁹

• 5.54    We discuss transfer operations that effect a state change, transfers by a change of control, derivative transfers of title and the development of a common law special defence of good faith purchaser for value without notice for third category things in detail in Chapter 6 (Transfers).

Intention

• 5.55    In our consultation paper, we considered the factual concept of possession applicable to tangible things by analogy with the concept of control applicable to third category things including digital objects. There are two elements to the concept of possession: (1) a sufficient degree of custody and control; and (2) an intention to exercise such custody and control on one’s own behalf and for one’s own benefit. ⁴⁵⁰We said that the intention element of the concept of possession is potentially harder to apply in the digital or online context than in the physical world. ⁴⁵¹This is particularly the case when questions of intention relate to systems, technology and digital objects that rely on the deterministic operation of software or computer code.

• 5.56    Notwithstanding this difficulty, consultees said that intention is likely to remain important in determining the legal consequences of control in the context of transfers of a digital object (either by a transfer operation that effects a state change, or by a change of control).⁴⁵²

• 5.57    We agree. We conclude that intention should form part of the jurisprudential consideration of how control applies to third category things including digital objects.

However, we re-iterate our caution that always to search for an element of human intentionality risks introducing an unreal human element to what can often be automatic or deterministic processes.⁴⁵³

• 5.58    We conclude that the common law will again need to develop such that intention is considered in the context of the specific technology or application in question. Consultees said that this is already the case in other areas of law where intention is an objective test that is equally capable of being applied in the context of automated processes.⁴⁵⁴

• 5.59    Equally, the common law will need to develop principles that allow it to recognise how control (and intention) are used as part of processes and arrangements within DeFi systems. Linklaters LLP explained this point well:⁴⁵⁵

If there is no person that can be found to have the requisite intention, it may be appropriate to conclude that there is no person with Control of the asset and/or that a particular arrangement has given rise to a relinquishment of Control rather than a transfer of Control. This could potentially be the case in certain circumstances where crypto-tokens are deposited in certain token contracts governed on a completely decentralised basis or with addresses in respect of which no person has knowledge of the private key. We do not see any policy rationale for excluding that possibility.

• 5.60    We conclude that intention will help to resolve disputes where legal consequences flow from control ⁴⁵⁶over a digital object. For example, as Linklaters LLP noted, cryptotokens can be airdropped ⁴⁵⁷to public addresses over which a person has factual control but where that person had no knowledge of the existence of the tokens and/or had no desire to use them. If that change in factual control resulted in a transfer of superior legal title or a control-based legal proprietary interest, the person could potentially suffer adverse tax, regulatory or criminal consequences, where such consequences flow from “ownership” or “control”.

• 5.61    Again, considering how intention might be analysed in a range of different real-world factual scenarios could be an important function of the technical expert group.

Perfection requirement for qualifying non-possessory security interests

• 5.62    In Chapter 8 (Collateral arrangements), we consider how collateral arrangements in respect of crypto-tokens and cryptoassets can be structured under the law of England and Wales. We conclude that although the law of England and Wales does provide options for granting security in respect of crypto-tokens and cryptoassets, those options are not adequate. As such, we recommend that, as a matter of priority, the Government sets up a multi-disciplinary project to formulate and put in place a bespoke statutory legal framework that better and more clearly facilitates the entering into, operation and enforcement of (certain) crypto-token and (certain) cryptoasset collateral arrangements.

• 5.63    As part of that recommendation, we discuss the concept of “provision” — a flexible and responsive perfection requirement for qualifying non-possessory security interests that incorporates factual control as a core constitute element but is not defined exclusively by reference to factual control.⁴⁵⁸

• 5.64    We think that this concept of “provision” could be more flexible and better able to deal with the complex way in which collateral arrangements work in respect of digital objects. As the CLLS-FLC argues, it is useful to distinguish control for different purposes, and defining control in legislation (as opposed to a more flexible concept, such as provision), might not facilitate this:⁴⁵⁹

While the concept of control is prima facie attractive, unfortunately English law precedent in the context of the Financial Collateral Arrangements (No 2) Regulations 2003 ("FCARs"), has declared that the concept of control (for the purpose of perfecting a financial collateral arrangement) has the same [or similar] characteristics as when the term is used in English law to distinguish a fixed charge from a floating charge [subject to certain qualifications of uncertain scope]. Lesser measures of factual or legal control, which are almost inevitable in the context of the practical operation of control arrangements in a blockchain or DLT-based system, would run the risk of not being recognised by the English courts as constituting sufficient control to found a legal or equitable proprietary title to a relevant digital asset (in the third category). And this area would remain unsatisfactory with regard to taking security over [third category things], as well as over other forms of financial collateral (i.e. financial instruments, cash and credit claims).

• 5.65    For those reasons (and as we discuss in more detail in Chapter 8 (Collateral arrangements), we model our concept of “provision” on CLLS-FLC’s proposed changes to the FCARs to deal with existing uncertainties in the context of financial instruments and cash.⁴⁶⁰

The difficulties with rigid legislative definitions

• 5.66    Where the concept of control appears in legislation and is then interpreted by the courts, it has sometimes led to further confusion. ⁴⁶¹An example is the Financial Collateral Arrangements (No 2) Regulations 2003 (FCARs). ⁴⁶²Professors Fox and Gullifer said in their response to our call for evidence on digital assets that:

The concept of control, as [used] by the FCARs and the courts, has caused a great deal of difficulty.

• 5.67    We consider that defining the concept of control in legislation could potentially increase uncertainty in the context of digital objects. This is for three principal reasons.

• 5.68    First, defining control in legislation could either risk undermining the law’s flexibility to apply control to different technologies, or risk the definition becoming so vague and subject to so many carve-outs that it becomes too complex or practically useless (or both).

• 5.69    Second, in the context of digital objects, control is likely to be an evolving concept. For example, in the context of security features, users currently manipulate control through physical security, hardware wallets, multi-signature arrangements, software-as-a-service based security features or intermediated holding arrangements. But new safety and security arrangements continue to be designed and tested. For example, social recovery wallets and account abstraction techniques are now available.⁴⁶³ Defining control in legislation risks crystallising the concept by reference to technology available today.

• 5.70    Third, we conclude that the concept of control is often most useful as an important constituent element of higher-level organising or framing principles. For example, and as we discuss in Chapter 7 in relation to intermediated holding arrangements, the concept of control might not be the only determining factor in the legal analysis of the arrangement. Similarly, in Chapter 8 (Collateral arrangements) we discuss how a concept of “provision” ⁴⁶⁴might improve on the concepts of possession and control used in the FCARs.

The legal consequences of control — remaining differences

• 5.71    Consultee responses were therefore broadly aligned on how superior legal title and factual control over a digital object can be separated, on how a change of control can effect a legal transfer of a digital object and on the importance of intention to the analysis. Those responses that considered the issue were also broadly aligned on the difficulties with using control as a standalone and determinative principle for the purposes of a flexible and responsive perfection requirement for qualifying non-possessory security interests.

• 5.72    However, below we identify remaining differences in the views of consultees, which focus on one principal conceptual issue. That is whether the law should treat a certain level of control of a digital object as giving rise to a legal proprietary interest in that digital object (which can, in some circumstances, be a lesser interest than a superior legal title).

• 5.73    Below, we explain the differences in more detail and conclude on what we consider to be the better view. We accept that it is likely not possible to reconcile the difference in approach between consultees.

Relative title

• 5.74    Almost all consultees agreed that superior legal title and control over a digital object can be separated. However, a small number of consultees ⁴⁶⁵said that control over a third category thing should not be capable of founding some form of legal proprietary interest in the thing short of (that is, lesser than) a superior legal title. On the other hand, a greater number of consultees⁴⁶⁶ said that treating third category things in the same way as things in action (that is, as incapable of a relative title analysis) implicitly limits the flexibility of the common law to develop legal principles that are best suited to third category things.

• 5.75    Below, by way of example, we describe some scenarios in which the superior legal title to a crypto-token ⁴⁶⁷can be separated from either: (i) the recorded state of the distributed ledger or structured record (for example, where there is an offchain transfer by a change of control); or (ii) a factual controller of the crypto-token (for example, following a void transfer). We also explain that it is common for a person to apply technical encumbrances and conditions within a crypto-token system to regulate how a particular crypto-token can be controlled. Examples include multi-signature arrangements, and some DeFi arrangements such as validator staking and other types of non-validator staking, and the operation and deployment of certain smart contracts.

• 5.76    We conclude that disputes are likely to arise as to the legal consequences of a person having factual control over a crypto-token (and the actions taken, or inaction, by such controllers). Some particularly difficult scenarios could involve circumstances where the person with control:

• (1)    is not the same as the person that is recorded within the state of the cryptotoken system as having control (offchain transfers of control could give rise to this scenario);

• (2)    is not the person with superior legal title; and/or

• (3)    exercised some level of joint, split, or otherwise encumbered factual control and took (or failed to take) certain action(s).

• 5.77    We think that some of those disputes might need to be resolved by reference to “title”. For these purposes, we adopt the explanation of Professor Fox that:⁴⁶⁸

Title is best understood as referring to a claim to an asset arising from a proprietary interest. In its strict sense, title does not refer so much to the content of the claimant's proprietary interest — i.e., to the incidents of practical enjoyment which his interest in the asset confers on him — but to the strength of his claim to that enjoyment, relative to other people who have similar interests in the same asset. Inherent in the notion of title is the question whether one person's claim to those incidents is stronger or weaker than that of other potential claimants. Professor Goode has expressed the point as depending on the difference between the quantum of the claimant's right to the asset and the strength of that right as against others...

To summarise then, relativity of title is about the respective enforceability of competing claims to vindicate the incidents of enjoyment inherent in some particular proprietary interest.

• 5.78    The fundamental difficulty with third category things is that they are neither things in possession (to which principles of relativity of title apply) nor are they things in action (to which principles of relativity of title do not apply).

• 5.79    Professor Fox continues:⁴⁶⁹

The relativity principle has no part to play in determining competing legal claims to [things] in action, except where the [thing in action] is embodied in a documentary intangible or in those rare instances where the title arises from an unauthorised substitution of an original asset. In general, the relativity principle illustrates the pragmatic interaction in common law reasoning between the rules of evidence and procedure, and the substantive principles governing the creation and transfer of property interests.

• 5.80    We agree with the view of Professor Fox and the authors of The Law of Personal Property that to conceive of a sort of relative title to things in action — rights or claims enforceable by action — is “a concept which is alien, illogical and contrary to authority”.⁴⁷⁰

• 5.81    But third category things are different to things in action, as this report makes clear. Almost all consultees recognised that superior legal title and factual control over third category things can be separated. And, in the context of crypto-tokens, almost all consultees acknowledged that a controller might not be the same person as recorded by the state of the crypto-token system ⁴⁷¹and that control can be joint, split, or otherwise encumbered in a variety of different ways.⁴⁷² Once this is recognised, a fundamental question remains as to whether control can found a legal proprietary interest in a third category things such as a digital object which, in certain circumstances can be separated from (and be inferior to or short of) a superior legal title.

• 5.82    We conclude that it can, for the following reasons.

• 5.83    First, there is existing (albeit criticised)⁴⁷³ authority ⁴⁷⁴that the common law of England and Wales does not absolutely exclude the possibility of concepts of relativity of title applying to third category things.⁴⁷⁵

• 5.84    Second, regardless of the way in which factual control is acquired, it gives the person in control many of the incidents of a proprietary interest.⁴⁷⁶ The controller will obtain the factual ability to (1) exclude or permit access to a third category thing; and (2) put the third category thing to the uses of which it is capable. If an innocent acquisition rule applies, a controller will also be able to transfer good legal title to an innocent acquirer. Under the article 12 of the UCC and principle 16 of the UNIDROIT Principles, a secured creditor that perfects its security right by control will take priority over a creditor that perfects it security right by another method.⁴⁷⁷ It is odd for the law to recognise this factual state of affairs and to acknowledge that some legal consequences (such as perfection of a security interest) can flow from that factual state but not describe it as giving rise to some form of legal interest or “title”. ⁴⁷⁸We conclude that the better view is that the law should recognise the factual state of affairs and that control does give the controller some form of legal proprietary interest — and therefore some form of title — to the third category thing in question. The nuance on the relative strength of that title will remain to be determined by other legal principles. For example, the relative strength of that title will depend on the way in which factual control was acquired. A lawful acquisition would in many circumstances give the acquirer the superior legal title to the asset (good against the whole world), while factual control acquired under a transaction that is void or has been avoided would be good against the whole world except the person from whom control was acquired. This is a function of the system of relative title, which is capable of recognising multiple concurrent interests in the same asset.

• 5.85    Third, in this report we distinguish things in action from third category things and conclude that things in action function differently to third category things. Our conclusions are based on our view that the law of England and Wales must remain flexible enough to recognise and respond to the idiosyncratic features of third category things. Specifically, the law must recognise that control over third category things is manifested and mediated by software and works differently to control over things in possession and things in action. We said above that control gives a person in control of a third category thing many of the incidents of a proprietary interest. This is a feature of the way in which third category things are designed.

• 5.86    Market participants intuitively understand this and expect to be able to “self-hold” or “self-custody” their crypto-tokens. ⁴⁷⁹Crypto-token systems utilise software which permits the imposition or creation of varying degrees of technical encumbrances in respect of the crypto-token. ⁴⁸⁰Imposing or creating such technical encumbrances within a crypto-token system makes it possible for a person to have a factual relationship of control ⁴⁸¹with a crypto-token. Crypto-token systems and the

decentralised finance markets which have evolved around them recognise this core feature and allow the layering of further technical encumbrances, including the splitting or distribution of control. These markets and many related products are also more often designed by reference to control and the powers and incentives/incentive mechanisms of participants, rather than in terms of claims/rights, corresponding duties and obligations.⁴⁸²

• 5.87    To treat third category things as directly akin to things in action would be to ignore the core underlying reason why such third category things were created in the first place and to overlook their principal functionality, both in terms of their distinguishing practical purpose and their technical operation. Moreover, it would risk blunting their unique functionality by directly applying legal concepts that evolved solely by reference to things in action that do not have that functionality. We conclude that this would be a retrograde step for the law of England and Wales. While the law might benefit from increased certainty in the short-term, its lack of flexibility and its failure to recognise the defining features of new third category things would eventually lead to reduced international competitiveness. This is particularly so in the face of law reform initiatives worldwide which do recognise the defining features of new digital objects.⁴⁸³

• 5.88    Fourth, in the context of third category things, it is important that the law permits market participants to make assumptions as to the superior legal title of a controller of a third category thing. This is most obviously important for onchain transfers, where a recipient may be unaware (or unable) to find evidence as to the superior legal title conferred by an (anonymous or pseudonymous) chain of conveyances. It might also be helpful for offchain transfers: without more a recipient ought to be able to rely on the presumption of a superior legal title from the fact of control (plus intention). Professor Fox suggests (albeit in the context of things in possession and land) that when supported by rules on the limitation of actions, ⁴⁸⁴a presumption of title by possession helps to reduce the risk of competing claims, whether spurious or well founded. ⁴⁸⁵This is also likely to be true in the context of third category things, where the risk of competing claims could otherwise act as a disincentive to transfers.

• 5.89    Fifth, relativity of title is important because in the absence of a conclusive system of title registration, it may be difficult for a claimant to prove that they have the superior title to the disputed thing. As we said in our consultation paper and as we discuss in more detail in Chapter 6 (Transfers), we do not consider that crypto-token systems operate as conclusive legal title registration systems. The functionality that they manifest relates to control. To force a claimant to establish that they are the superior legal title holder could require an extremely high level of proof: that there is no other person with a title superior to theirs.⁴⁸⁶ This is unlikely to be possible or practicable in crypto-token systems which permit use by anonymous or pseudonymous persons, facilitate complex relationships of control and interactions with smart contract systems or DeFi protocols and which are global, public, open and permissionless in nature. What participants are likely to be able to demonstrate however, is some form of control.

• 5.90    Sixth, as Professor Fox argues (albeit again in the context of land), the rule that allows a person to sue on a merely relative title to land shows the preference of the common law for discouraging the social disruption that comes when one person unjustifiably interferes with the settled possession of another. The presumption of title in a mere controller of a third category thing could be very helpful for many of the difficult cases involving wrongdoers and new technology.⁴⁸⁷ For example, there seems little sense in a legal system which does not permit a holding intermediary who holds digital objects on behalf of a large number of users to pursue a hacker who wrongfully removed assets from the control of that holding intermediary. ⁴⁸⁸Indeed, one might consider that large, well-funded holding intermediaries would be better placed than individuals to pursue such actions.

• 5.91    Seventh, as we discuss in more detail below and in Chapter 9 (Remedies), allowing for a relativity of title analysis will assist with the protection of the personal property rights of individuals in relation to their things (including their third category things). This might be important in relation to unlawful or wrongful interference (whether by a hacker or by a government or other legal body acting beyond their powers).

• 5.92    Eighth, to accept this position in relation to third category things is not simply to apply the rules relating to things in possession. Instead, the common law will need to develop rules for the resolution of disputes between claimants who assert equivalent proprietary interests in third category things. We think that it is well placed to do so.

Shared control

• 5.93    The question of who has factual control might not always be easy to determine. Many consultees gave examples of where this could be the case (see from paragraph 5.13 above).

• 5.94    In some (but not all) of those situations, multiple parties might share control, no single party might have control, or control might be split in a variety of different ways. Sometimes this is achieved by reference to the protocol rules of a crypto-token system itself, and sometimes through additional software that is used externally to a cryptotokens system (often for security reasons).⁴⁸⁹

• 5.95    However, we do not think these issues are insurmountable for the law.

• 5.96    As we discuss above, the concept of intention will be a helpful factual element which will likely assist courts in distinguishing between the “relativity” or “priority” of controlbased legal proprietary interests, particularly where parties are acting in concert or share control. As with intention in other areas of the law, intention can readily be inferred by analysing the contractual and other arrangements between the parties, including by looking to the laws of agency and representation.⁴⁹⁰

• 5.97    We also note that, as a purely practical matter, where factual control is shared by two (or more) competing parties, in practice one party will most likely deploy a self-help remedy, by effecting a transfer operation that effects a state change, thus extinguishing the factual control of the other competing parties. ⁴⁹¹But in those situations, distinguishing between the “relativity” or “priority” control-based legal proprietary interests⁴⁹² will still be important as doing so will determine whether any causes of action exist against the party that effected the transfer operation, or whether any remedies are otherwise available.

• 5.98    More broadly, we acknowledge that similar issues around shared control apply equally in relation to the law of possession, even if they are less common in practice. An often-cited example is a physical object in a warehouse the door to which has multiple keys held by different people. We think that this analogy can only go so far in the context of digital assets. Nevertheless, the point is that in the context of things in possession, the courts have developed complex and nuanced principles that can help resolve disputes in a multitude of different situations involving different types of physical thing. ⁴⁹³We think the courts are well-placed to do so again in relation to digital objects. The factors applicable to digital objects are likely to differ in many ways from those applicable to physical things, but that should not prevent the courts from undertaking that task. Again, the technical expert group will be well-placed to provide guidance on these complex issues, which could assist the courts in identifying what control looks like in relation to different technological forms.

Interference with control

• 5.99    Perhaps the clearest difference of opinion in consultee responses was not on the circumstances in which superior legal title and factual control could be separated in respect of a digital object. Nor was it necessarily the application of the concept of relativity of title (or priority) ⁴⁹⁴itself. Rather, the differences in opinion centred on the possible causes of action and associated remedies available to a person with a control-based legal proprietary interest in a digital object (that was subject to a superior legal title of another person).

• 5.100 As we discuss in detail in this chapter, it is difficult to discuss control in the abstract, without reference to specific technology or facts. We also recognise that if the law of England and Wales is to adapt to new technology in a way that preserves its international competitiveness, it should do so incrementally.⁴⁹⁵ As such, we conclude as follows.

5.101 There was strong support among consultees for the proposition that factual control (and intention, see paragraph 5.55 above) can be taken together to provide “best evidence” of the superior legal title to a digital object. This conclusion is limited to digital objects within the third category, and does not affect rights or claims enforceable by action. We think the better and more practical view — supported by those consultees who are closely involved in markets involving such digital objects, and market practice ⁴⁹⁶— is that factual control (plus intention) can found a legal proprietary interest in a digital object. We conclude that in certain circumstances such a control-based legal proprietary interest can be separated from (and be inferior to or short of) a superior legal title.

5.102 We think that in certain circumstances, claims involving third parties asserted by or against a person with a control-based legal proprietary interest (that was subject to a superior legal title of another person), could be based on and determined by reference to the controller’s control-based legal proprietary interest. We discuss this in more detail in Chapter 7 (Intermediated holding arrangements) in relation to intermediated holding arrangements and in Chapter 9 (Remedies) in relation to available causes of action and associated remedies.

5.103 In coming to that conclusion we acknowledge that the courts are best placed to develop the law by reference to the facts and technology in front of them (and in cases where genuine disputes arise). We think in this area, common law development is a more appropriate way for the law to evolve than by reference to the less nuanced hypothetical factual scenarios discussed in this paper and in academic and market commentary. Indeed, as we discuss in more detail in Chapters 7 (Intermediated holding arrangements) and 9 (Remedies), it has already begun to do so.⁴⁹⁷

Conclusion 1.

5.104 We conclude that factual control (plus intention) can found a legal proprietary interest in a digital object. We conclude that in certain circumstances such a controlbased legal proprietary interest can be separated from (and be inferior to or short of) a superior legal title.

Chapter 6: Transfers

INTRODUCTION

• 6.1    In this chapter, we discuss legal issues specific to the transfer of crypto-tokens. While some other third category things will transfer in similar ways, some will not. Because of that, and because crypto-token systems raise some specific questions as to transfers that are currently relevant to market participants, we generally refer to crypto-tokens (and not third category things) in this chapter.

• 6.2    Consultees were not uniform in their characterisation of crypto-tokens, or the legal consequences which they considered ought to flow from factual transfers of cryptotokens. In this chapter, we discuss two alternate views of consultees as to how best to characterise transfers and consider the practical and legal consequences of those views.

• 6.3    We discuss offchain transfers and conclude that it is possible to effect a legal transfer of a crypto-token offchain by a “change of control” (along with the requisite intention).

• 6.4    We also discuss the application and rationale for an “innocent acquisition” rule and conclude that a special defence of good faith purchaser for value without notice applicable to crypto-tokens can be recognised and developed by the courts through incremental development of the common law. We conclude that this reasoning can also be extended to other third category things.

BACKGROUND TO CONCEPTS USED IN THIS REPORT

A transfer operation that effects a state change

• 6.5    In general, an onchain⁴⁹⁸ factual “transfer”⁴⁹⁹ of a crypto-token occurs by way of a state change within a crypto-token system. ⁵⁰⁰A state change is, broadly speaking, a valid modification of the data recorded by the distributed ledger or structured record within a crypto-token system effected by a (transfer) operation conforming to the protocol rules of the system.

• 6.6    Typically, when a crypto-asset is “transferred”, the transfer operation will result in the imposition or creation of varying degrees of technical encumbrances in respect of the crypto-token (most commonly, the referencing or association of the crypto-token with a receiving public key address). Imposing or creating such technical encumbrances within a crypto-token system makes it possible for a person to have a factual relationship of control with a crypto-token. It is typically also possible for that person to divest themselves of the factual relationship of control with the crypto-token. ⁵⁰¹Again (and in general), this will be achieved by the technical completion of a transfer operation that effects a state change in accordance with the protocol rules of the relevant crypto-token system. So, in typical cases, a transfer operation that effects a state change will result in a change of control.⁵⁰²

• 6.7    The transfer operation, once confirmed, results in a change of state of the distributed ledger or structured record according to the protocol rules of the crypto-token system. We use the terms “state” to refer to the canonical⁵⁰³ and chronological order of transactional events as recorded within the distributed, transaction-based ledger or structured record of a crypto-token system (and “change of state” to changes to that record). This is consistent with technical descriptions authored by designers of cryptotoken systems. For example, in the Ethereum White Paper, Vitalik Buterin refers to the Bitcoin system as a “state transition system”:⁵⁰⁴

From a technical standpoint, the ledger of a cryptocurrency such as Bitcoin can be thought of as a state transition system, where there is a ‘state’ consisting of the ownership status of all existing bitcoins and a ‘state transition function’ that takes a state and a transaction and outputs a new state which is the result.

• 6.8    He goes on to describe the state within Ethereum as follows: ⁵⁰⁵

In Ethereum, the state is made up of objects called ‘accounts’, with each account having a 20-byte address and state transitions being direct transfers of value and information between accounts.

• 6.9   We therefore referred to “state”, “changes of state” and a “transfer operation that

effects a state change” in our consultation paper.⁵⁰⁶ We also noted that a change of state or a transfer operation that effects a state change does not necessarily happen instantaneously in crypto-token systems. ⁵⁰⁷In addition, depending on the system involved, it may take some time for the transfer operation that effects a state change to become probabilistically irreversible (in the context of some proof of work based systems) or finalised (in the context of some proof of stake based systems).⁵⁰⁸

A system for state transitions, not “just a record”

• 6.10    In our consultation paper, we said that the state of a distributed ledger or structured record within a crypto-token system should not necessarily be regarded as a definitive record of superior legal title to a crypto-token.⁵⁰⁹ We said that the state of the distributed ledger or structured record “may provide a definitive record of the links between discrete [factual states]⁵¹⁰ within the [crypto-token] system, but it cannot be a record of their legal effect”.⁵¹¹

• 6.11    However, the state of the crypto-token system does not merely record the factual situation; that is, it does not only show with which address a particular crypto-token is referenced or associated. It also records and enforces the technical encumbrances and conditions that regulate how crypto-tokens referenced or associated with that address can be used or spent.⁵¹² So the state of the crypto-token system cannot realistically be separated from its functionality such that it can be thought of as a record alone — it is not simply “a distributed database”.⁵¹³ The composite of recorded data and the functionality of that data within the crypto-token system give cryptotokens characteristics that are very different to mere entries in a database.

• 6.12    As we discuss in more detail in Chapter 4 (Third thing in practice), a crypto-token system manifests notional quantity units which the law of England and Wales treats as things to which personal property rights can relate. The crypto-token system relies on the active operation of software over network-instantiated data to manifest the operative functionality of crypto-tokens within the system. Specifically for the purposes of transfers, that functionality is the ability/power of a controller of the crypto-token uniquely to perform an operation (or an action)⁵¹⁴ in respect of that crypto-token.

• 6.13    In that sense, crypto-token systems are better described as manifesting functional crypto-tokens as distinct things. It is an incomplete analysis to say that crypto-token systems merely record a factual state of affairs — specifically the state of the system as a matter of fact at any particular time (by reference to a particular block height).⁵¹⁵

Factual consequences of a transfer operation that effects a state change

• 6.14    In our consultation paper we made the following three observations on the factual characteristics of a transfer operation that effects a state change within a crypto-token system.⁵¹⁶

• (1)    Such a transfer operation will typically involve the replacing, modifying, destroying, cancelling, or eliminating of a pre-transfer crypto-token and the resulting and corresponding causal creation of a new, modified or causally-related crypto-token.⁵¹⁷

• (2)    Such a transfer operation will typically involve the imposition or creation of varying degrees of technical encumbrances in respect of the causally-related crypto-token, which will typically amount to a change of control as between the pre-transfer crypto-token and the causally-related crypto-token.⁵¹⁸

• (3)    Such a transfer operation will typically result in a change of state of the distributed ledger or structured record in accordance with the protocol rules of the crypto-token system.

• 6.15    Consultees broadly agreed with this factual description. ⁵¹⁹Some consultees were however critical of the language used in sub-paragraph 6.14(1) above and its potential legal consequences. We discuss this point in detail from paragraph 6.21 below.⁵²⁰

Legal consequences of a transfer operation that effects a state change

• 6.16    Consultees were less uniform in their views as to how the law should best accommodate the factual reality we describe above.

• 6.17    In general, consultees considered that the formulation of our first criterion ⁵²¹(and, as a consequence, our conception of a crypto-token as described in our consultation paper) risked being overly data-centric. ⁵²²We think that a more or a less data-centric approach to conceptualising a crypto-token can potentially give rise to different analysis when applying legal principles to transfers of crypto-tokens.⁵²³

• 6.18    As we discuss in detail in Chapter 4 (Third thing in practice), we have taken on board consultees’ criticisms of our first criterion. We also reiterate our characterisation of a crypto-token as a composite thing. ⁵²⁴Specifically, in Chapter 4 (Third thing in practice) we summarise our conclusion as follows:

A crypto-token exists as a notional quantity unit manifested by the combination of the active operation of software by a network of participants and network-instantiated data.

• 6.19    We conclude in this chapter that this characterisation is consistent with consultees’ views as to the legal characterisation of transfers of crypto-tokens and remains consistent with the reasoning in our consultation paper. Broadly, consultees — even those with differing views — chose to focus their legal analysis on the ability/power of persons uniquely to perform an operation (or an action)⁵²⁵ within the crypto-token system in respect of notional quantity units.

• 6.20    Below, we address three different legal issues relating to a transfer of a crypto-token. First, we discuss two alternate views put forward by consultees as to how best to characterise transfers. Second, we discuss offchain transfers and conclude that it is possible to effect a legal transfer of a crypto-token offchain, by a “change of control” (along with the requisite intention). Third, we discuss the application and rationale for an “innocent acquisition” rule and conclude that a special defence of good faith purchaser for value without notice applicable to crypto-tokens can be recognised and developed by the courts through incremental development of the common law.

A TRANSFER OPERATION THAT EFFECTS A STATE CHANGE

• 6.21    When a crypto-token is transferred onchain, what the transferee receives is not (at least technically speaking) exactly the same thing as the transferee transferred, unlike when a tangible object is handed from one person to another. That is because the record within the crypto-token system is updated to reflect the transfer and a new factual state of affairs. Specifically, the state of the system is transitioned, updating the chain of cryptographically signed and validated transactions relating to the relevant entry in the distributed ledger or structured record.

• 6.22    This fact was perhaps on the mind of Lord Justice Birss when he discussed transfers of crypto-tokens after concluding that bitcoin was a notional quantity unit manifested by software:⁵²⁶

Since every transaction relating to that token adds to its chain, some would say a fresh piece of property is created every time bitcoin is transferred, but there is no need on this appeal to get into that debate.

• 6.23    We refer to this analysis as the “the extinction/creation analysis”.

• 6.24    In our consultation paper, we broadly agreed with the extinction/creation analysis and said that a transfer operation within a crypto-token system typically involves the replacement, modification, destruction, cancellation, or elimination of a pre-transfer crypto-token and the resulting and corresponding causal creation of a new, modified or causally-related crypto-token. ⁵²⁷We explained this analysis by reference to a variety of crypto-tokens.

• 6.25    While many consultees agreed with this analysis,⁵²⁸ many did not. ⁵²⁹Those who disagreed did not dispute the factual analysis of a transfer operation that effects a state change — that new data is added post-transaction and that the state changes.⁵³⁰ Instead, they rejected the description of the legal consequences of the updated state — most specifically the use of the terms “destruction”, “cancellation”, “elimination” and “new” to describe a post-transfer crypto-token. ⁵³¹In short, they rejected the proposition that a transfer operation that effects a state change extinguishes a pretransfer object of personal property rights and creates a “new”, post-transfer object of personal property rights.

• 6.26    Many consultees instead argued that a transfer operation that effects a state change involves the persistence of an object of personal property rights through the transfer. We refer to this as the “persistent thing” analysis.

• 6.27    We asked consultees for feedback on this issue directly.⁵³² As is clear from the responses analysed below, consultees considered that the issue was important. The reasoning behind the two differing analyses can be summarised as follows.

• 6.28    The extinction/creation analysis: The first argument recognises that a crypto-token is a notional quantity unit manifested by the combination of the active operation of software by a network of participants and network-instantiated data. The technological implementation of the system necessarily ensures that the functionality of that cryptotoken as a notional quantity unit is extinguished and re-created by a transfer operation that effects a state change. This is an intentional design choice taken precisely to avoid the double-spend problem.⁵³³ On that basis, the extinction/creation analysis describes a core (or perhaps the core) feature of a crypto-token system. Any ability/power uniquely to perform an operation (or an action) within the crypto-token system in respect of that crypto-token as a notional quantity unit is therefore also necessarily extinguished and re-created on transfer because the ability/power relates to different data instantiated within the system pre- and post-transaction.⁵³⁴

• 6.29    The persistent thing analysis: The alternative argument also recognises that a crypto-token is a notional quantity unit manifested by the combination of the active operation of software by a network of participants and network-instantiated data. It also recognises that a crypto-token as a notional quantity unit is manifested through different data instantiated within a system pre- and post-transaction. However, the legal analysis focuses more on a crypto-token as a notional quantity unit manifested at the system level, rather than on the changing recorded data. Because of this, it is possible to treat a crypto-token as a notional quantity unit (and the ability/power in respect of the notional quantity unit) as persisting through a transaction. The argument is that a core (or perhaps the core) feature of the system is that it tracks and records the location of notional quantity units as they persist and flow through the system — tracking and recording value as it flows through the system.⁵³⁵

• 6.30    Each analysis characterises the object of personal property rights in broadly the same way: as a notional quantity unit. Each analysis also recognises that the data instantiated within the system — and so the means of manifesting the object — changes pre- and post-transaction. The extinction/creation analysis simply places a greater analytical focus on data at the blockchain level, while the persistent thing analysis places a greater analytical focus at the system level.

• 6.31    We conclude that there is no “correct” analysis; both could be said to be accurate in their own ways. However, consultees offered a number of arguments as to why the most practically accurate analysis is the persistent thing analysis.

• 6.32    First, as we said in our consultation paper, users might treat transfers of their cryptotokens as transfers of (persistent) value,⁵³⁶ or as transfers of persistent things.⁵³⁷ Users are certainly likely to treat transfers of NFTs as transfers of persistent things.⁵³⁸

• 6.33    Second, in respect of UTXO-based systems, ⁵³⁹it is more accurate to focus on the notional quantity units that persist through transactions than the UTXO data. COMBAR said:⁵⁴⁰

There can be no UTXO without there being basic units from which it is calculated. In the case of the Bitcoin system, the basic unit is the satoshi. The satoshi is a construct by the system, as is the UTXO. Both are defined by the software of the system (the information attribute). The function of the satoshi (the functional or operational attribute) is an object that can be the subject of instructions to the system, to be recorded on the blockchain... What they are sending are satoshis, not the UTXO.

• 6.34    Third, the way in which UTXO-based transactions are constructed is better recognised by a focus on (persistent) notional quantity units. Very broadly speaking, a UTXO-based transaction can be constructed to include both a quantity of notional units to be sent to the transferee and a quantity of notional units to be sent back to the composer of the transaction as “change”. ⁵⁴¹As COMBAR argued, such a transaction should be regarded as simply transferring the net quantity of notional quantity units. Indeed, this position is reflected in HM Revenue & Customs’ approach to capital gains tax on transactions ⁵⁴²and pooling.⁵⁴³

• 6.35    Fourth, the fact that Account-based tokens⁵⁴⁴ are necessarily mixed on transfer to a single public address (where that address already is associated with a quantity of the same tokens) is better reflected by a focus on notional quantity units that persist through transactions, as opposed to discrete tokens associated with various specific data parameters. This is to some extent also true of UTXO-based tokens, ⁵⁴⁵but is not true of NFTs. Nonetheless, the persistent thing analysis might be important in the context of DeFi arrangements such as deposits into a liquidity pool smart contract that forms part of an automated market maker protocol.⁵⁴⁶ The analysis might also be important in DeFi collateral arrangements.⁵⁴⁷

• 6.36    Fifth, as we discuss at paragraph 6.92 below, an extinction/creation analysis might not, without more, provide a legal method by which new, indefeasible title to a thing can arise at law (at least in the context of bank account money). ⁵⁴⁸As we discuss below, an extinction/creation analysis is in any case not necessary to protect a good faith purchaser for value without notice of a crypto-token, given that we conclude that the common law could provide such protection.⁵⁴⁹

• 6.37    Sixth, the persistent thing analysis would potentially allow for the application of the evidential process of following to digital objects. We discuss this further in Chapter 9 (Remedies). Following is “the process of pursuing assets from one location to another”, as opposed to tracing, which involves the identification of substitute assets. ⁵⁵⁰While substitute assets will still be relevant (for example, where a self-held digital object was deposited into a non-custodial intermediary holding account), the broader argument is that “we need not draw any automatic conclusions from changes in physical (electronic or other) constitution.” ⁵⁵¹A less rigid approach to the applicability of rules concerning following/tracing to digital objects would allow the courts to engage in more complex evidentiary processes to characterise the result of (multiple) transactions involving digital objects.

• 6.38    Seventh, the persistent thing analysis better reflects the characterisation of a cryptotoken as a composite thing. It does not narrowly focus on only the data or the state of the record but instead recognises how a crypto-token system can manifest a notional quantity unit through the combination of the active operation of software by a network of participants and network-instantiated data.⁵⁵²

TRANSFERS BY “A CHANGE OF CONTROL”

• 6.39    In our consultation paper, we provisionally proposed that the law should be clarified to confirm that a transfer operation that effects a state change is a necessary (but not sufficient) condition for a legal transfer of a crypto-token. However, we also considered the possibility that a potentially wider condition could apply — namely, a “change of control”.⁵⁵³

• 6.40    By change of control, we broadly mean a change of “control” (as described in Chapter 5 (Control)) to another person.⁵⁵⁴

• 6.41    Consultees largely said that it is possible to effect a legal transfer of a crypto-token offchain by a “change of control”. ⁵⁵⁵They also said that it was possible to effect a legal transfer of a crypto-token onchain by a transfer operation that effects a state change (which, as we discuss above, itself would result in a change of control). In each case, most consultees said that intention would be a necessary element in determining whether a valid legal transfer occurred.⁵⁵⁶

• 6.42    We agree, and conclude that it is possible (with the requisite intention) to effect a legal transfer of a crypto-token offchain by a “change of control”⁵⁵⁷ or onchain by a transfer operation that effects a state change.

• 6.43    Consultees said that recognising that a “change of control” (plus the requisite intention) could effect a valid transfer at law accurately reflects the current use of crypto-tokens in the market. ⁵⁵⁸Limiting legal transfers to a transfer operation that effects a state change would focus the law too closely on particular data within a crypto-token system — specifically, the state of the distributed ledger or structured record. Consultees expressed a clear preference for the law of England and Wales to also facilitate and support offchain transfers.⁵⁵⁹

• 6.44    Other arguments in support of a change of control (rather than a transfer operation that effects a state change) being a necessary (but not sufficient) condition for the transfer of title include the following.

• (1)    Supporting sufficient flexibility within the law of England and Wales for market participants to structure their transfers as they see fit (including over the counter or offchain transfers, which are common in the market).⁵⁶⁰

• (2)    The varying degrees of technical encumbrances that can relate to crypto-tokens within crypto-token systems facilitate new ways in which control can be manipulated. In particular, there exist certain business models that enable parties to transfer exclusive control offchain, including (but not limited to) multi-signature ⁵⁶¹and key-sharding⁵⁶² techniques that relate to private keys, private pin numbers,⁵⁶³ the transfer of physical control over/access to private keys,⁵⁶⁴ and certain Layer 2 (or other scaling solution) functionalities.⁵⁶⁵

• (3)    Many DeFi products focus on the manipulation and transfer of control. Examples given by consultees include smart contracts used for making offers on NFT platforms,⁵⁶⁶ DeFi related collateral arrangements, staking arrangements⁵⁶⁷ and various different software-as-a-service arrangements⁵⁶⁸ which manipulate control in some way.

• (4)   Other international law reform initiatives focus on a change of control. For

example, both the UNIDROIT Working Group ⁵⁶⁹and the UCC Committee⁵⁷⁰ acknowledge that a transfer of a crypto-token can be made by a change of control.

• 6.45    In Chapter 5 (Control), we explain our reasoning for not defining control in a static way. Nonetheless, we think it is entirely appropriate for the courts to develop jurisprudence concerning a concept of control (and intention) which would properly reflect technology-neutral principles while being responsive to the specific technological implementation question, including on what constitutes a valid transfer at law. This has long been the way in which the law relating to relationships between parties and things has developed.

• 6.46    We conclude that our recommended technical expert group will be able to build up practical, technology-specific guidance over a period of time. Cumulatively, that guidance would help develop certainty under the law of England and Wales as to the precise elements of a “change of control” which are required as a necessary (but not sufficient) condition for the transfer of title. Neither the courts nor the technical expert group should be fettered by a static, incomplete or blunt definition of control. Most consultees were entirely confident in the ability of the courts in this respect as were members of the judiciary in our discussions with them. As COMBAR and the Chancery Bar Association said:⁵⁷¹

If [a concept of control and intention specific to digital objects] ⁵⁷²is available, then the consequences can be worked out by the court as necessary to resolve the disputes before them. The courts have not shied away from deciding novel points.

Conclusion 2.

• 6.47    We conclude that it is possible (with the requisite intention) to effect a legal transfer of a crypto-token offchain by a change of control or onchain by a transfer operation that effects a state change.

DERIVATIVE TRANSFERS OF TITLE

• 6.48    The majority of legal interests that persons acquire in things or objects are said to be dependent or derivative. That is, a person who acquires an object does not normally acquire a new, original interest in that object. Instead, a person will receive by transfer the interest that pre-existed in another (for example, through a sale or by taking delivery of a gift), or acquire some lesser interest, carved out of the better interest of another (as when a person becomes a pledgee). ⁵⁷³The acquirer’s rights in relation to the object will not be independently acquired rights, ⁵⁷⁴but instead rights that derive from the rights of others.

• 6.49    We conclude that the rules of derivative transfer of title can apply to crypto-tokens. In our consultation paper, we said that the rules of derivative transfer of title can apply to crypto-tokens notwithstanding the extinction/creation analysis argument that an onchain transfer of a crypto-token involves the acquisition of a new, causally-related thing by the transferee. We also discuss above that there are good arguments for the persistent thing analysis — that the law could conceptualise crypto-tokens as persisting through transfers. The persistent thing analysis would make the argument that the rules on derivative transfers of title apply to onchain transfers even stronger (given that the crypto-token would not be extinguished/created on transfer).⁵⁷⁵

• 6.50    We conclude that it does not make a difference that notional quantity units (or even a power/ability in respect of such notional quantity units) are necessarily intangible. As Professor Fox says (in the context of incorporeal money):⁵⁷⁶

... incorporeal money can indeed be transferred. To speak of incorporeal transfers gives full effect to the principle that the law should aim for functionally equivalent outcomes regardless of whether money is paid in corporeal or incorporeal form. Moreover, on a closer investigation, the principle of derivative transfer of title can never have entailed that the payer’s possession or ownership of a corporeal asset actually passed, physically, to the recipient. That would be to take an extreme, and unrealistic, approach to the reification of property interests. On balance. it is justifiable to treat the incorporeal transfer [of money by means of an inter-bank payment] as involving a derivative means of acquiring title. First, the creation of the recipient’s title depends on the expression of the payer’s will at the outset. It is the payer who initiates the payment instruction. Secondly. the fact that the recipient’s claim against his or her bank is newly created does not necessarily entail that he or she takes it free from competing titles. Indeed, it would take an extreme and unrealistic conception of derivative means of acquiring title to sustain the view that transfers of incorporeal money should be treated differently from corporeal transfers. Even in the simplest case of a transfer of ownership by delivery of a chattel, the network of jural relations constituted by the transferee’s possession and ownership of the chattel is different from that constituted by the transferor’s former possession and ownership. ⁵⁷⁷The transferee does not in fact succeed to the same possession and ownership as the transferor. Possession and ownership are legal constructs. Unlike the corporeal assets they relate to, they cannot be transferred in space from one person to another.

• 6.51    On that basis, we reiterate our provisional conclusion from our consultation paper that this line of reasoning can be extended to the transfer of crypto-tokens where such a transfer is effected by a transfer operation that effects a state change, or a change of control. ⁵⁷⁸We also think that the concepts of possession described by Professor Fox in the passage above can apply by analogy to a concept of control and intention specific to crypto-tokens. Our starting point is that such a transfer (either onchain or offchain) is a necessary (but not sufficient) condition for the transfer of superior legal title to a crypto-token.

• 6.52    Such a transfer is not sufficient in itself to transfer superior legal title to a crypto-token for two reasons. First, in general, a transferor can confer no better title to a transferee than they have (the “nemo dat principle”). ⁵⁷⁹Second, the transaction between the transferor and the transferee must be legally valid in terms of the common law and equitable rules governing derivative transfers of title.⁵⁸⁰

• 6.53    The nemo dat principle is codified in law in relation to goods by section 21(1) of the Sale of Goods Act 1979. As a general principle, it is subject to statutory exceptions, as well as the common law special defence of good faith purchaser for value without notice that applies to money and to negotiable instruments, and the equitable principle of good faith purchaser for value without notice.⁵⁸¹

• 6.54    The common law special defence of good faith purchaser for value without notice that applies to money creates a fresh, indefeasible legal title in a purchaser of money in good faith and for value. The common law special defence operates to make the purchaser immune from the claim (legal or equitable) of any previous holder who might otherwise have retained a proprietary interest in the money. ⁵⁸²The common law special defence of good faith purchaser for value without notice that applies to negotiable instruments such as bills of exchange and promissory notes ⁵⁸³operates in a similar way. It creates a fresh, indefeasible legal title in a transferee (a “holder in due course”) who takes the negotiable instrument in good faith and for value, and without notice of any defect in the title of the person who negotiated it. Possession of the negotiable instrument is required for a holder to enforce the right to claim performance of the obligations embodied by the negotiable instrument.⁵⁸⁴

• 6.55    Below we conclude that a special defence of good faith purchaser for value without notice applicable to crypto-tokens can be recognised and developed by the courts through incremental development of the common law. We conclude that this reasoning can also be extended to other third category things. As we discuss below, this would significantly limit the application of the nemo dat principle to crypto-token transfers. However, for a number of reasons we conclude that such a development would represent only a limited practical extension to the legal position were a common law special defence not available.

A COMMON LAW SPECIAL DEFENCE OF GOOD FAITH PURCHASER FOR VALUE WITHOUT NOTICE

• 6.56    In our consultation paper, we provisionally proposed that a common law special defence of good faith purchaser for value without notice should apply to a transfer of a crypto-token by a transfer operation that effects a state change.⁵⁸⁵ Given that we mooted the possibility of codifying such a rule in statute, we referred to the concept as an “innocent acquisition rule”.

• 6.57    We provisionally proposed that an innocent acquisition rule should apply to both “fungible” and “non-fungible” technical implementations of crypto-tokens. ⁵⁸⁶We said that an innocent acquisition rule could not and should not apply automatically to things (such as legal rights) that are linked to a crypto-token.

• 6.58    We said that it is possible that the common law could develop an innocent acquisition rule in respect of crypto-tokens by way of analogy with the existing common law special defences of good faith purchaser for value without notice that apply in respect of money and negotiable instruments. However, we provisionally concluded that market participants would benefit from a statutory provision which explicitly recognised an innocent acquisition rule in respect of transfers of crypto-tokens by a transfer operation that effects a state change. We noted that this is the approach recommended by both the UCC Committee⁵⁸⁷ and the UNIDROIT Working Group.⁵⁸⁸

• 6.59    Consultee responses to this set of consultation questions⁵⁸⁹ were strongly in favour of an innocent acquisition rule. This was one of the topics in our consultation paper that received the most consistent and strongly argued set of responses.⁵⁹⁰

• 6.60    We agree with the views of the majority of consultees. However, based on consultee responses and the recommendations that we make in the rest of this report, our conclusion is different from the provisional proposal in our consultation paper in two principal ways.

• 6.61    First, we now conclude that the law should recognise that it is possible for a transfer of a crypto-token to be made offchain, through a change of control (along with the requisite intention). ⁵⁹¹As we discuss from paragraph 6.39 above, this is a wider concept than the more limited concept of a transfer operation that effects a state change (an onchain transfer). As such, the application of any common law special defence of good faith purchaser for value without notice would not in principle be limited to situations involving an onchain transfer. The common law special defence could also apply to a transfer of a crypto-token made offchain, through a change of control, although difficult evidential issues could arise where a subsequent (and conflicting) onchain transfer occurs.⁵⁹²

• 6.62    Second, we do not recommend ⁵⁹³a hard definition of “data object”, “digital object”, “crypto-token”, controllable “electronic record” or any other type of digital asset,⁵⁹⁴ in contrast to the definitions proposed by the UCC and UNIDROIT ⁵⁹⁵(and in contrast to our proposals in our consultation paper).

• 6.63    This means that there is and will be no hard definition under the law of England and Wales on which to base a statutory innocent acquisition rule for crypto-tokens or other third category things. Therefore, while many consultees said it would be preferable or useful to implement a statutory innocent acquisition rule, we do not recommend that at this time.

• 6.64    Instead, we conclude that a special defence of good faith purchaser for value without notice applicable to crypto-tokens can be recognised and developed by the courts through incremental development of the common law.

• 6.65    Below, we (1) discuss how our conclusion might take effect at common law; (2) discuss how our conclusion represents only a limited practical extension to the legal position were a common law special defence not available; and (3) reiterate the policy rationale for our conclusion.

The provisional proposal in our consultation paper

• 6.66    In our consultation paper we said that as far as the law currently stands, we do not consider that crypto-tokens or third category things benefit from the existing common law special defences of good faith purchaser for value without notice applicable to money or to negotiable instruments. We said that because crypto-tokens are generally not considered to be either money or negotiable instruments, the existing defences would not apply. We said however that this would not prevent the courts from developing a similar defence by analogy with the existing defences applicable to money or negotiable instruments:⁵⁹⁶

We think that the best way to avoid any lingering uncertainty as to the application of the common law special defence of good faith purchaser for value without notice is explicitly to recognise its application in relation to transfers of crypto-tokens.

• 6.67    Most consultees agreed with this position, but some took it one step further and argued that an applicable special defence already exists as a matter of common law.⁵⁹⁷

How our conclusion might take effect at common law

• 6.68    For the reasons discussed in our consultation paper ⁵⁹⁸and which we discuss below, we conclude that the common law special defences of good faith purchaser for value without notice that apply to money and to negotiable instruments do not apply directly to crypto-tokens or to other third category things.

• 6.69    Instead, we think that it is more appropriate for the common law to recognise, refine and develop a similar special defence of good faith purchaser for value without notice that applies to crypto-tokens and to other third category things. That defence can be recognised and developed by analogy with the existing defences applicable to money and negotiable instruments. As the Crypto Council for Innovation said:⁵⁹⁹

We do not advocate the wholesale importation of the law of negotiable instruments as it currently stands into the law of cryptoassets... However, the concepts which lie behind the law of negotiable instruments match closely the structural elements which a law of cryptoassets should achieve, and this body of law provides some useful conceptual scaffolding which could be relatively easily adapted to the position in respect of cryptoassets.

• 6.70    The benefit of this approach is that crypto-tokens and other third category things need not be characterised as either money or negotiable instruments for a person to benefit from legal protection when they are a purchaser in good faith for value and without notice.

Analogy with the common law special defence of good faith purchaser for value without notice that applies to money

• 6.71    The characterisation of crypto-tokens as money or otherwise is outside the scope of this report. However, we make three observations on the point. First, the debate on whether a crypto-token is “money” has largely focused on a specific sub-set of cryptotokens over the last few years (that is, those that are intended to function in a similar way to money), rather than on crypto-tokens in general. ⁶⁰⁰However, other cryptotokens are not designed to replicate the functionality or characteristics of money. So the special defence applicable to money would remain inapplicable to those types of crypto-token.

• 6.72    Second, whether a crypto-token (or specific crypto-token) is characterised as money is likely to depend on the law of the particular jurisdiction in question and this characterisation might be different across different jurisdictions. Indeed, some cryptotokens are already recognised as legal tender in some jurisdictions. ⁶⁰¹It is also possible that some jurisdictions will adopt central bank digital currencies, which might be given “monetary” status by legislation.⁶⁰²

• 6.73    Third, “the property regime applying to money is just one of a number of legal devices by which the state can support the efficiency of monetary functions”. ⁶⁰³It is therefore less accurate to ask “are crypto-tokens money?” than to ask the specific question as to whether the law is able to characterise crypto-tokens as something that should enjoy a privileged proprietary status which other kinds of assets or things do not enjoy. This question can be asked separately for each distinct “legal device” by which the state can support the efficiency of monetary functions, including for the application of a common law special defence.

• 6.74    Crypto-tokens therefore do not need to be “money” to benefit from a special defence of good faith purchaser for value without notice.

• 6.75    This is the approach taken by the UCC Committee in Article 12 of the UCC,⁶⁰⁴ which does not comment on the monetary status of crypto-tokens but still:⁶⁰⁵

Preserve[s] the decentralized nature of bitcoin and other cryptocurrencies, enable[s] secured lenders to enjoy legal benefits of their existing commercial practices, and protect[s] the negotiability of bitcoin by allowing onward transferees to take bitcoin and cryptocurrencies free of existing encumbrances.

• 6.76    While it does not apply directly, the common law special defence of good faith purchaser for value without notice that applies to money can be used as a useful analogy for a similar common law special defence that could be developed for cryptotokens. The analogy is particularly apt given that money can exist as an intangible thing in action — bank account money — or as a physical thing in possession in the form of coins (or, in the case of a bank note, a combination of the two).⁶⁰⁶

Analogy with the common law defence of good faith purchaser for value without notice that applies to negotiable instruments

• 6.77    Many consultees⁶⁰⁷ made a strong argument that analogies could be drawn between crypto-tokens and negotiable instruments for the purpose of an innocent acquisition rule. This is the approach taken by the UNIDROIT Working Group and the UCC Committee, both of whom refer explicitly to the concept of negotiability.⁶⁰⁸

• 6.78    For negotiable instruments, the common law requires evidence of a mercantile practice of the relevant instrument “passing by delivery from hand to hand”⁶⁰⁹ and historically only tangible things have acquired negotiable status through mercantile custom.

• 6.79    In addition, the law on negotiable instruments developed based on the transfer of a right to sue an obligor (that is, a transfer of a thing in action) being embodied in a physical instrument capable of physical delivery (that is, a thing in possession). As we discuss in Chapters 3 (Third thing) and 4 (Third thing in practice), a crypto-token in general does not confer or consist of such a right (it is not a thing in action), and it is not tangible (it is not a thing in possession).

• 6.80    Applying the concept of negotiability to crypto-tokens therefore necessarily requires analogy. The analogy is between tangible things such as paper documents which are used as “tokens [tangible embodiments] of abstract rights”⁶¹⁰ on the one hand (that is, negotiable instruments), and intangible things used as tokens of notional quantity units (which exist independently of any rights in relation to them) on the other. Cryptotokens are an example of the latter. When a crypto-token is linked in some way to a right or claim enforceable by action (such as when creating a “digital bearer instrument”) ⁶¹¹the analogy with negotiable instruments is much clearer. Nevertheless, that situation still raises the question as to whether the focus should be on the transfer of the legal rights, or the transfer of the thing that embodies those legal rights.

• 6.81    We think the better approach is to focus on the transfer of the thing that embodies the rights.⁶¹² This must be correct, because we know that crypto-tokens can exist as things in themselves and not as legal rights: they are not things in action. ⁶¹³With a negotiable instrument, the rights to claim performance of the obligations that the negotiable instrument embodies move with the holder in due course of the negotiable instrument. The promise from the maker can be considered as “an original and direct promise to the bearer”.⁶¹⁴ This is not the case for crypto-tokens: the crypto-token as a thing to which personal property rights can relate is much better analogised as a token with some inherent bearer-like qualities, but not as a promise from any person or issuer. So focusing on the transfer of a “right” does not work with a crypto-token.

• 6.82    Instead, we consider that the appropriate exercise is critically to analyse the mercantile custom that has arisen in respect of crypto-tokens. The question is whether the mercantile custom that has arisen in respect of crypto-tokens is sufficiently similar to the mercantile custom in relation to negotiable instruments so as to justify the application of a common law special defence to crypto-tokens.

• 6.83    It is clear that the law of England and Wales will treat crypto-tokens as a thing to which society attaches legal consequences (including the recognition that property rights can relate to that thing). ⁶¹⁵In this way, the law is responsive to the actions and intentions and expectations of parties and recognises that (much of) society regards such assets as socially and economically important.

• 6.84    With this in mind, it is difficult to see why the law should treat crypto-tokens as things to which personal property rights can relate, but not treat them as “negotiable”⁶¹⁶ where this is the clear intention and developed custom of users. Indeed, crypto-token systems are “state transition systems” ⁶¹⁷designed to facilitate the communication of value ⁶¹⁸on a trust-minimised (and often pseudonymous or anonymous) basis. As such, the technology is arguably explicitly designed such that the notional quantity units within the system mimic some of the functionality of negotiable bearer instruments.⁶¹⁹ This distinguishes crypto-tokens from physical things in possession which are not explicitly designed to facilitate peer-to-peer transfers of value on a global and online basis. In respect of those physical things in possession, there is a better argument that the law should prioritise the personal property rights of individuals in respect of their physical things over the sanctity of transactions involving an innocent acquirer of those things.⁶²⁰

• 6.85    In short, drawing an analogy with the common law special defence of good faith purchaser for value without notice that applies to negotiable instruments is apt. The category of negotiable instruments (or, for our purposes, those third category things including crypto-tokens that are treated by market participants as “negotiable” by analogy) is not closed. ⁶²¹The category of negotiable instruments has not always necessarily required significant time to develop, and so the relative nascence of crypto-token markets should not be a barrier to common law development. ⁶²²Indeed, the common law is highly responsive to market practice ⁶²³and it is a simple fact that markets for crypto-tokens exist today, regardless of how quickly they developed.⁶²⁴

• 6.86    Crypto-tokens are not, in any case, brand new. ⁶²⁵The Bitcoin whitepaper was published in 2008 ⁶²⁶and the Ethereum Yellow Paper in 2014.⁶²⁷ Further, cryptotokens are used in a large number of transactions, by a large number of people, and in high total volumes. ⁶²⁸The number of transactions on the Bitcoin network regularly exceeds five million per month, with monthly volumes ranging significantly between $200 billion and $8 trillion since 2020.⁶²⁹ Since 2018 the number of active addresses (monthly) (not active persons) on the Bitcoin network ranged between 18 million and 35 million. The number of transactions on the Ethereum network regularly exceeds 30 million per month, with recent monthly volumes since 2020 ranging significantly between $6.5 billion and $404 billion.⁶³⁰ Since 2018 the number of active addresses (monthly) (not active persons) on the Ethereum network ranged between five million and 20 million. ⁶³¹In contrast, some existing negotiable instruments are used much less today: in 2022, nearly 131.5 million cheques with a total value of around £187 million were processed by the image clearing system.⁶³²

• 6.87    We conclude that this level of adoption is evidence that mercantile custom has arisen in respect of such crypto-tokens — particularly because users (or the software they rely on) must comply with the protocol rules of the system as designed to be able to transact. In that respect, we agree with Linklaters LLP that:⁶³³

It would be a relatively straightforward incremental step for the courts to recognise a mercantile custom treating onchain transfers of crypto-tokens as negotiable, where such a custom can be demonstrated. ... We believe such custom is highly evident in the case of many frequently traded crypto-tokens. ... We see no reason why the same kind of technical extension to the common law to recognise a global mercantile custom should not also apply in the case of crypto-tokens. We also believe that market practice established in relation to certain crypto-tokens can easily (and very quickly) be extended to similar crypto-tokens by analogy.

• 6.88    Lord Devlin said in Kum v Wah Tat Bank⁶³⁴ that the function of the commercial law is to allow, so far as it can, commercial people to do business in the way in which they see fit. If the law of England and Wales is to navigate the technological changes of the 2020s, 2030s and beyond, ⁶³⁵it will need to adapt quickly. This is not a new phenomenon for the law. Over 120 years ago, Mr Justice Bigham in Edelstein v Schuler & Co said:⁶³⁶

It is no doubt true that negotiability can only be attached to a contract by the law merchant or by a statute; and it is also true that, in determining whether a usage has become so well established as to be binding on the courts of law, the length of time during which the usage has existed is an important circumstance to take into consideration; but it is to be remembered that in these days usage is established much more quickly than it was in days gone by; more depends on the number of the transactions which help to create it than on the time over which the transactions are spread; and it is probably no exaggeration to say that nowadays there are more business transactions in an hour than there were in a week a century ago. Therefore, the comparatively recent origin of this class of securities in my view creates no difficulty in the way of holding that they are negotiable by virtue of the law merchant; they are dealt in as negotiable instruments in every minute of a working day, and to the extent of many thousands of pounds. It is also to be remembered that the law merchant is not fixed and stereotyped; it has not yet been arrested in its growth by being moulded into a code; it is, to use the words of Cockburn CJ in Goodwin v Robarts, capable of being expanded and enlarged so as to meet the wants and requirements of trade in the varying circumstances of commerce, ⁶³⁷the effect of which is that it approves and adopts from time to time those usages of merchants which are found necessary for the convenience of trade.