Eversheds Sutherland response to UK Jurisdiction Taskforce consultation on cryptoassets, distributed ledger technology and smart contracts

• United Kingdom

• Financial services and markets regulation
• Financial services and markets regulation - Hedge funds
• Other
• Financial services

28-06-2019

Introduction

The Eversheds Sutherland Cryptoassets team regularly contributes to consultations launched by governments and regulatory bodies, and has given evidence to Parliament regarding the promotion of blockchain and cryptoasset technologies in the United Kingdom.

On 9 May 2019 the UK Jurisdiction Taskforce (“UKJT”), chaired by Chancellor to the High Court, the Rt Hon Sir Geoffrey Vos, published a consultation paper to identify key issues of legal uncertainty regarding cryptoassets, DLT and smart contracts with the aim of providing clarity in this area of law through the means of a legal statement.

The UKJT is one of the six taskforces of the LawTech Delivery Panel. Its objective is to demonstrate that English law and the jurisdiction of England and Wales together provide a state-of-the-art foundation for the development of DLT, smart contracts and associated technologies.

The consultation sought input from stakeholders on the principal issues of perceived legal uncertainty about the status of cryptoassets and smart contracts under English private law, to inform what should be addressed in the legal statement.

The consultation closed on 21 June 2019. Below is our submission.

To read the consultation paper, click here.

To visit the Law Society webpage, click here.

Our submission

Legal status of cryptoassets: Under what circumstances, if any, would the following be characterised as personal property: a cryptoasset; and a private key?

The starting point for determining the characterisation of a “cryptoasset” or a “private key” must start with a consideration of what these are, and whilst the UKJT paper (the “Consultation”) provides a useful exposition of the useful features of DLT, it is worth noting the description put forwards by the Financial Conduct Authority (“FCA”):

“While there is no formal definition of DLT, it can be described as a set of technological solutions that enables a single, sequenced, standardised and cryptographically-secured record of activity to be safely distributed to, and acted upon by, a network of varied participants. This record could contain for example, transactions, asset holdings or identity data. This contrasts with a traditional centralised ledger system, owned and operated by a single trusted entity. We consider a blockchain to be a type of DLT where records are collated into “blocks” and linked using a cryptographic signature.¹”

DLT is, therefore, a form of electronic register, which is sui generis in the way it uses validators and nodes to create an immutable record which is open to all parties. Importantly, DLT solves the double-spend problem using a trustless mechanism², which enables participants to provide bits with certain attributes, in particular making them rivalrous, so that they can hold independent value³. In providing attributes to bits, participants often refer to them as “tokens” or “cryptoassets”.

Whilst, therefore, there “is currently no legal definition of ‘crypto-assets’ in the EU financial securities laws”⁴, given that the nature of the crypto-assets is determined by its characterisation, in our view a functional approach can be taken to determine the role ascribed to the bits within the DLT framework⁵.

To analyses this further, it is worth considering how property is generally defined, and in this respect it is worth noting the following:

“The starting point for any discussion of digital ownership is to examine how property law functions in real functions in the real world. Definitions property differ but they all appear to have some elements in common. The first is that property defines a relationship between a person and a thing. Unlike obligations which normalise relations between persons, one tangent of the axis in a property relationship must be a thing. This is because property, and property law, regulate one’s right to own, buy and sell, dispose or destroy. These rights may only be exercised overt things: it has been illegal to take rights such as these over persons in the UK for almost 200 years. The second common theme of property law is that it is exclusive. The rights that property law confers upon the owner, or other rightsholder such as a lessee, are of the nature of rights in rem as opposed to rights ad personam. This means that the property rights holder has a right which may be exercised against any individual who attempts to interfere with his or her property right without the need for a prior relationship with that person. This may be contrasted with obligations which arise out of a prior relationship such as a contractual relationship or a relationship which establishes a duty of care in tort. As James Penner explains in his book The Idea of Property Law, the essential element of these rights is the right to exclude others from exercising competing rights over your property.”⁶

The core distinction this helpfully draws out is the difference between rights between persons (rights ‘in personam’) - whereby, in the event of dispute any action is between the persons only (e.g. for breach of contract) - and rights against the world (rights ‘in rem’) - whereby, in the event of a dispute rights are enforceable against any other person (e.g. for trespass). In our view, this distinction is determinative in deciding whether a cryptoasset or private key is personal property.

Cryptoassets

As stated above, the genesis of cryptoassets is the attribution of characteristics to bits, and it is the nature of these characteristics which determine whether they are personal property. Also, as stated in the Consultation, there are various taxonomies for cryptoassets, which seek to split them into different grouping, however, they are not generally helpful for the current analysis to the extent that they merely reflect different ways in which cryptoassets fall within the regulatory framework, rather than an analysis of the different fundamental roles that can be played by different cryptoassets. To achieve this, therefore, we would suggest looking at a purely functional system of classification, and in particular we suggest that cryptoassets can be split into three core functions⁷:

1. Cryptoassets which operate as vouchers: these are tokens which refer to something outside the network (so-called “off-chain” assets), which may be a reference to a security (security tokens), service or goods (utility tokens) or fiat money (e-money) (“crypto-voucher tokens”).

2. Cryptoassets which act as a means of investment / exchange: these are different to crypto-vouchers, in that they do not act as a reference to an external asset or service, but rather it is the existence of the token in itself which is valued, and include, for example, bitcoin (“crypto-transaction tokens”).

3. Cryptoassets which are the coding base for tokens: these effectively constitute coding platforms, on which the other token types can be built, and include, for example ERC-20 (“crypto-fuels”).

Once cryptoassets are split into these three types, it becomes possible to determine their legal nature, as delineated below.

Crypto-voucher tokens

1.8 Crypto-voucher tokens function as a register in relation to the underlying asset / service. The legal nature of the register is, therefore, determined by the evidential value placed on it. In other words, if it is accepted that the legal nature of the register is such that the owner of the cryptoasset listed on it is the owner of a real world item (which would be the case, for example, if the land registry were placed on the blockchain), the cryptoasset would be evidence of real world ownership of the off-chain asset. Taken one step further, to grant security in such a scenario, one would not grant security against the cryptoasset itself (this simply constituting evidence of title), but rather one could register a security on the blockchain against the off-chain asset (in the same way as registering a charge at Companies’ House).

Crypto-exchange tokens

Crypto-exchange tokens do not have an off-chain reference point, but rather the token is given a value in and of itself. In this sense, it can be compared to money, noting, however, that as yet no crypto-exchange is deemed money, on the basis it does not meet the test generally required to be deemed money, i.e.⁸:

1. A medium of exchange: crypto-exchange tokens are not generally accepted as a form of payment (although some stores do accept them). This is generally regarded as the key feature⁹, which may explain the emphasis placed on this aspect by those seeking to depict crypto-exchange tokens as cash substitutes.

2. A store of value of wealth: the volatility of crypto-exchange token has prevented this so far.

3. A measure of value or as a standard for contractual obligations: again, the issue of volatility and lack of acceptance as means of payment have generally prevented this so far.

4. A unit of account: generally, crypto-exchange tokens do actually meet this requirement.

From an economist’s perspective, therefore, crypto-exchange tokens are not currently money. The state theory of money would reinforce this, given that, as yet, no jurisdiction has given a crypto-exchange the attribute of currency¹⁰. In this respect, it is worth noting that crypto-exchange tokens function as an object, rather than a medium of exchange¹¹ - a useful comparison to demonstrate this is with gambling chips, which act as a store of value, may be used as a basis of exchange at participating casinos, but are not, in themselves, a form of money.

The best characterisation, therefore, of crypto-exchange tokens is as a form of commodity, and indeed it is worth noting that, in older legal analyses of foreign monies, they were also characterised as commodities¹². In taking this view, we note that the final position of crypto-exchange tokens is not yet settled, and, for example, in certain jurisdictions crypto-exchange tokens have been integrated within payment services regulation, suggesting that in the long term it is possible for a crypto-exchange to become a type of money¹³.

The consequence of determining that a crypto-asset is a commodity allows for it to be the subject of security, and to form part of the assets of a person, e.g. on insolvency. It has already been held by the Singapore Supreme Court that bitcoin can be the subject of a trust (B2C2 Ltd v Quoine Ltd SGHV (l) 03), and by the UK High Court that buyers of bitcoin may benefit from protections afforded to consumers (Ang v Reliantco Investments Ltd [2019] EWCH 879 (Comm)). Lastly it is worth noting, as an aside, that in the bitcoin whitepaper, the rationale provided for the setup of bitcoin was to function the same way as gold, which itself is treated as a commodity.

Giving crypto-exchange token the status of commodity, whilst it does not fulfil the criterion of money, would provide legal clarity without requiring substantial changes to the current operation of the existing legal framework.

Crypto-fuels

Crypto-fuels, as stated above, operate as the coding base for designing tokens. As they do not create enforceable rights, neither in personam nor in rem, but rather act simply as a basis for designing other constructs, we do not see them as able to constitute personal property.

Private keys

Private keys are not assets in themselves, but rather are the effective means of controlling cryptoassets. Practically, therefore, they are very important - for example, a custodian of cryptoassets would require control over the private key in order to prevent unauthorised use of the cryptoasset. That being said, keys do not share the attributes of:

1. Contractual rights: there is no “meeting of the minds”, consideration or other hallmarks of a contractual basis for private keys.

2. Property rights: they are not exclusive (keys can be copied, as any password can be), and as such there is no central ‘thing’ against which right can be exercised in rem.

In the context of the current analysis, therefore, private keys would not be personal property, albeit they might be the mechanism for exercising control over a cryptoasset which, itself, could constitute personal property.

Enforceability of smart contracts: In what circumstances is a smart contract capable of giving rise to binding legal obligations, enforceable in accordance with its terms (a “smart legal contract”)?

The starting point for determining what a smart contract is capable of is to determine what a smart contract actually is. At their simplest, a smart contract can be summarised as “if ... then ...”. In this respect, the expression “smart contract” is misleading, in that they are not designed to be contracts. For example, the smart contract used in the DAO (acknowledging this is a gross simplification) stated “if sufficient investors approve, the recipient receives ETH”. As such:

1. There is no requirement for consideration for a smart contract to be valid (e.g “if it is Wednesday” would be sufficient to trigger a smart contract).

2. There is no requirement for a “meeting of the minds” between participants for a smart contact to be valid and executed. AI algorithms, for example, can enter into and execute smart contracts without human intervention (as occurred in B2C2 Ltd v Quoine Ltd SGHV (l) 03).

3. There is no scope for contractual interpretation and only binomial outcomes are possible. This means that concepts such as reasonableness cannot be coded into a smart contract - in the case of the DAO, it is arguable that this was a major contributor to its downfall, as the “code is law” approach meant that the DAO infrastructure could not cope with unexpected events.

4. There are no implied terms within a smart contract: code executes on the terms on which it is set, and terms implied by law, for example under the Consumer Rights Act 2015, are excluded to the extent that they are not specifically included within the code.

For these reasons, it is clear that a smart contract is not ipso facto a smart legal contract. Nor, in our view, would it make sense to somehow redefine the nature of contract law such that a smart contract would be a legal smart contract.

Instead courts should, as they have done so far, look at the context in which the relevant smart contract is developed, and it is in taking this approach the ancillary questions can be answered.

How would an English court apply general principles of contractual interpretation to a smart contract written wholly or in part in computer code?

Although smart contacts are relatively new, the question of dealing with events originating with human actions but which have consequences outside of human control has existed since Rylands v Fletcher [1868] UKHL 1. That case established, as a general principal, that a person is responsible for the consequences of that which is under his control coming out of control - a sentiment which is echoed in the B2C2 case, under which the actions of a smart contract were attributed to its creator.

Using this approach, therefore, the approach of the courts to a smart contract are, in our view correctly, likely to be twofold. Firstly, if in the event that two parties contract, and as part of that contract they agree to use smart contract, then the smart contract itself is not the contract, but rather the means of discharging rights and obligations under the contract. In such a scenario, the courts would interpret the smart contract in light of the overarching contract, to which they would apply the general principals of contractual interpretation.

The second case would be where a smart contract had unintended consequences, for example causing unacceptable market outcomes. In such a scenario, the court would likely hold the creator of the smart contract responsible for these consequences - and indeed, this is the general approach already taken in relation to high frequency trading algorithms and potential market abuse.

Under what circumstances would an English court look beyond the mere outcome of the running of any computer code that is or is part of a smart contract in determining the agreement between the parties?

Please see response above - this is likely to be the case where it is clear to the court, e.g. because of an overarching contract that, in using smart contracts, the relevant parties do so cognisant that broader rules of construction may apply. This will, in our view, generally be the case, noting, for example, that in the case of the DAO, where parties actively chose to disapply any broader legal framework, there was arguably still an overarching contract - albeit one which disapplied any contractual terms not encoded into the system. The courts are particularly likely to take account of the broader relationship between participants where they include consumers/ retail customers, as various statutes, such as the Consumer Rights Acts, may mandate additional contractual provisions, even if not written in the code itself.

Is a smart contract between anonymous or pseudo-anonymous parties capable of giving rise to binding legal obligations?

For legal obligations to be binding there must be a “meeting of the minds”, and so, for example, a person cannot accept a contract of which he is unaware. This does not import a requirement to have a specific degree of knowledge regarding the attributes of the other party for a contract to be binding (except possibly as regards capacity), and as such we see no issue with a smart contract between pseudo-anonymous parties being able to give rise to legal obligations. We do note, however, that, as a matter of practice, the requirements of anti-money laundering and regulatory obligations may complicate matters, however this is outside the scope of the current consultation.

As regards completely anonymous parties, the position will be highly fact dependent, and as such it is hard to give a concrete response. We would, however, note that even if participants within a DLT ecosystem are completely anonymous, they are still required to comply with the rules of that ecosystem. In turn, DLT ecosystems are reliant on an administrator to ensure that they continue to operate smoothly (a concept referred to as legal entity dependence¹⁴). The result is that DLT ecosystems must develop an operational framework, which work as “membership rules” which participants are bound to. Even if, therefore, the parties to a smart contract are anonymous, they would in the normal course be contracting through an ecosystem whereby their relationship would be governed by an administrator which would not be anonymous, and as such capable of enforcing legal obligations.

Could a statutory signature requirement be met by using a private key?

In our view, a statutory signature requirement could be met by using a private key.

A signature, in broad terms, is a mark or sign made by an individual on an instrument or document to signify knowledge, approval, or acceptance of obligations - the purpose of which is to bind the signatory to the relevant provisions. The purpose of a signature requirement is, therefore, one of proof: i.e. that the person purporting to agree to an agreement is actually that person.

In this respect, a wet ink written signature performs the proof function by acting as:

1. Proof of knowledge: i.e. the owner of the signature knows how it is created; and

2. Proof of nature: i.e. the handwriting of signature is a natural expression of the writer of the signature.

As acknowledged in the Department for Business, Energy & Industrial Strategy Guide to Electronic Signatures and Trust Services¹⁵, the impact of technological advances has been to expand what constitutes an acceptable signature, so that an e-signature may include “an electronic representation of a handwritten signature” or “a unique representation of characters”. These signatures also perform the proof function, usually through:

1. Proof of knowledge: i.e. only the person who knows the relevant representation of characters can complete the signature; and / or

2. Proof of possession: i.e. only the person who possesses the relevant data can complete the signature.

In this respect, therefore, the role of the signature is the same - it proves acknowledgement of binding terms - albeit the means by which that proof is provided may have altered.

Under the eIDAS Regulation¹⁶, a qualified electronic signature has the equivalent legal effect as a handwritten signature. “Qualified” refers to the fact that it will be an advanced electronic signature that is: uniquely linked to the signatory; is capable of identifying the signatory; is created using electronic signature creation data that the signatory can control; and is linked to the data signed therewith in such a way that any subsequent change in the data is detectable. A qualified signature goes one step further and is also given a qualified certificate which will be issued by a trusted service provider.

A key requirement and benefit of electronic signatures is the ability to trace where the signature has come from and therefore an audit trail would be important when considering the use of private keys. Typically, electronic signatures allow for the tracking of an IP address, a time, date and sometimes even a geo-location if signing on certain networks or servers.

Turning to private keys, therefore, these would naturally fit within the definition of a type of e-signature. In particular, as only the holder of the private key could use it, it would provide proof of possession, and, to the extent it could be copied, then it would constitute proof of knowledge. The alphanumeric code would, therefore, naturally fall within the category of “unique representation of characters”.

Lastly, it is worth quickly looking at this issue of fraudulent signatures. In this respect, wet ink signatures are relatively susceptible to fraud: once a wet ink signature is provided, a fraudster can copy that signature. In fact, to the extent that the signature is unknown by the recipient, a fraudster may not event have to copy a particular signature in order to impersonate another. By contrast, a private key cannot be copied through witnessing the public key - in order to make an impersonation, therefore, the fraudster would have to have actual access to the private key (and there would be no reason for the holder of that key to share it). So long, therefore, as the holder of the private key takes care to look after it, the risk of a fraudulent signature is far less when using a private keys than wet ink signatures. Again, a visible audit trail by electronically signing reduces the risk of fraudulent behavior. In this respect, therefore, there is some merit is preferring private key signatures to traditional wet ink signatures.

Could a statutory “in writing” requirement be met in the case of a smart contract composed partly or wholly of computer code?

In our view, this is the case. Again, as with signatures, there is a need for the meaning of “in writing” to develop to take account of changes in technology. In the ordinary course, an “in writing” requirement does not indicate a particular language, and indeed where this is disapplied it is generally done so using express language (for example as is the case as regards certain obligations under Directive 2014/65/EU, which in the UK requires certain communications to be “in English”). As coding languages, such as R, Ruby and Python, are still forms of language and communication, it would therefore be disingenuous to suggest that contractual terms in these languages are somehow not in writing, whereas more traditional languages are. To the extent that the recipient of a contract written in these terms does not understand the language used to express the relevant obligation, we would, in the same way, treat this as equivalent to an individual receiving a contract written in a traditional language not understood by the recipient.

Additional Areas for Consideration

It would be helpful to understand what status the “Legal Statement” will have in law. Equally it would be interesting to know if the Lord Chancellor intends to reserve any cases in the High Court to himself in the first instance.

It would be helpful for the Legal Statement to provide guidance as to what parties should do to ensure that English law/English courts will apply to the resolution of any disputes arising in relation to them. If different jurisdictions treat these agreements in different ways, we can see parties issuing declaratory proceedings in their preferred jurisdiction in order to avoid having another deal with it if a dispute arises.

It would be helpful if the Legal Statement could consider the appropriateness of existing remedies under English law to deal with breaches of property/contractual rights in relation to cryptoasset or smartcontract transactions (assuming they are characterized as such). Are the traditional remedies practicable and, if not, what are the likely alternatives to be?

While the consultation doesn’t touch on it, it would be helpful to know if the UKJT intends to do any work in relation to AI. It would be interesting to address how AI will be treated in law, who would own the IP and who is responsible for the acts of such code (particularly where that code is open source and anyone can contribute to its development).

How Eversheds Sutherland can help

We have an award winning crypto assets practice, with our work being “Highly Commended” at the Financial Times Innovative Lawyers Europe awards 2019, and were runners up for the Banking and finance Team of the Year with the British Legal Awards (Legal Week).

Since advising on the first successful initial coin offering in the United Kingdom, we have been at the forefront of advising firms on crypto assets, blockchain and the FCA sandbox process.

We take a flexible approach to providing advice, taking a cross-sectoral and practical approach tailored to the diverse needs of our clients. For further information, please do contact us.

---

1. p. 10, FCA Discussion Paper on distributed ledger technology: https://www.fca.org.uk/publication/discussion/dp17-03.pdf

2. https://bitcoin.org/bitcoin.pdf

3. For further discussion re the role of the attribute of data being rivalrous, see Professor Andrew Murray, Information Technology Law: The Law and Society.

4. P. 18, European Securities and Markets Authority Advice Initial Coin Offerings and Crypto-Assets

5. See Andrew Burnie, James Burnie, Andrew Henderson ‘Developing a Cryptocurrency Assessment Framework’ 10.5915/LEDGER.2018.121.

6. p. 84, Professor Andrew Murray, Information Technology Law: The Law and Society.

7. See Andrew Burnie, James Burnie, Andrew Henderson ‘Developing a Cryptocurrency Assessment Framework’ 10.5915/LEDGER.2018.121.

8. Para 1.107, Mann on the Legal Aspect of Money

9. Para 1.107, Mann on the Legal Aspect of Money

10. Para 1.17, Mann on the Legal Aspect of Money

11. Para 1.53, Mann on the Legal Aspect of Money

12. Para 1.84, Mann on the Legal Aspect of Money

13. E.g. see the Singapore Payment Services Act

14. See Andrew Burnie, James Burnie, Andrew Henderson ‘Plugging into a new world order, Reconciling cryptocurrency heterogeneity and regulatory continuity’, Journal of International Banking and Financial Law (2018) 2 JIBFL 83

15. https://www.gov.uk/government/publications/electronic-signatures

16. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=uriserv%3AOJ.L_.2014.257.01.0073.01.ENG