Blockchain and the law - Regulation for smart contracts on the way?

The Research and Documentation Centre of the Dutch Ministry of Justice and Security (WODC) has recently published the report "Blockchain and the law". The August 2019 report is fairly critical towards blockchain technology, and questions whether the potential benefits of blockchain (mainly efficiency gains) are proportional in relation to the potential disadvantages. The report also analyses the benefits and disadvantages of smart contracts. The Dutch government is expected to present a formal response to this report before the end of the year.

The growing interest in smart contracts, alongside the potential they offer, should encourage legal practitioners and prospective users and developers of smart contracts to consider if and how smart contracts fit into the existing framework of Dutch contract law. This blog post starts with a brief description of what smart contracts are. The following paragraphs discuss some of the legal issues related to smart contracts that are addressed in the report: namely, the conclusion and interpretation of smart contracts, and how to deal with normative standards. Another issue, related to groups of contracts, will also be discussed. This blog post ends with a short conclusion.

What are smart contracts?

Smart contracts are computer protocols that execute pre-determined rules of code: the contractual terms are embedded in software, generally on a blockchain (although smart contracts do not necessarily have to be blockchain-based). Smart contracts cannot be altered retroactively, and will automatically execute or effectuate legal agreements (i.e. without the cooperation or intervention of the contracting parties or a third party) once they are recorded on a blockchain.

Smart contracts are not really 'smart'; they cannot (or at least not yet) think or develop, and are not a form of Artificial Intelligence (AI). A smart contract can merely effectuate the automatic execution of certain terms of an existing 'traditional' contract, but it can also be considered as proof of the conclusion of a legal agreement (i.e. the smart contract itself sets out the legal relationship between the contracting parties). However, as the report also notes, some contractual terms may be hard to put into code, such as forum selection clauses.

One of the main benefits of smart contracts is that they will always be executed in the way they were coded, without requiring the involvement of the parties or a judge, significantly reducing the chances of non-performance of the contract. A downside of this is that once smart contracts have been placed on a blockchain, they cannot be modified - even if the parties later agree or the law requires that the contract is altered or annulled - unless the option to modify the smart contract was built into the original code. However, it is possible to place an updated ('repaired') version of the smart contract on the blockchain that will be used instead of the old version. 

Conclusion of a smart contract

The report considers that the rules of Dutch contract law apply to smart contracts. Pursuant to section 6:217 of the Dutch Civil Code (DCC), a contract is formed by an offer and its acceptance. To determine whether a contract has been formed, one should always consider what the parties can deduce from each other's statements or conduct regarding their mutual intentions, and what legal consequences can be derived from this (this is termed 'wilsvertrouwensleer' and follows from sections 3:33 and 3:35 DCC).

Dutch law generally does not require any specific formalities for a contract to be formed: a contract can be concluded in computer code on a blockchain, as with smart contracts. Therefore, Dutch contract law in principle acknowledges and allows for smart contracts as 'contracts' and the report therefore concludes that no additional regulation is required to ensure the validity of smart contracts under Dutch law.

Interpretation of smart contracts

Pursuant to Dutch law, the interpretation of contracts is governed by what is termed the Haviltex standard. According to this standard, to determine the parties' legal rights and obligations towards each other, one should not only look at the wording of a contract but also take into consideration all the specific circumstances of the case, subject to the principles of reasonableness and fairness.

How should the Haviltex standard be applied to smart contracts? Since smart contracts are laid down in computer code, some argue that there is no room for any interpretation; it is solely the lines of code that determine the legal relationship between the contracting parties ("code is law"). However, the report also stipulates, if a smart contract is governed by Dutch law, the general interpretation rules (i.e. the Haviltex standard) will apply. This means that to determine the legal relationship between the parties, not only the actual code of the smart contract but also other facts and circumstances (such as the parties' intentions), may be relevant.

The report acknowledges that the fact that the parties chose to contract through a smart contract could indicate that, for the interpretation of the contract, decisive weight may be given to the code of the contract. In this way, the interpretation of smart contracts could perhaps be similar to the interpretation of commercial contracts between professional parties where, under certain circumstances and as a starting point, overriding significance may be given to the text of the agreement.

Dealing with open, normative standards

Another problem highlighted in the report is that some contractual principles under Dutch law are hard to implement into smart contracts. For example, a continuing performance contract (such as a lease) may be terminated on the grounds of the 'additional effect of the principle of reasonableness and fairness'. In addition, under certain circumstances banks and other financial services providers may have a special duty of care when dealing with their customers. Another example of a normative concept is 'force majeure': a debtor is not liable for damages as a result of failure in performance of an obligation if this failure was the result of a force majeure.

Such open, normative standards cannot (or cannot easily) be laid down in computer code since there are generally no strict rules that determine whether the standard is met; it depends on the specific circumstances of the case. A way to implement such open standards into smart contracts may be to implement in the code a role for an external, third party (an 'oracle') to judge whether the normative standard is met. However, the use of an oracle implies that the execution of the smart contract depends on a third party’s judgment. Automatic execution of the smart contract, one of the main benefits of the technology, may therefore no longer be guaranteed. The report suggests that this problem may be solved by AI in the future. However, this remains somewhat doubtful as AI generally relies on extensive and relevant datasets which may not always be readily available in the context of smart contracts.

Groups of contracts

One issue that the report does not address concerns the use of smart contracts in the context of groups of contracts. Under certain circumstances Dutch law acknowledges a connection between separate contracts, which may result in legal consequences, e.g. the termination of a related contract. This may lead to complications if one of these contracts is a smart contract. For example, if a seller fails to perform its obligations under a purchase agreement, this may give the purchaser the right to suspend its obligations under a related financing agreement. However, if the financing agreement is a smart contract, the interest payments could automatically be executed regardless of the purchaser's right to suspend these payments once the seller defaults under the purchase agreement.

One way to solve this problem would be to draft both agreements as smart contracts and link the codes in such a way that a default under one contract immediately stops the execution of the other contract. However, this may be complicated depending on the number of contracts and contracting parties involved, and even more difficult if a normative standard is involved (see above).

Conclusion

In remains to be seen to what extent smart contracts will play a role in contract law in the Netherlands. How legal relationships may be established through smart contracts will depend on how judges, lawyers and the legislature deal with the legal challenges posed by smart contracts. The research report "Blockchain and the law" addresses some interesting issues, but does not provide particularly concrete directions as to how the legislature should act now. It will be interesting to see what steps the Dutch government intends to take - if any - following this report.