Smart Contracting And The New Digital Directives: Some Initial Thoughts

4

Smart contracts raise interesting questions about their legal nature. It is often only said that the existing smart contracts are neither particularly smart nor they are even strictly speaking legally binding contracts at all. [6] Any discussion about smart contracts must begin with the definition of the concept. [7] There are numerous definitions of what smart contracts are. [8] They are often defined as a special protocol intended to contribute, verify or implement the negotiation or performance of the contract in a trackable and irreversible manner without the interference of third parties. [9] One can go back to Nick Szabo, who in the 1990s, defined for the first time a smart contract as a:

“computerized transaction protocol that executes the terms of a contract. The general objectives of smart contract design are to satisfy common contractual conditions (such as: payment terms, liens, confidentiality, and enforcement etc.), minimize exceptions both malicious and accidental, and minimize the need for trusted intermediaries like banks or other kind of agents.ˮ [10]

5

Related economic goals of smart contracts include reducing loss by fraud, enforcement costs, or other transaction costs. They are presumed to be able to provide full transparency of the transaction and to grant a high degree of privacy contemporaneously.  [11] Szabo's definition can be simplified to a computer code that is created to automatically execute contractual duties upon the occurrence of a trigger event as a “digital condition precedentˮ, [12] or agreements wherein execution is automated, usually by a computer programme. [13] A minimum consensus definition can be distilled: a smart contract is a form of computer code which is self-executing and self-enforcing. [14] As the current smart contracts work without self-learning systems, it has to be emphasised that they neither need artificial intelligence nor deep learning. [15]

6

Needless to say, there are many debates and confusions on the legal concept of smart contracts. For blockchain-based smart contracts which will be discussed soon in this contribution, a useful dichotomy can be drawn between the “smart contract code”, which is the computer code stored, verified and executed in a blockchain, and the “smart legal contract”, which is a complement (or maybe even a substitute) for a legal contract to apply such technology. [16] In essence, a “smart legal contract” is a combination of the “smart contract code” and traditional legal language. [17] A smart contract is a computer code that specifies in ‘if this happens that shall happen’ language, in a way understandable to a computer. Once verified, it will self-execute and self-enforce by recognizing an occurred triggering event and dispensing the assets accordingly. [18]

7

It is evident that the term smart contract is a misnomer. [19] A smart contract, as we know it right now, is independent from the applicable law, as it is not a contract in the legal meaning. The choice of such name for the concept of a self-executing and computer-coded agreement is unfortunate as it exacerbates confusion. Some theoretical similarities, however, exist between smart contracts and legal contracts insofar as both ‘are frameworks for regulating the interaction between different entities. [20]

8

As for the question regarding how a smart contract works in practice and how it is concluded, Szabo uses his famous vending machine analogy. [21] A vending machine takes coins and dispenses change and product according to the displayed price. Once the coins are inserted, there is no further human intervention required to conclude and later execute the contract. Similar to a smart contract, a contract concluded through a vending machine is also in principle immutable and self-enforcing. Even if a person were forced to buy something from the vending machine, the machine would still give the product to the person regardless the fact that the transaction is legally invalid ex tunc due to duress. Furthermore, in theory, anybody with coins can participate in an exchange with the vendor regardless of the legal capacity of the contracting parties. Where smart contracts go further is “in proposing to embed contracts in all sorts of property that is valuable and controlled by digital meansˮ. [22] Essentially, once both parties agree on a smart contract, its execution is taken from their control.

9

Smart contracts do not necessarily require blockchain technology. [23] However, there is little doubt that the main reason for the increasing importance of smart contracts is the rise of blockchain technology, as it allows smart contracts to use their full automation potential. Bitcoin, which proliferated this technology, led ultimately to the establishment of Ethereum, a sophisticated and prominent blockchain platform allowing more complicated (i.e. smart contract) transactions beyond transfers of virtual currencies. [24] In the meanwhile several other blockchain-based smart contract platforms such as Hyperledger Burrow, Hyperledger Fabric, Open Transactions, and Quorum have also entered the market. The blockchain technology demonstrates how a network could be set up so that once a transaction is set in motion, the network can produce outputs autonomously without the direct intervention of any party or other intermediaries. [25] Because of this feature, it is often said that the contracting parties do not need to trust each other, they can rely on the system as a whole to carry out transactions knowing that the other party cannot frustrate the intended outcome. Blockchain not only allows verification of each transaction through the nodes (the computers in the chain), but it also, by storing the contract in a “blockˮ and sending it to each node, makes the execution automatic and, in principle, immutable. Thus, smart contracting allows the “digitization of trust through certainty of execution” and the “creation of efficiency through removal of intermediaries and the costs they bring to the “transactionsˮ. [26] These characteristics are perhaps the greatest appeal of blockchain-based smart contracts.

10

When describing the actual process of formation of smart contracts, the concept can be best explained through Ethereum's process. [27] First, the user first types out the contract in Ethereum’s coding language called “solidity”, [28] for which the user has to download the Ethereum software and be part of its network. Then he will “propose” a specific contract by making it available in the system. The contract will have its own identification number and functions as an autonomous entity within the system. Another user may then accept the proposed contract by communicating to it. For instance, he communicates by making a payment, regularly in “Ether (ETH)”, the virtual currency of Ethereum. After that communication of the other party, the smart contract will execute itself. It is important to note that to conduct a transaction or to execute a contract on the Ethereum blockchain platform the users need to pay “gas”, which is a computation fee. [29] Gas is priced in small fractions of Ether called “gwei” and it is used to allocate resources of the Ethereum Virtual Machine (EVM) so that decentralized applications such as smart contracts can ultimately self-execute in a secured but decentralized way. The fee is paid to the miners for mining transactions, putting them into blocks. [30] The exact price of the gas is determined by supply and demand between the network's miners. They can decline to process a transaction if the gas price does not meet their threshold, and users of the network who seek processing power.

11

There are many (potential) fields of application for smart contracts. Besides the well-known smart refrigerator example (the refrigerator “orders” automatically food or beverages within a previously concluded delivery smart contract) the “pay as you drive-principle” is subject to discussions in the insurance industry right now. [31] Here the policyholder concludes a (smart) car insurance contract with the insurance company. The contract contains a “pay as you drive-provision” which means the riskier the policyholder drives, the higher his premium. For data collection, the policyholder's car has a blockchain interface and the blockchain-based smart (insurance) contract automatically adjusts the amount of the payable premium according to the manner the insured car is driven. A similar idea is “drive as long as you pay” where a car can only be driven as long as the premiums are paid. If premiums have not been paid, the blockchain-based smart insurance contract uses the smart lock of the car to block the further use of the vehicle. [32] There is also the idea of combining smart contracts and smart meters in order to automatically cut off the supply of gas, water, and electricity in case of unpaid bills.  [33]

14

The opening of the scope of application of the New Consumer Sales Directive could be problematic in the present case because no payment in a regular currency such as the Euro or US-dollar was agreed; however, a “payment” in a virtual currency (here “Ether”) was provided for. This is not necessarily a sole smart contract problem, because “ordinary non-smart contracts” can provide for a “payment” in a virtual currency. Nevertheless, it is currently the case that this problem arises primarily with smart contracts, which is why it seems justified to identify this primarily as a “smart contract problem”.

15

But what exactly is the problem with the New Consumer Sales Directive and “payment” in virtual currency? Let us take a closer look at the provisions of the Directive. Art. 3(1) of the New Consumer Sales Directive states that “(t)his Directive shall apply to sales contracts between a consumer and a seller.” [34] Art. 2 no. 2 of the New Consumer Sales Directive defines the term “sales contract” as follows: “'sales contract' means any contract under which the seller transfers or undertakes to transfer ownership of goods to a consumer, and the consumer pays or undertakes to pay the price thereof”.  [35] Certainly, “payment of price” covers payments in regular currency. But are “goods for virtual currency contracts” also covered by the New Consumer Sales Directive? The Directive itself does not provide any further explanation of what is meant by a “payment of price” according to Art. 2 no. 2 of the New Consumer Sales Directive. The legislative history and the literature are also, as far as can be seen, unproductive in solving this question.

16

If this were not in itself already a significant problem for the interpretation of the New Consumer Sales Directive, the situation becomes even more confusing when one looks at the Digital Content Directive. Both directives are to be understood as twin directives, where the New Consumer Sales Directive covers the area of goods and goods with digital elements, while the Digital Content Directive regulates the supply of digital content and digital services. According to Art. 3(1)1 of the Digital Content Directive, the Directive applies “to any contract where the trader supplies or undertakes to supply digital content or a digital service to the consumer and the consumer pays or undertakes to pay a price.”  [36] However, unlike the New Consumer Sales Directive, the Digital Consumer Sales Directive defines the term “price” in Art. 2(7) Digital Content Directive, according to which “price” means money “or a digital representation of value that is due in exchange for the supply of digital content or a digital service”. [37] Recital 23 of the Digital Content Directive provides further information on the background to the inclusion of “digital representations of value” in the Digital Content Directive and what exactly is meant by this. According to that recital:

“(...) (d)igital representations of value should also be understood to include virtual currencies (...). Differentiation depending on the methods of payment could be a cause of discrimination and provide an unjustified incentive for businesses to move towards supplying digital content or a digital service against digital representations of value. (…)”. [38]

17

The Digital Content Directive thus makes it unmistakably clear that the scope of application of the Directive is also open in the cases of “payments in virtual currency” and gives a convincing reason for this. After all, with the increasing popularity of virtual currencies as a means of payment, this is the only way to prevent companies from escaping the requirements of the Digital Content Directive by demanding “virtual currencies payments” with consumers. If our example case had been about the supply of digital content or digital service and not about goods (with digital elements), the scope of application of the Digital Content Directive would undoubtedly have been given. Why the European Union is obviously pursuing two different approaches regarding payment in virtual currencies, or at least is introducing a great deal of interpretational uncertainty into the New Consumer Sales Directive, is not apparent and also eludes a deeper logic. Of course, there is the same incentive for businesses to escape the scope of the New Consumer Sales Directive by demanding a “payment” in virtual currencies and thus to undermine its protection standard as with the Digital Content Directive. Ultimately, this is precisely why there could be an increased use of virtual currencies in the area of consumer sales in the future.

18

Overall, the result to the problem of “payment in virtual currencies” can be described as disappointing. Firstly, because the New Consumer Sales Directive seems to contain a loophole that could make it possible for businesses to systematically undermine the standard of protection. [39] And this issue might even become a potential preliminary question for the European Court of Justice to answer whether “goods for virtual currency contracts” must be considered as “sales contracts” under the New Consumer Sales Directive. On the other hand, it also seems to be a missed opportunity to further harmonise the acquis communautaire in a meaningful way, even though the EU member states are free to extend the scope of the New Consumer Sales Directive to contracts in which goods are purchased for virtual currencies. Nevertheless, a clear solution for both digital directives in this regard is preferred.

19

Let us now assume that in our example the present contract is undoubtedly a “sales contract" in the sense of the New Consumer Sales Directive, because payment potentially had to be made in euros and not in a virtual currency. The starting point, that it is a sale of “goods with digital elements”  [40] according to Art. 3(1), (2) and Art. 2 no. 5 of the New Consumer Sales Directive, is also undoubted. The car sold is a tangible movable item according to Art. 2 no. 5 of the New Consumer Sales Directive that incorporates or inter-connects with digital content or a digital service (here, the smart lock with its digital functions) in such a way that the absence of that digital content respectively digital service would prevent the good (here, the car) from performing its functions. The New Consumer Sales Directive therefore applies to the car itself including its smart lock.

20

The elephant in the room is, of course, whether the New Consumer Sales Directive also covers the (defective) smart contract component of Ethereum (as a “digital element” of the car), which was ultimately responsible in the example for the consumer no longer being able to use the car. Alternatively, is the defective smart contract component rather regulated by the Digital Content Directive since this Directive is ultimately aiming at regulating the supply of digital content and digital services? The importance of deciding which directive covers the smart contract component quickly becomes apparent when considering the legal consequences of this decision. If one concludes that the New Consumer Sales Directive also covers the defective smart contract component, the consumer has direct rights arising from that non-conformity against the seller. The seller's liability risk would then increase accordingly, even though he would have a right of redress under Art. 18 of the New Consumer Sales Directive against Ethereum after the consumer has made a claim against the seller. [41] If, on the other hand, the Digital Content Directive were to apply to the smart contract component, the consumer would have to turn directly to Ethereum, as the digital content comes from the digital service provider itself; in this case, the seller could not be held liable for the defective smart contract element.

21

On the whole, the better arguments seem to speak in favour of also subjecting smart contract elements to the scope of the New Consumer Sales Directive as a “digital element” and not to the Digital Content Directive. The wording of Art. 3(3)2, Art. 2 no. 5 of the New Consumer Sales Directive is broad enough to justify such an interpretation. It could be argued that the car is interconnected in such a way with the smart contract that the absence of it prevents the car from performing. Finally, the example shows that the car cannot be used without a faultless smart contract. The fact that the smart contract element comes from a third party (in the example, Ethereum) and not from the seller is irrelevant. [42] Also, a broad interpretation of Art. 3(3)2 of the New Consumer Sales Directive could justify that the digital service running the smart contract is also provided with the car under the sales contract.

22

Nevertheless, it must be admitted that the wording of Art. 3(3)2, Art. 2 no. 5 of the New Consumer Sales Directive would also allow for another, narrower interpretation. Accordingly, the strongest reason for including the smart contract component as a “digital element” in the scope of the New Consumer Sales Directive seems to be the aim of an effective consumer protection. For if one takes the perspective of the consumer, it becomes clear that it will often be impossible, or at least very difficult, for him to see why the purchased goods do not work. In our example, how is it possible for the consumer to realise why the smart lock cannot be opened? In the end, he will not be able to recognise whether the smart lock itself is defective or whether the problem comes from the area of the smart contract. The consumer cannot be burdened with such an obligation to examine, especially since he will usually lack the necessary expertise anyway. The objective of the New Consumer Sales Directive was to establish a one-stop-only policy for the consumer in the marginal area of goods and digital content and digital services in order to ensure effective consumer protection. If one really wants to meet this objective, the smart contracts elements should be subject to the New Consumer Sales Directive. That this leads to an extended liability of the seller for defective digital content or digital services of third parties is as previously mentioned not alien to the New Consumer Sales Directive, but is part of its concept. [43] As a final “ultima ratio argument” for the inclusion of the smart contracts component into the scope of the New Consumer Sales Directive, Art. 3(3)3 of the New Consumer Sales Directive can be cited, according to which “(i)n the event of doubt as to whether the supply of incorporated or inter-connected digital content or an incorporated or inter-connected digital service forms part of the sales contract, the digital content or digital service shall be presumed to be covered by the sales contract.”  [44]