© Philipp Paech 2017
The inventors of blockchain technology aimed at creating self-governing and state-remote networks, as epitomised by Bitcoin. Nobody should be able to interfere in the governance of the network from outside the circle of its nodes: in particular, States should be unable to censor or regulate it. Instead, internal processes are deemed to balance all the relevant interests so that no judicial or regulatory intervention is needed. Nevertheless, since blockchain-based virtual currencies provide individuals with a means of payment and an easy and near-anonymous method of transferring value, States are considering relevant regulation, to date mainly targeting money laundering and terrorist financing. Beyond this very specific rationale, the role of blockchain financial networks in which securities, fiat money and derivatives are held could become so relevant in the future that societies will need to regulate and supervise them more consistently.
No addressee for enforcement
Effective regulation requires a suitable addressee against which the rules can be enforced. In the case of virtual currencies, such as Bitcoin, it appears difficult or well-nigh impossible effectively to regulate the person or persons controlling the software (which I call the ‘software platform provider’) as they are typically informally associated individuals that may be scattered around different jurisdictions. Regulators could, therefore, attempt to regulate these networks by forcing local Internet providers to block the relevant data traffic. However, this approach is only partly effective and politically and legally difficult to justify in a democratic setting as long as equally efficient, less intrusive means are at the regulators’ disposal. Against this background, regulatory initiatives at present target the intermediaries at the intersection between the virtual currency and the financial market, in particular, the so-called virtual currency exchanges, ie those entities exchanging fiat money for virtual currency. Regulators could take the same approach in relation to blockchain financial networks on which securities, fiat money and derivatives are held and transferred. However, it is moot whether this approach would suffice, in particular as there might be risks in this context that can only be addressed for a blockchain financial network as a whole.
What are the risks associated with blockchain financial networks?
As soon as financial assets such as fiat money, securities or derivatives are held and transferred through blockchain financial networks, the regulatory perimeter will need to extend to many more areas than just money laundering. This is due, first of all, to considerations of (market) scale: for the time being, transaction volumes in virtual currencies are tiny compared to those in financial assets. Then there is the question of interconnectedness: currently, incumbent market participants, ie banks and other financial institutions are likely to become nodes in blockchain financial networks and to administer their and their clients’ financial assets on them. Potential negative externalities rooted in the operation of the relevant blockchain network, as discussed in the following section, will immediately impact on those financial institutions and their clients. For instance, if a blockchain financial network was to produce unexpected outcomes because of a software bug or loophole, all financial institutions using this network would instantly face the same operational difficulties, and there would be no option to work around them individually. Also, the connection to the real economy would become much more immediate since, unlike virtual currencies, financial assets embody claims against corporate and State debtors.
Hence, legislators need to consider a number of risks. First, blockchain financial networks may influence the stability of financial markets. Secondly, self-governance within a network may cause distortions that lead to discrimination against parties that are unable to adjust their behaviour. Thirdly, the possibility of transferring financial assets on blockchain networks may render anti-money-laundering measures and similar rules partly ineffective.
Regulating poly-directional relationships
But how can regulation (and private law) be extended to blockchain financial networks in the most efficient manner? Disintermediation, leading to the abolition of accounts and intermediary-client relationships more generally will render traditional regulatory strategies largely inefficient and remove an important element to which private law rules traditionally attach. Instead, we must focus on what actually replaces the two-party relationship: a distributed network, built on poly-directional relationships among its nodes, which are linked solely through a software platform. Hence, regulation and law could target the software platform or the nodes, or both.
Regulating platform providers
Platform providers for first-generation blockchain applications are generally informally organised groups of individuals. Today, Fintech start-ups, well-established financial institutions and infrastructures, and even central banks may venture into setting up blockchain financial networks. There are a number of regulatory and legal aspects that can only be addressed for a blockchain financial network as a whole, regardless of how the circle of nodes is made up. The platform provider is the only suitable point of entry for network-wide regulatory and legal rules. Starting from basic requirements regarding safety, availability, integrity and continuity of service, any rules that can only be implemented centrally must be imposed on the platform provider. As a consequence, platform providers need to be legal persons (natural persons are too elusive) regulated by the State. There may still be state-remote, unregulated blockchain networks where the platform is provided under a more informal arrangement, such as for Bitcoin or Ethereum. However, it should be impossible to issue securities through these networks and they should not be dealing with fiat money. To achieve this goal, it is not necessary to close them down or block access to their websites. It is sufficient to prohibit regulated financial institutions from dealing with such networks.
Platform providers have to ensure the soundness and continuity of the software platform. Most importantly, this includes aligning the internal rules governing the acquisition of rights and the execution of contracts with private law. Turning the spotlight onto issues of systemic stability, the platform provider has to help prevent flash crashes and bubbles, not only by shaping the software accordingly but also by providing for relevant reporting mechanisms. Furthermore, in case of a permissioned network, the platform provider must administer admission to the network, ensuring non-discriminatory access to it and respecting relevant restrictions as to the circle of users or as to territorial reach.
Whether the platform provider should be the addressee of all relevant regulatory or legal rules depends on who the nodes of the blockchain network are: if the circle of nodes consists exclusively of regulated financial institutions (which may act as intermediaries and therefore maintain account-based relationships with clients), the regulatory burden can be shared between them and the platform provider. In this case, regulatory and legal rules that do not need to be implemented centrally are addressed to nodes. Generally, regulated financial entities will already be subject to relevant rules, such as an anti-money-laundering regime. However, relevant nodes must be authorised for the specific type of service provided by the network. For instance, if the network provides payment services, nodes authorised as banks will automatically be subject to all relevant regulation. Obversely, in a network administering securities, nodes authorised as payment services providers alone are not sufficiently regulated.
Where the nodes of a given network are entities not regulated as financial institutions, or individuals, the situation is completely different. In this case, there are no intermediaries that could apply relevant regulation to their relationships with clients. The only entity capable of applying the relevant regulation to the network and its nodes is the platform provider itself. In that situation, the platform provider would need to be the addressee of the full range of relevant regulatory and legal rules, thereby becoming a fully regulated financial institution itself which does not, however, maintain accounts with its nodes but controls the network through means of access control and programming of the network software.
Thus, ‘structure’ as the first determinant refers to who the nodes of a network are and what services it provides. As a rule of thumb, the application of regulatory and private law rules to blockchain financial networks requires less adaptation of existing rules to the extent that such networks are homogeneous as regards their circle of nodes and the services provided. For example, a network specialising in payments that has as its nodes only authorised payment service providers or banks will not pose any great problem from the point of view of regulation and private law. By contrast, a network for clearing securities transfers against cash settlement that also offers collateral management and has both non-financial corporations and regulated financial institutions as its nodes will be significantly more complicated to govern
 De Filippi and Loveluck, n 28, 3-4.
 See European Parliament, n 21); EU Commission, ‘Proposal for a Directive of the European Parliament and of the Council amending Directive (EU) 2015/849 [etc]’ (5 July 2016), COM(2016) 450 final; New York Codes, Rules and Regulations, Title 23 Chapter I Part 200 – Virtual Currencies at http://www.dfs.ny.gov/legal/regulations/adoptions/dfsp200t.pdf.
 See De Filippi and Loveluck, n 28, 8-10; V. Lehdonvirta and R. Ali, ‘Governance and Regulation’, in UK Government Chief Scientific Advisor, Distributed Ledger Technology: Beyond Blockchain (2016), 42 at https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/492972/gs-16-1-distributed-ledger-technology.pdf, visited 30 Nov. 2016.
 Wright and De Filippi, n 23, 51.
 EU Commission, n 70, 7; Lehdonvirta and Ali, n 71, 42. See New York State Regulation on Virtual Currencies (n 70) Section 200.2(q).
 Broadbent, n 16, 3.
 Lehdonvirta and Ali, n 71, 42, 43.
 See New York State Regulation on Virtual Currencies n 73 above, Section 200.8 (capital requirements), 200.9 (custody and protection of customer assets), 200.15 (anti-money laundering rules), 200.19 (consumer protection). European Commission, n 70, 7.
 See Bitstamp, http://www.bitstamp.net/payment-institution-license/, visited 30 Nov. 2016.