© Philipp Paech 2017
Blockchain came to be counted among the ‘disruptive’ technologies very early on, i.e., it was spotted as one of those typically Internet-based platforms that have the potential of unravelling traditional market structures, as has happened in other areas such as transport by taxi (Uber), holiday accommodation (Airbnb) and telecommunications (WhatsApp). Typically, disruptive technologies may modify the value chain of a traditional business, thereby threatening the incumbents’ income models.
Blockchain technology enables disrupters and incumbents to reconceptualise business models in financial markets. As a result, existing ways of trading and administering financial assets might change considerably were blockchain technology to be adopted on a wider scale. However, the resulting changes will affect some aspects that today serve as linchpins linking regulation and private law to market practice. These elements are deeply anchored in our understanding of how financial markets work and how we govern them. If they disappear or change, governance strategies will need to be adapted accordingly.
Three characteristics of blockchain technology have the potential of turning our understanding of how the market functions upside down, and may affect the current governance framework accordingly. First, the concept of the distributed ledger that lies at the heart of blockchain affects the central role of intermediation and client accounts or, more broadly, intermediary-client relationships; secondly, fail-proof, automated acquisition processes and immutable records replace trust in intermediaries and create a new form of truth (I explore this aspect separately here); and lastly, blockchain technology renders the execution of smart contracts truly unstoppable, which means that, in the absence of built-in circuit breakers, all human discretion is excised from the execution and enforcement of contractual duties (see here).
The three databases concepts
Financial transactions, such as the payment of money, the sale and purchase of securities, the exchange of currencies or derivative contracts, in principle represent a bilateral relationship between the relevant parties. However, they are typically concluded, administered or settled using intermediaries such as banks or brokers, and financial market infrastructures such as stock exchanges, payment systems, securities settlement systems or derivatives central counterparties. Intermediaries and infrastructures form networks that link financial market actors with one another. These networks are traditionally ordered either in a centralised or a decentralised fashion.
Centralised networks rely on a single record in which all transactions and holdings are recorded by a trusted central entity; only thus can market participants reach consensus on relevant facts, in particular, their holdings. In several countries, for instance, a central securities depository maintains securities accounts for all market participants that invest in securities. All acquisitions and dispositions are recorded in that register, and each individual balance is retrievable there.
Decentralised networks, on the other hand, are characterised by a structure in which different records together provide complete information on transactions and holdings. No single record on its own holds that comprehensive information. For instance, in some jurisdictions, the central securities depository records the transactions and holdings of banks and brokers but not of end-investors. The assumption is that these banks and brokers will record the identity of investors to whom the securities ultimately belong in their own ledgers.
Different as they may be, the centralised and decentralised financial network models do share an important feature: the original two-party relationship between the parties to a transaction (seller-buyer) is replaced by several two-party relationships between the parties and their intermediary and, as the case might be, between additional intermediaries providing the necessary links in the network. The technical process of recording an entitlement to an asset takes place on the IT system of the relevant intermediary. This record is associated with the legal relationship between the intermediary and its client, generally called an account or, more broadly, the client relationship. In modern financial markets, this account or client relationship is one of the linchpins of financial regulation and private law: property rights are defined by and contractual duties arise from it, as do a plethora of behavioural rules set by financial regulation.
By contrast, blockchain technology is based on the idea of a distributed record. Here, each participant in the network (‘node’), in practice a computer server controlled by a market participant and fitted with the relevant blockchain platform software, maintains a complete record of past transactions. All nodes are constantly updated with information on the latest transactions. As a consequence, all transaction information is available at any node at any given point in time, is identical and has equal constitutive value, ie there are no master and subordinated records. Thus, blockchain introduces an organising principle into the financial markets that is not built on a two-party relationship between investors and intermediaries and between intermediaries and infrastructures. There are no intermediaries, hence no accounts or other intermediary-client relationships within the blockchain network, so that an important linchpin of financial regulation and private law concepts is missing within the network itself.
Combined decentralised-distributed network
However, intermediation may still occur outside the network. Nodes may have clients which are not part of the network. In such a scenario they may transact on the network in their own name but on behalf of these clients, ie operate as intermediaries for persons outside the network.
The effect of disintermediation
Considering the ‘disintermediation’ within the network itself it becomes clear that there is enormous potential to change the market. In order to understand it, we must consider the current ecosystem of financial holdings and transactions. Financial intermediaries and infrastructures are only rarely involved in moving tangible assets around. Banks hold book-money in electronic accounts and transfer it through electronic payment systems. Similarly, shares, bonds and derivatives are typically incorporeal and purely account-based. In fact, the lion’s share of the services provided by the financial service industry relates to data storage and data processing.
Elimination of multiple records
However, the relevant IT systems they use differ considerably: as between different types of asset, different types of service provided in relation to an asset, different jurisdictions and even as between individual financial institutions. The same asset is typically mirrored repeatedly in different systems maintained by different entities, potentially in different jurisdictions. This historically generated multiplication and diversification of records and account relationships leaves ample room for inefficiencies and operational and legal risk: the constant reconciliation of these records is costly and slow; there are frequent temporary mismatches; investors are increasingly disconnected from issuers because the relevant investor rights are degraded down to the smallest feature common to all accounts used to hold a specific security; extracting aggregate data, for example for supervisory purposes, is a cumbersome exercise that often results in unsatisfactory results; as a given asset appears in different independent records it may be unclear which record is constitutive and which is only for book-keeping purposes; or, for the same reason, an asset may be used simultaneously by different parties, eg it might be pledged by different market participants for their own purposes, simply because the same asset appears in various accounts.
Greater data depth
In the case of distributed records used in blockchain networks, all parties involved in holding and administering an asset have an up-to-date copy of the same record at their disposal at all times, a record that is so designed as to exclude mismatches with the other copies. In addition, blockchain technology allows for greater data depth. That is, records are able to store more complex information than accounts typically can today. For instance, a traditional securities account with a broker records ownership of securities but nothing else. More in-depth information in relation to these securities needs to be generated and held in separate records. In a future blockchain-based setting, information as to ownership of a specific share could extend to information as to which service providers are involved in its administration, whether the share is encumbered and if so, in whose favour.
Recording and executing programme code and smart contracts
In addition, self-executing programmes, so-called ‘smart contracts’ (which I will discuss below), can be recorded together with the ownership information and could, for instance, automatically process dividend or interest payments once they are due.
In other words, the industry could move from a multitude of records relating to the same asset and maintained for different purposes, and which are not properly co-ordinated, to a single record distributed amongst and used by all parties, or at least significantly reduce the number of different records. Because the blockchain record is distributed amongst all nodes, the relevant financial institutions and infrastructures are able to provide their services in relation to a specific asset on the basis of the same information. Significant parts of the financial industry, including most ‘global players’, have identified these benefits as their common interest and have formed consortia supporting technology start-ups, such as the R3CEV and Hyperledger, that are currently developing the relevant blockchain software.
As a consequence, the considerable operational complications caused by multiple records could be removed in the future, as would be the associated uncertainty and cost. The speed of settling transactions would increase. At the same time, reporting to the competent supervisor would be facilitated, as the relevant data could be made available by giving the supervisor access to the blockchain record. 
 ibid, 5.
 See Unidroit, ‘Working Paper regarding so-called Transparent Systems’ (2006 Unidroit S78-44), http://www.unidroit.org/english/documents/2006/study78/s-78-044-e.pdf, visited 30 Nov. 2016.
 See P. Paech, ‘Securities, Intermediation and the Blockchain—An Inevitable Choice between Liquidity and Legal Certainty’, (2016) Uniform Law Review 21(4), 8-10.
 ibid, 15-16.
 Nakamoto, n 4, section 5; P. De Filippi and B Loveluck, ‘The invisible Politics of Bitcoin: Governance crisis of a decentralised Infrastructure’, 5(3) Internet Policy Review (2016), 7-8 at https://policyreview.info/articles/analysis/invisible-politics-bitcoin-governance-crisis-decentralised-infrastructure, visited 30 Nov. 2016.
 Wright and De Filippi, n 23, 2.
 See below .
 Paech, n 26, 15-22.
 See E. Micheler, ‘Custody Chains and Asset Values: why crypto-securities are worth contemplating’ (2015) Cambridge Law Journal 74(3), 509-519.
 See Euroclear and Oliver Wyman, n 14, 7.
 See Peters and Panayi, n 4, 22-23 for an overview of the various ledgers held within a financial institution for accounting and regulatory purposes.
 See Nakamoto, n 4, 3-4; Wall and Malmo, n 4, 8-16.
 P. Ortolani, ‘Self-enforcing Online Dispute Resolution: Lessons from Bitcoin’ (2016) Oxford Journal of Legal Studies 36(3), 595, 608.
 See Peters and Panayi, n 4, 24.
 See Peters and Panayi, n 4, 17, 27.
 See ibid, 18.