Is European
competition law ready
for the blockchain?
F.J.P.M. Haverhals (12371076) Supervised by Dr. M. Weimer
LLM European Competition Law and Regulation (International and European Law)
6 January 2020 12.940 words
florence.haverhals@outlook.com
ABSTRACT
Blockchain is still a little-explored territory, yet a more frequently used platform for various purposes. European competition law issues regarding the blockchain technology are still at an early stage of academic debate. This thesis has identified several types of anticompetitive behaviour on the blockchain that raise challenges for the legal concepts of European competition law. In establishing an abuse of a dominant position, the use of the as efficient competitor-test is problematic. An agreement to constitute a blockchain may be considered anticompetitive under Article 101 TFEU when it has the object to restricting competition. However, it is unsure if collusive behaviour on the blockchain should be labelled as a unilateral act or an agreement or coordination amounting to anticompetitive conduct. The determination of who is to be held accountable for the collusive behaviour on the blockchain, based on the AC-Treuhand-judgement, leaves open the broad interpretation of the concept of indirect contact. Furthermore, following the Eturas-judgement, if collusive behaviour is just the result of the use of the system, without other indicia, how can the 'anticompetitive' conduct fall under the scope of Article 101 TFEU? Moreover, the Anic-presumption circumvents the inherent features of immutability and decentralisation of the blockchain, leaving this exemption useless. Problematic is tacit collusion occurring through the use of smart contracts, which is currently not labelled as an agreement, and consequently escapes the loop of European competition law. This thesis calls for a novel regulatory approach, explicitly targeting the aspects of blockchain that have the potential to escape the prohibitions laid down in European competition law. This regulation should be based on a 'results-based approach' to effectively prohibit anticompetitive behaviour occurring on the blockchain. European competition law is presently not optimally suited to fight anticompetitive conduct on the blockchain as it misses out on fundamental characteristics of the blockchain, such as immutability and decentralisation. A novel regulatory approach targeting blockchain is needed before European Competition law becomes ineffective in regards to the technology.
TABLE OF CONTENTS
List of Abbreviations ... 3
Definitions ... 4
Introduction ... 6
1 How does blockchain technology work, and how is it related to European competition law? ... 9
1.1 The blockchain paradigm ... 9
1.2 Types of blockchain ... 10
1.2.1 Public and permissionless blockchain ... 11
1.2.2 Private and permissioned blockchain ... 11
1.3 Decentralised applications ... 12
1.4 Competition on and off the blockchain ... 14
1.5 Blockchain and its link to European competition law ... 14
1.6 Interim Conclusion ... 15
2 What are the implications of the blockchain for the legal concepts of European competition law? ... 17
2.1 The blockchain market ... 17
2.1.1 The 'token effect' ... 17
2.1.2 Blockchain as zero-price market ... 19
2.2 Concept of undertaking ... 20
2.2.1 Blockchain as a single economic-entity ... 22
2.3 Interim conclusion ... 23
3 Does blockchain give rise to any (new) legal issues for European competition law? 24 3.1 Competition law analysis of the blockchain under Article 102 TFEU ... 24
3.1.1 Acquiring a dominant position ... 24
3.1.2 Refusal to deal ... 26
3.2 Competition law analysis of the blockchain under Article 101 TFEU ... 27
3.2.1 The agreement creating a blockchain ... 27
3.2.2 Blockchain as a tool for sharing sensitive data ... 28
3.2.3 Smart contracts and liability issues ... 31
3.2.3.1 Smart contracts easing collusion ... 31
3.2.3.2 Ambiguity in regards to liability ... 32
3.2.4 Concerted practices on the blockchain ... 34
3.2.4.1 Concurrence of wills obsolete for blockchain? ... 35
3.2.4.2 Flaws of the Anic-presumption ... 37
3.2.4.3 Prohibited tacit collusion ... 39
3.3 Interim conclusion ... 41 Conclusion ... 42 Bibliography ... 46 Appendix I ... 51
LIST OF ABBREVIATIONS
AEC-test As-efficient competitor test
CJEU Court of Justice of the European Union
ESMA European Securities and Markets
Authority
Guidelines Guidelines on Horizontal Cooperation
Agreements
OECD Organisation for Economic Co-operation
P2P network Peer-to-peer network
TFEU Treaty on the Functioning of the
European Union
DEFINITIONS
Bitcoin A cryptocurrency of which every single
transaction is recorded on a blockchain, operating independently of a central bank.
Block Data structure used in blockchains to
group transactions.
Consensus protocol When using the blockchain, every user agrees to a specific set of procedures. This ensures convergence towards a single, immutable version of the ledger. Consortium blockchain A private blockchain in which the
consensus protocol is set up by a pre-selected set of nodes.
Ethereum Open and permissionless blockchain;
most common blockchain where smart contract operate on.
Hash A unique fingerprint of each block that
represents each block through a set of characters and numbers.
Lex cryptographia A set of algorithms and rules in the blockchain.
Miners The nodes that group transactions into
new blocks and suggest them to the existing network.
Nodes Computers that save a local version of the
distributed ledger.
Oracles A single person or a group of persons,
who feeds the software relevant information. Oracles verify whether an execution-relevant event occurred.
Proof-of-work The most commonly used consensus
protocol for public and permissionless networks.
Private and permissioned blockchain A blockchain that can run on a private network, only participants can read and
Public and permissionless blockchain A blockchain that anyone can read and on which anyone can propose new transactions.
P2P network There is no central authority that controls
the blockchain; instead it uses a network with the nodes serving as different peers. Semi-private blockchain A blockchain in which a single company
manages the blockchain, and access to enter the blockchain is granted to any qualified user.
Single entity blockchain A private blockchain in which a single entity will design the protocol and
manage the blockchain.
Smart contract
Token effect
Pieces of code stored on the blockchain, that are automatically performed once deployed, enhancing the trust and security of the blockchain network. A lower incentive to join the blockchain when more users use it, contrary to a network effect in other digital platforms.
INTRODUCTION
As the price for Bitcoin reached unexpected heights since its launch in 2009, it has garnered more attention of users as well as of the European Commission. This shifting focus of the Commission towards creating a digital market single is not solely based on its popularity; the technology has the potential to disrupt important economic, social, and political structures.1
The potential use of blockchain is extensive and includes tracking, verification, authentication, and record-keeping of data. Moreover, its use could include storing and automating contracts, loans, titles, mortgages, and records.2 Blockchain is more about value rather than just information.3 Blockchain's potential to create new ecosystems in trade, investments, and commercial transactions could have profound impacts on market competition. Therefore, blockchain technology inherently calls for the application of European competition law.
A blockchain is an open and distributed ledger that can record all sorts of transactions between users. The fundamental characterisations of the blockchain include decentralisation,4 pseudonymity, or even anonymity for that matter, immutability, and automation.5 The transactions in a blockchain follow a set of algorithms and rules, the lex cryptographia.6 In theory, all users of the blockchain see each transactions existence, but the purpose or rationale behind a transaction is kept secret because the transactions are encrypted.7 A unique alphanumeric address identifies all users of the blockchain - accordingly, users of a blockchain act in a pseudonymous manner. 8 Once recorded on the blockchain, information and transactions are permanent.9
1 EU Blockchain Observatory and Forum, 'Thematic Report - Legal and Regulatory Framework of Blockchains and Smart Contracts' (27 September 2019) 9 <https://www.eublockchainforum.eu/reports> accessed 4 January 2 Rosario Girasa, Regulation of Cryptocurrencies and Blockchain Technologies (1st edn, Palgrave Macmillan 2018) 44.
3 Lokke Moerel, 'Blockchain & Data Protection … and Why They Are Not on a Collision Course' (2019) 6 European Review on Private law 825, 833.
4 No central authority governs the blockchain. 5 EU Blockchain Observatory and Forum (n 1) 5.
6 Thibault Schrepel, 'Is blockchain the death of antitrust law? The blockchain antitrust paradox' (2018) 3 Georgetown Law Technology Review 281, 287.
7 Robert P. Murphy & Silas Barta, Understanding Bitcoin (1st edn, Createspace independent publishing platform 2015) 52.
8 Phil Champagne, The book of Satoshi: the collected writings of Bitcoin creator Satoshi Nakamoto (1st edn, e53 Publishing LLC 2014) 136.
European competition law issues regarding the blockchain technology are still at an early stage of academic debate. Speeches of Margrethe Vestager, the current Commissioner for Competition, indicate a shifting focus towards the use of algorithms by undertakings as well as upcoming large data platforms and its applications.10 The blockchain technology has not been explicitly mentioned in this context. However, at the working staff level, the Commission closely follows current developments regarding blockchain technology in the field of financial services.11 Recently, the Commission launched an examination relating to the legal implications of the blockchain.12
The Blockchain Observatory and Forum13 (launched by the European Commission and the European Parliament), as well as the Organisation for Economic Co-Operation14 ("OECD") call for scrutiny under (European) competition law. The EU Blockchain and Observatory Forum mentions collusive conduct covered by Article 101 Treaty on the Functioning of the European Union15 ("TFEU") and abuse
of dominant position on the blockchain as prohibited by Article 102 TFEU, as the antitrust domains in which the blockchain might have the most disruptive impact.16
All significant new technologies that will act as catalysts for society, such as the blockchain, will at some point intersect the existing legal framework.17 Nevertheless, until today, the Court of Justice of the European Union ("CJEU") has not yet rendered case law specifically mentioning blockchain technology and its relationship with European competition law. Moreover, the European Union has thus far not taken legislative measures that are explicitly regulating the blockchain technology.
10 Falk Schöning & Myrto Tagara, 'Blockchain: Mind the gap! Lessons learned from the net neutrality debate and competition law related aspects' (2018) 3 Concurrences 5.
11 Falk Schöning & Myrto Tagara, 'Competition Law in the competition era' <
https://coinlaw.io/competition-law-in-the-blockchain-era-falk-schoning-and-myrto-tagara-of-hogan-lovells-brussels/> accessed 4 January 2020. 12 European Commission, 'Study on Blockchains, Legal, Governance and Interoperability Aspects' (12 December 2018) <
https://ec.europa.eu/digital-single-market/en/news/study-blockchains-legal-governance-and-interoperability-aspects> accessed 4 January 2020. 13 EU Blockchain Observatory and Forum (n 1).
14 OECD 'Algorithms and Collusion: Competition Policy in the Digital Age' (2017)
<https://www.oecd.org/daf/competition/Algorithms-and-colllusion-competition-policy-in-the-digital-age.pdf> accessed 4 January 2020.
15 Consolidated Version of the Treaty on Functioning of the European Union [2012] OJ C326/47. 16 EU Blockchain Observatory and Forum (n 1) 20.
Consequently, the main research question in this thesis is formulated as: 'Does blockchain technology call for the application of European competition law, and if so, what challenges will it raise for the legal concepts of European competition law?' The research is split into three parts. The first part addresses the working of the blockchain technology and its link to European competition law (1). The second part focuses on the implications of the blockchain on the legal concepts of European competition law (2). The third part analyses if the blockchain gives rise to any (new) legal issues that are not covered by the current legal framework (3).
1 HOW DOES BLOCKCHAIN TECHNOLOGY WORK, AND HOW IS IT RELATED TO EUROPEAN COMPETITION LAW?
Blockchain is a technology that is rapidly evolving, together with new functionalities. This chapter will discuss the working of the blockchain technology and its link to European competition law. The analysis will only focus on the fundamental parts of the technology that are relevant to European competition law.
1.1 The blockchain paradigm
As the etymology suggests, a blockchain is a chain of blocks. Blockchain is a shared and synchronised database that is stored on multiple nodes (computers that save a local version of the distributed ledger). A block bundles different transactions and then adds a new block to the existing blockchain. Instead of a central authority that controls the chain, blockchain uses a peer-to-peer network18 ("P2P network"), with the nodes serving as different peers.19 All in all, blockchain technology eliminates the need for a fiduciary, such as a bank.
The blocks in the ledger contain different vital components. A hash is the unique fingerprint that represents each block through a set of characters and numbers.20 Besides its unique hash, a block contains a timestamp and the hash of the previous block, which in turn creates the sequential chain of blocks.21 Data is blocked in groups that are chained to the ledger through a hashing process. The hash is created automatically by the blockchain, making it nearly impossible to modify or hack, as it will automatically change a hash in the case the data recorded in the block is modified. A copy of a blockchain will be invalidated as such.22 This cryptographic hash chaining makes the log tamper-evident, which increases the accountability of the blockchain. Because of this theoretical immutability of the blockchain, it is said, 'unlike Pinocchio, the blockchain does not lie.'23
18 A peer does not necessarily mean an individual joining the blockchain; a node may very well be a corporate or any other entity.
19 Michèle Finck, Blockchain Regulation and Governance in Europe (1st edn, Cambridge University Press 2019) 6. 20 Ibid 7.
21 Andreas Antonopoulos, Mastering Bitcoin (2nd edn, O'Reilly 2017) xxxiii. 22 Schrepel (n 6) 288.
23 Don Tapscott & Alex Tapscott, A Blockchain revolution: how the technology behind bitcoin is changing money,
When using the blockchain, every user agrees to a specific set of procedures, called the consensus protocol. Once this protocol is set, it is theoretically impossible to deviate from the protocol unless the majority of users agree to modify it. The immutability contributes to creating trust in the blockchain.24 The consensus protocol enables the distributed network to agree on the current state of the network in the absence of a central authority. The consensus protocol governs the way new blocks must be added to the chain.25 A possible scenario that intersects the decentralisation in the blockchain is a 51% attack. In that case, an entity is somehow able to take control (through the majority vote) of the consensus protocol.26
For nodes to accept new blocks, they need to be generated by miners (the nodes that group transactions into new blocks and suggest them to the network) following the relevant consensus protocol. All nodes participate in the validation of new blocks. When the node concludes the new block is valid under the consensus protocol, it adds the block to its local copy of the ledger and broadcasts it to other nodes in the network.27
1.2 Types of blockchain
There is a wide variety in the types of blockchains and their governance structures. There is no predefined set of structures as blockchains are most commonly utilised in a hybrid form. The European Securities and Markets Authority ("ESMA") has issued a report28 in which it sets out that blockchains are programmed to operate as a public and permissionless blockchain or as a private and permission-based blockchain.
These different types of blockchains are classified based on several factors, such as how consensus is achieved, through its software management, visibility, the possibility to identify transactions on the ledger, and the right to add new data to the ledger.29
24 Schrepel (n 6) 289. 25 Finck (n 19) 7.
26 Jamil Civatarese, 'Risk Aversion and 51% Attack Resistance in Proof-of-Work Cryptocurrencies' (2018) SSRN 1 <https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3289317> Accessed 4 January 2020.
27 European Commission, 'Remarks by Vice-President Dombrovkis at the Roundtable on Cryptocurrencies' (Speech) <https://europa.eu/rapid/press-release_SPEECH-18-1242_en.htm> accessed 4 January 2020. 28 ESMA, 'Report on Distributed Ledger Technology' (7 February 2017) 4.
1.2.1 Public and permissionless blockchain
A public and permissionless blockchain is a blockchain that anyone can read and on which anyone can propose new transactions. In such a permissionless system, there are no restrictions for participation. Furthermore, there is no central source of power. Transparency is a critical feature in a public blockchain. Anyone can download the ledger and view transaction data. Moreover, anyone can access the blockchain without the prior consent of a gatekeeper.30 Transactions are generally secured by merely requiring new entries to include a proof of work.31
Currently, proof-of-work is the most commonly used consensus protocol for public and permissionless networks, relied upon in, for example, Ethereum and Bitcoin. Through this consensus mechanism, nodes contribute computing power to secure and maintain the system and engage in economic competition. This competition is generated by the requirement for nodes to solve a computational puzzle before they can propose for a block to be added to the chain, as to verify the integrity of transactions.32 In return, the nodes receive a transaction fee.33 Accordingly, public
blockchains are crypto-economically secured.34 Generally, because of the proof-of-work mechanism and its economic rationale, and the lack of a central source of power, implementing unilateral strategies is not possible.
1.2.2 Private and permissioned blockchain
A private and permissioned blockchain is a blockchain that can run on a private network, such as an intranet or a virtual private network. Permissioned ledgers are frequently designed for a specific purpose. Parties experiment with permissioned blockchains to achieve efficiency gains or to execute transactions more easily regarding devoid of trust.35 The completion of new transactions is generally limited to a list of predetermined participants.36
30 Finck (n 19) 15. 31 Schrepel (n 6) 290. 32 Finck (n 19) 20. 33 Schrepel (n 6) 290-291. 34 Civatarese (n 26) 1. 35 Finck (n 19) 15. 36 Schrepel (n 6) 290.
A private blockchain can be divided into two different categories. The first is called a single entity blockchain. In this blockchain, a separate entity will design the consensus protocol and manage the blockchain. Permission to enter the blockchain may be public or only granted to specific participants. The second category concerns the consortium blockchain. In such a blockchain, the consensus protocol is set up by a pre-selected set of nodes. Where a single entity blockchain runs under the guidance of a separate entity, a consortium blockchain operates under the leadership of a group.
In a private blockchain, there is no mining, no proof of work, and consequently no economic rationale. The benefits of a private blockchain come from is valuation and applicability. Uses include serving as a transfer value (for instance, securities, votes, industrial patents),37 serving as a register to verify exchange of products and assets38 and as a smart contract.39
Contrary to the consensus protocol in a public and permissionless blockchain, a single or group of users may be in control of the consensus protocol, making implementing unilateral strategies easier on the private blockchain. Therefore, when analysing anticompetitive conduct prohibited by Article 102 TFEU, only private and permissionless blockchains will be evaluated (3.1).
In addition to the open and permissionless blockchain and the private and permissioned blockchain, there is the semi-private blockchain. In such a blockchain, a single company manages the blockchain, and access to enter the blockchain is granted to any qualified user.40
1.3 Decentralised applications
Entries in the blockchain appear in multiple forms. Besides recordkeeping for cryptocurrencies such as Bitcoin, blockchain technology has a wide variety of uses.
37 Dominique Guegan, 'Public blockchain versus private blockchain' (2017) Documents de travail du centre d'economie de la Sorbonne 5.
38 Ibid.
39 Pierluigi Cuccuru, 'Beyond Bitcoin, an early overview on smart contracts' (2017) 25 International Journal of Law and Information Technology 189, 25.
For instance, blockchain can store a range of records, including payment transactions, retail and pricing history, future pricing, sales records, intangible properties, rights, personal data, licenses, wills, and business.41
One of the decentralised applications of the blockchain is the use of smart contracts. Contrary to what its name suggests, smart contracts are not contracts in a legal sense. However, they can serve as a complement or substitute for legal agreements and, as such, produce legal effects. In essence, smart contracts are pieces of computer nodes capable of verifying, executing, and enforcing a set of instructions.42 The most
commonly used blockchain for smarts contracts is Ethereum.
One specific feature of smart contracts is its automatic execution. For instance, a smart contract installed by a supplier can check the incoming purchase order for conformity with the sales contract in the 'real world' and release an order to ship the products to the customer if the purchase order matches the terms as agreed in the smart contract. So, smart contracts work as an independent agent, which controls certain digital assets, and manages them based on the programmer's instructions. Once launched in the blockchain, they follow their rules until the established goal is reached.43
Smart contracts may exist in different blockchain types but are considered influential in public and permissionless blockchains as there is nothing that can interfere with the software's execution.44 An example of a smart contract is a bank using a smart contract to process automatic payments and issue loans, or an insurance company using a smart contract to process claims.
Oracles are needed to verify whether an execution-relevant event occurred. An oracle can be a single person or a group of persons, which feed the software relevant information. Oracles are the necessary bridge between the blockchain and the off-blockchain world.45
41 Cuccuru (n 39) 185.
42 'The ultimate guide to understanding Smart Contracts' <
https://www.blockchaintechnologies.com/smart-contracts/> accessed 4 January 2020. 43 Cuccuru (n 39) 185.
44 Finck (n 19) 24. 45 Finck (n 19) 25.
1.4 Competition on and off the blockchain
Competition law concerns regarding the blockchain platform and the software operating on it can arise in a twofold manner. First, concerns may arise when considering the blockchain as a platform, and second, concerns may arise through the applications working on the blockchain.46
Regarding the first option, literature distinguishes between three categories of blockchain technology, each increasing in its degree of complexity.47 Blockchain 1.0
refers to the currency applications of the blockchain and using it as a public ledger system for validating transactions, digital currencies, and services like Bitcoin with the ability to disrupt the financial sector. With blockchain 2.0, the focus shifts from currencies to entire markets and economies, storing not just transactions, but also smart contracts and applications. Finally, blockchain 3.0 can be seen as the complete diffusion and adoption of blockchain technology throughout society, potentially disrupting government services and identification systems.48
Not all blockchains allow software (also referred to as 'layer 2') to work on top of the root blockchain ('layer 1'). However, most blockchains do. For instance, Ethereum was designed to allow users to build decentralised applications on the root blockchain. Such applications include smart contracts or agreements between different users to automatically send or release tokens as soon as certain specific conditions are met.49
1.5 Blockchain and its link to European competition law
Blockchain technology makes the substitution of centralised systems possible. The critical factor to take into account when analysing the blockchain from a European
46 Schrepel (n 6) 295.
47 Ioannis Lianos, 'Blockchain Competition – Gaining Competitive Advantage in the Digital Economy: Competition Law Implication' s, in Philipp Hacker, Ioannis Lianos, Georgios Dimitropoulos & Stefan Eich,
Regulating Blockchain: Political and Legal Challenges (forth. OUP, 2019).
48 Ethan Kane, 'Is Blockchain a General Purpose Technology' (2014) SSRN 10
<https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2932585> accessed 4 January 2020. 49 Schrepel (n 6) 295.
competition law perspective is to consider that all data on the blockchain is accessible to all its users, regardless of whether it concerns an open or private blockchain.
As explained before, blockchains are set up in a P2P network. These peers might include (potential) direct competitors in order to increase efficiency. The information held in the blockchain is anonymised or encrypted. Regardless of the anonymised nature of the data, it nevertheless provides an insight into commercial transactions of a respective competitor, making participation on the blockchain problematic as such.50
Another situation that might trigger the application of European competition law in a blockchain is facilitation of collusion. Facilitation of collusion may appear as a result of the public character of the blockchain and the enhanced data visibility it offers. Undertakings may constitute a private blockchain on which they will exchange data regarding their prices, output in real-time, and other sensitive information. Anticompetitive agreements between rivals could be algorithmically controlled, and pseudonymous participants would consequently be harder to trace. This behaviour might constitute a breach of Article 101 TFEU.
Moreover, a firm providing a blockchain could offer terms that would have the effect of excluding smaller or potential competitors. This behaviour will be prohibited pursuant to Article 102 TFEU. When access is needed to participate in a permissioned blockchain, and a refusal to access cannot be objectively justified, this refusal might constitute a breach of Article 102 TFEU.
1.6 Interim Conclusion
A blockchain is an automatically functioning technology, in which consensus of its users is a central feature together with the lack of a central authority regulating it. Blockchains can be used in different ways with different decentralised applications working on it, such as smart contacts. As the data stored on the blockchain is accessible for all its users, the technology may trigger the application of Article 101
50 Morten Nissen & Martin von Haller Grønbæk, 'Blockchain and competition law- issues to be considered' (Bird & Bird News Centre, August 2018) < https://www.twobirds.com/en/news/articles/2018/global/blockchain-technology-and-competition-law-issues-to-be-considered> accessed 4 January 2020.
and 102 TFEU. The following chapter focuses on the implications of the blockchain on the legal concepts of European competition law and how these concepts must be applied to the blockchain.
2 WHAT ARE THE IMPLICATIONS OF THE BLOCKCHAIN FOR THE LEGAL CONCEPTS OF EUROPEAN COMPETITION LAW?
The legal definitions of the relevant market, market power, and undertaking lie at the heart of European competition law and are essential in a competition law analysis. For instance, in order to establish an infringement with Article 102 TFEU, substantial market power of the undertaking concerned needs to be established.51 Before embarking upon an analysis of the substantive provisions of European competition law, it is therefore essential to analyse if the legal definitions inherent to European competition law, can be applied to blockchain, similarly as to markets that have already been known for a longer time.
This chapter will analyse the legal definition of the product market and market power on the blockchain (2.1) and if European competition law captures the blockchain as an 'undertaking' in the legal sense (2.2).
2.1 The blockchain market
In order to establish a market definition, European competition law first addresses the relevant product market. When defining the relevant market, it is essential to analyse the competitive strategy that undertakings are engaging in.52 Consequently, the market power in the relevant markets is determined.53
2.1.1 The 'token effect'
Costs, and other impediments faced by customers in switching to a new supplier, may be a barrier to expansion or to enter. Such costs may arise from network effects.54 A direct network effect arises where the value of a product increases with the number of other customers consuming the same products.55 Direct network effects were
51 European Commission, 'Guidance on the Commission's Enforcement Priorities in Applying [Article 102 TFEU] to Abusive Exclusionary Conduct by Dominant Undertakings' (2009/C 45/02), para 10.
52 The analysis to define the relevant product market entails the relevant product market and the relevant
geographic market. Because of the universal nature of the blockchain, the geographic market will not be discussed in this thesis.
53 Richard Whish & David Bailey, Competition Law (9th edn, Oxford University Press 2018) 26. 54 Whish & Bailey (53) 193.
considered in the Commission's decision in Microsoft56 when the Commission referred to a self-reinforcing dynamic. The more users were on the Microsoft platform, the more software was written for it, and vice versa.57 The Commission again referred to network effects in its Google Shopping58 decision. The more consumers use a search engine like Google, the more attractive the platform becomes to advertisers.
Similar to blockchain, assessing the cryptocurrency market, the more popular a currency, the more useful, and the easier it attracts new users.59 At the so-called 'point
of critical mass', when a certain number of users is reached, the value derived from the product or service is more significant than its price.60 Consequently, the eagerness of users to join increases, and the opportunity to compete on the given market decreases.
The critical question is if the blockchain could be used to compete more efficiently with non-blockchain applications enjoying a powerful network effect.61 Typically, when the critical mass would not be met, this would amount to a low incentive for the users to join the blockchain. On the other hand, the incentive system created on the blockchain makes it attractive for users to join the platform when not many users have joined, as the blockchain 'adds financial utility when application utility is low.'62 Therefore, on the blockchain, when more users use the platform, the lower the incentive for users to join. This counter effect is referred to as the 'token effect.'63 This effect is used when calculating and defining market power on the blockchain (appendix I).
56 Commission decision of 24 March 2004, upheld on appeal case Case T-201/04 Microsoft Corporation v.
Commission, ECLI:EU:T:2007:289.
57 Whish & Bailey (n 53) 193.
58 Case AT.39740 Google Search (Shopping) (2017).
59 Neil Gandal & Hanna Halaburda, 'Can We Predict the Winner in a Market with Network Effects? Competition in Cryptocurrency Market' (2016) 7, 16 Games, 1.
60 Schrepel (n 6) 297. 61 Ibid.
62 Chris Dixon, 'Crypto Tokens: A Breakthrough in Open Network Design' (2016) Medium
<https://medium.com/@cdixon/crypto-tokens-a-breakthrough-in-open-network-design-e600975be2ef> accessed 4 January 2020.
2.1.2 Blockchain as zero-price market
First and foremost, a market definition is based on what product or service is offered on the market. The definition of a relevant product market in the classical sense, as defined by the CJEU in Continental Can64 and United Brands65, is based on product differentiation66, meaning that if goods or services are interchangeable, they belong to the same product market. When assessing the blockchain market, the first difference with a classical market that arises is that blockchain offers services on a so-called 'zero-price market.' 67 In the aforementioned Google Shopping decision, the Commission explained this phenomenon as 'contributing to the monetisation of the service by providing data with each query.'68 Users of the blockchain contribute to the
market power of a blockchain through sharing value in the form of data. Therefore, the blockchain is a 'zero-price market.'
Moreover, the assessment of the blockchain market needs to take into account the specifics and nature of the market. Every blockchain is used for different purposes. The approach to defining the relevant product market will have to determine market power based on the applications, which are regarded as the goods and services on 'layer 2 '(1.4). The type of blockchain69 will then be the focus of the market definition when analysing the functioning of the applications running on the blockchain.70
Following this approach, market power is assessed in comparison with other (non-digital) goods and services. Consequently, competition as such arises between different industries.71 Therefore, a further assessment can be made on whether the service on the blockchain competes with a service outside the blockchain, or services that are solely offered on the blockchain. In the first case, the blockchain can be analysed using a broader market, for example, by comparing the general sales market with online sales. In the second case, competition between mere blockchains needs to
64 Case 6/72 Continental Can v. Commission, ECLI:EU:C:1973:22 para 32. 65 Case 27/76 United Brands v. Commission, ECLI:EU:C:1978:22 para 22.
66 See also: Case T-229/94 Deutsche Bahn AG v. Commission, ECLI:EU:T:1997:155 para 54; Case T-219/99
British Airways v. Commission, ECLI:EU:T:2003:343 para 91 & Case T-321/05 AstraZeneca v. Commission,
ECLI:EU:T:2010:266 para 31.
67 John M. Newman, 'Antitrust in zero-price markets: foundations' (2015) 164:149 University of Pennsylvania Law Review 149, 154.
68 Google Search (Shopping) (n 58) para 158. 69 Meaning blockchain 1.0, 2.0, or potentially 3.0. 70 Schrepel (n 6) 304.
be evaluated.72 The first approach evaluates substitutes and is therefore in line with original case law as Continental Can and United Brands. The Commission followed this approach in its Google Shopping decision, in which the market definition was established on the basis of the totality of interchangeable products and services.73
Usually, in assessing market power, economic concentration indexes serve as a proper tool to assess market power. In a market absent of barriers to entry and which is highly concentrated (such as an open and permissionless blockchain), the company controlling large market shares is likely to possess the largest share of market power. In Google Shopping, the Commission used market shares by volume as a substitute.74 It considered that Google is a service free of charge and that advertisers using search engines look at usage shares. Moreover, the Commission was unable to set the precise value of revenue per search.75
These requirements do not seem to appear relevant for the blockchain as such. Therefore, a case-by-case analysis is required for the blockchain. In order to assess market power, one must consider, inter alia, the users of the blockchain, the number of recorded transactions, the position of the blockchain participants on off blockchain markets and the power of hashing to validate blocks.76 Product differentiation, through for example security, capacity, speed, and anonymity, is critical in this analysis.
2.2 Concept of undertaking
The provisions of Article 101 and 102 TFEU apply to 'undertakings.' A blockchain is a decentralised organisation that is not recognised as a legal entity.77 Accordingly, the main question is: are participants on the blockchain, controllers, or miners, undertakings in a legal sense? This issue is important because of three reasons. First, in order to prohibit anticompetitive conduct under European competition law, the users of a blockchain must be qualified as an undertaking. For instance, Article 101
72 Ibid.
73 Google Search (Shopping) (n 58) para 145. 74 Schrepel (n 6) 305.
75 Google Search (Shopping) (n 58) para 275. 76 Lianos (n 47).
77 John M. Newman, 'Procompetitive Justifications in Antitrust Law' (2019) Indiana Law Journal (forthcoming 2019).
TFEU applies to agreements between undertakings, decisions by associations of undertakings or concerted practices by undertakings. Second, classifying an entity as an undertaking is of essential importance in order to attribute liability for illegal conduct.78 Third, the definition of an undertaking must be clear when assessing market shares in the context of abusing a dominant position.
In the case Höfner and Elser Macrotron GmbH79, the CJEU clarified that the concept of an undertaking 'encompasses every entity engaged in an economic activity regardless of the legal status of the entity and the way it is financed.'80 In
Commission v. Italy,81 it was decided that the concept of economic activity, as mentioned in Höfner, must be interpreted as 'any activity consisting in offering goods and services on a given market.'82 European competition law takes a fundamental approach, focusing more on the concept of 'activity,' rather than 'entity.' The basic test is 'whether the entity is engaged in an activity, which could be carried out on by a private undertaking in order to make profits.'83 Hence, it primarily needs to be
examined what an economic activity entails.
In order to assess whether the participant engages in an economic activity on the blockchain, it is crucial to examine the various forms of anticompetitive conduct. This assessment is necessary in order to establish whether the activity in question is to be qualified as 'economic,' and the entity exercising it as 'undertaking.'
In the case of collusion on the blockchain, the economic activity depends on the activity of the participants as well as the miners and owners of the node (computers that save a local version of the distributed ledger). If these miners voluntarily store and validate transactions and earn a transaction fee, they deem to engage in an economic activity. These owners can be traced relatively effortless, because of the IP address linked to every computer.84
78 Ibid.
79 Case C-41/90 Höfner and Elser v. Macrotron GmbH, ECLI:C:1991:161. 80 Ibid para 21.
81 Case C-118/85 Commission v. Italy, ECLI:EU:C:1987:283. 82 Ibid para 7.
83 Case C-67/96 Albany International BV v. Stichting Bedrijfspensioenfonds Textielindustrie, ECLI:EU:C:1999:430, Opinion of AG Jacobs para 311.
For a European competition law analysis, the absence of a non-profit motive does not mean that the entity does not engage in an economic activity.85 Namely, in the absence of a transaction fee, such as the case in a private and permissionless blockchain, the CJEU ruled in AOK Bundesverband and othersthat if the activity is non-profit, it could still be labelled an economic activity if it has the potential to be exercised by private undertakings to generate a profit.86
Transaction fees will constitute an economic activity, and participants operating on the blockchain, as well as miners, may be labelled as undertakings. Such transaction fees are not characteristics of every blockchain, as the use of such a mechanism depends on the type of blockchain and its consensus protocol. However, in line with AOK Bundesverband and others, if the activities on the blockchain are exercised to
increase efficiency or to generate a profit somehow, they have the potential to be exercised by private undertakings to generate profit. Therefore, this analysis defines the activity on the blockchain as an economic activity in the majority of the cases. Consequently, participants on the blockchain, controllers and miners are to be classified as undertakings.
2.2.1 Blockchain as a single economic-entity
Contrary to the argumentation in the previous paragraphs, the distributed nature of the blockchain could lead to the argumentation of it being a single-economic entity. Miners contribute to the validation of blocks in open and permissionless blockchains. When focussing on the 'entity', rather than 'activity' in establishing an undertaking, miners could argue they are operating as employees, thus as an inherent part of the blockchain. Following that argumentation, miners cannot be qualified as an undertaking. Hence, they fall outside the scope of Article 101 TFEU.87
In recent case law, the CJEU took a cautious approach, focussing merely on the distinction between complete independence and dependence, leaving categorical distinctions such as self-employed out.88 Complete independence of actors on the
85 Whish & Bailey (n 53) 86.
86 Joined cases C-264/01, C-360/01, C-354/10 and C-355/01, AOK Bundesverband and Others, ECLI:EU:C:2004:150 Opinion of AG Jacobs para 27.
87 Lianos (n 47).
blockchain is unlikely, because industrial mining activity is concentrated in a restricted number of mining pools. Following this argumentation, it is ambiguous whether certain actors will fall under the scope of Article 101 TFEU.
However, focussing on what challenges blockchain raises for the legal concepts of European competition law, it will be assumed participants, miners and controllers of the blockchain are to be classified as undertakings in the legal sense.
2.3 Interim conclusion
The market of blockchain is assessed regarding applications running on the
blockchain, as well as the users, miners, controllers, number of recorded transactions, blockchain participants on markets off the blockchain, and the power of hashing to validate the blocks. Product differentiation through security, capacity, speed, and anonymity is the main focus of this analysis. The token effect is a helpful tool in establishing the market power of a blockchain. Moreover, activities on the blockchain can be labelled as an economic activity under most circumstances. Consequently, users of the blockchain are to be qualified as an undertaking and are therefore subject to Article 101 and 102 TFEU.
3 DOES BLOCKCHAIN GIVE RISE TO ANY (NEW) LEGAL ISSUES FOR EUROPEAN COMPETITION LAW?
Today, we witness a shifting focus of the Commission towards competition issues related to digital platforms such as Google, Facebook, and Uber. These platforms do not merely base their competition strategy on a traditional competitive advantage such as lower cost and higher quality of products. Instead, digital platforms find their competitive strategy on the consequences of network effects and data efficiencies. These competitive strategies could lead to anticompetitive behaviour prohibited by European competition law. This chapter analyses if the blockchain gives rise to anticompetitive behaviour and if it raises (new) legal issues that are not covered by Article 102 TFEU (3.1) and Article 101 TFEU (3.2).
3.1 Competition law analysis of the blockchain under Article 102 TFEU
Abuse of dominance, especially in private blockchains, raises acute competition concerns.89 Article 102 TFEU is directed towards the unilateral conduct of dominant firms acting in an abusive manner.90 It is evident that while the blockchain market grows, some blockchains may grow into bottleneck positions.91 If these blockchains are the sole providers of a service or application, Article 102 TFEU may come into play. This section will analyse what legal challenges the blockchain raises for the application of Article 102 TFEU.
3.1.1 Acquiring a dominant position
A prerequisite for analysing abuse of a dominant position is for an undertaking to have such a dominant position, or for more companies to be collectively dominant. In United Brands v. Commission92, the CJEU defined a dominant position as a 'situation when an undertaking enjoys a position of economic strength, preventing competition on the relevant market by the power to behave independently to an appreciable extent of its competitors, customers and ultimately its consumers.'93 The relevant product
89 Michael Ristaniemi and Klaudia Majcher, 'Blockchains in competition law – friend or foe' (Kluwer Competition Law Blog, July 2018) <
http://competitionlawblog.kluwercompetitionlaw.com/2018/07/21/blockchains-competition-law-friend-foe/>accessed 4 January 2020. 90 Whish & Bailey (n 53) 180.
91 Schöning and Tagara (n 10) 5.
92 United Brands Commission (n 65) para 38. 93 Ibid.
market, to determine an undertaking's actual competitors, potential competitors, and its customers, is essential when analysing a dominant position.94 Private and permissioned blockchains are more likely to raise competitive concerns regarding the abuse of a dominant position than public and permissionless blockchains because public blockchains create more transparency between users. Accordingly, this section will focus on private blockchains.
In private blockchains, the entity controlling the consensus protocol, and managing entries on the blockchain, is likely to be dominant on the relevant blockchain. The entity holding the dominant position could consist of more users, as in the single entity blockchain, and of a pre-selected set of nodes, as in a consortium blockchain.
On a general note, it must be borne in mind that it is not the purpose of Article 102 TFEU to prohibit an undertaking to acquire, on its own merits, the dominant position on the market. Nor does it mean that competitors less efficient than the undertaking with the dominant position should remain on the market.95 Nevertheless, a dominant undertaking has a special responsibility 'not to allow its behaviour to impair a genuine, undistorted competition on the internal market.'96 Therefore, a firm in a dominant position on the market has a 'special responsibility not to allow its conduct to impair undistorted competition'97 on the internal market.98 Only when a dominant undertaking abuses its position, it will violate Article 102 TFEU. Such abusive behaviour may consist of discrimination that helps certain parties to an advantage.99 Acquiring a dominant position on the blockchain market will not violate Article 102 TFEU as such. However, when the main objective is to abuse this dominant position on the blockchain market, it will violate Article 102 TFEU.
94 European Commission (51) para 12.
95 C-209/10 Post Danmark, ECLI:EU:C:2012:172 para 21.
96 Ibid para 23; Case C-322/81 Nederlandsche Banden-Industrie-Michelin v. Commission, ECLI:EU:C:1983:313 para 57.
97 Ibid para 57.
98 Whish & Bailey (n 53) 198.
99 Continental Can v. Commission (n 63) para 26; Cases C-395/96 P etc Compagnie Maritime Belge Transports SA
3.1.2 Refusal to deal
Refusal to deal on the blockchain could emerge through refusing a competitor to enter the blockchain, not validating new blocks and denying or not registering new transactions.100 Moreover, a refusal to deal could occur if the dominant undertaking in the blockchain has allowed a competitor to enter the blockchain in an early stage but later excluded that competitor without any legitimate reason.
The Commission101 and CJEU102 use the 'as-efficient competitor test' (''AEC-test'') to assess a refusal to deal. According to this test, the conduct of a dominant undertaking is prima facie abusive if it is excluding a competitor that is at least as efficient as the dominant undertaking. 103 In assessing the 'efficiency' factor, the economic costs and
sales prices are examined.104 In this assessment, the market definition plays a vital role (2.1).
Some companies are only engaged in providing a blockchain, without engaging in any other market. In that case, the AEC-test will not be sufficient to establish refusal to deal, because how can the undertaking that is being denied access on the blockchain be compared with the undertaking denying access on the blockchain? Moreover, if the efficiency of the gatekeeper is set, does this mean a freeway to all undertaking wanting to access the blockchains?
To establish a refusal to deal, dealing with the dominant undertaking must be 'indispensable' for others to compete on the neighbouring market. Second, the refusal must lead to the elimination of all competition on the market.105 Dealing with the competitor is indispensable if i) it is impossible to replicate, and ii) there is no available alternative distribution method.106
100 Ristaniemi & Majcher (n 89). 101 European Commission (n 51) para 23.
102 Case C-62/86 Akzo Chemie v. Commission, ECLI:EU:C:1991:286 para 72; Case C-280/08 P Deutsche Telekom
v. Commission, ECLI:EU:C:2010:603, para 194.
103 Renato Nazzini, The foundations of European Competition Law (1st edn, Oxford University Press 2011) 221. 104 European Commission (n 51) para 25.
105 Case C-7/97 Oscar Bronner Zeitungs-und Zeitshriftenverlag and Mediaprint Anzeigengesellschaft, ECLI:EU:C:1998:658 para 41.
Indispensability in the case of blockchain relates to the fact that in some situations, not being able to enter the blockchain will mean not being able to enter the market; therefore dealing with the dominant player is impossible to replicate. Furthermore, there is no available alternative distribution method, as every blockchain will be used for specific means, being distinguishable through its consensus protocol and establishment purpose. Moreover, access must be refused without any objective justification.107 For an infringement with Article 102 TFEU, indispensability will need to be proven through not only an assessment of the competitive position of the blockchain but also other off-blockchain competition.108
3.2 Competition law analysis of the blockchain under Article 101 TFEU
Article 101 TFEU prohibits agreements, decisions by associations of undertakings, and concerted practices that restrict competition. Article 101 TFEU prohibits certain practices, but not the medium on which these actions are exercised.109 Nevertheless, does the creation of a (public) blockchain in itself constitute an anticompetitive agreement, irrespective of its use? Moreover, what constitutes a concurrence of wills, a condition to establish an agreement?
3.2.1 The agreement creating a blockchain
In Bayer v. Commission110, the CJEU reviewed the case law on the notion of an agreement, and ruled that the concept: 'centres around the existence of a concurrence of wills between at least two parties, the form in which it is manifested being unimportant so long as it constitutes the faithful expression of the parties' intention.'111 The concurrence of wills must refer to 'the implementation of a policy, the pursuit of an objective, or the adaptation of a given line of conduct on the market, irrespective of the manner in which the parties' intention to behave on the market in accordance with the terms of that agreement is expressed.'112 There is no requirement that the
agreement is an agreement in the legal sense.113 The agreement creating a blockchain
107 Schöning and Tagara (n 10) 5. 108 Ibid.
109 Thibault Schrepel, 'Collusion by blockchain and smart contracts' (2019) 33 Harvard Journal of Law and Technology 117, 128.
110 Case T-41/96 Bayer v Commission, ECLI:EU:T:2000:242. 111 Ibid para 69.
112 Ibid para 173.
must be labelled as an agreement in the legal sense because there is a concurrence of wills between the users and facilitators of the blockchain to adapt to a given line of conduct.114
In the Bitcoin blockchain protocol, the number of Bitcoins is limited to twenty-one million.115 This limitation raises the concern whether this amounts to an illegal output restriction, which consequently means agreeing to set up the blockchain is illegal. It certainly is a restriction regarding output. However, this can only be illegal when it qualifies as collusive, not unilateral, conduct. Miners contribute to the operation of the Bitcoin blockchain through validating blocks and receiving economic remuneration for this, thus engaging in an economic activity. In performing their activity, they adhere to the consensus protocol. It is debatable whether their behaviour can be qualified as unilateral, or an agreement or coordination amounting to anticompetitive conduct.116
A closed and permissioned blockchain may be analysed as a cooperative joint venture.117 According to case law, the analysis must examine if the cooperative joint venture restricts competition among undertakings forming the joint venture and competition between the parent companies.118 In a closed and permissioned blockchain, competitors could potentially share sensitive business data, and participants could be harder to trace due to their pseudonymous identity.119 In that case, the agreement to set up a blockchain will constitute an agreement and will fall under the scope of Article 101 TFEU.
3.2.2 Blockchain as a tool for sharing sensitive data
A vital feature of the blockchain is that it turns private information into public information, generally for all users of the blockchain. After the blockchain is created, as discussed in the prior section, a situation that could occur is that information is
114 Schrepel (n 109) 129.
115 Peder Østbye, 'The case for a 21 Million Bitcoin conspiracy' (2018)
<https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3136044> accessed 4 January 2020. 116 Lianos (n 47).
117 Philip Hacker, Ioannis Lianos, Gergios Dimitropoulos and Stefan Eich, Regulating Blockchain: Techno-Social
and Legal Challenges (1st edn, Oxford University Press) 382.
118 Case T-112/99 Métropole Télévision (M6) and Others v. Commission, ECLI:EU:T:2001:215.
119 Martin Coleman, 'Blockchain: competition issues in nascent markets' (Norton Rose Fulbright Competition
shared on a (public) blockchain but is not equally accessible to competitors. In that regard, the Commission120 as well as the CJEU121 consider that Article 101 TFEU applies the same way to any agreement, and therefore also for collusive behaviour on the blockchain.
Sharing of information regarding future prices may constitute a restriction of competition under Article 101 TFEU.122 First must be analysed what kind of information is shared on the blockchain. The Guidelines on Horizontal Cooperation Agreements123 ("Guidelines") explain that it is likely that the exchange of strategic
data between competitors will infringe Article 101 TFEU because it will intervene with a parties' decision-making independence.124 The Guidelines identify that information on prices and quantities is the most strategic, followed by information concerning costs and demand.125 Moreover, the strategic nature of information depends on the aggregation126 and age127, as well as on the nature and characteristics of the market and frequency of exchange.128
On the blockchain, any data may be shared. The Guidelines prescribe that the sharing of strategic data129 relates to data that reduces strategic uncertainty in the market.130 It must be impossible for an undertaking not to take the information into account when determining its competitive strategy. As a result, sharing of information on a blockchain that is strategic may constitute a cartel. Until now, case law has not identified the sharing of current prices to constitute a restriction of competition under Article 101 TFEU.131
120 European Commission, 'Guidelines on the applicability of Article 101 of the Treaty on the Functioning of the European Union to horizontal co-operation agreements Text with EEA relevance' (2011/C 11/01) para 92. 121 Case T-587 Fresh Monte v. Commission, ECLI:EU:T:2013:129 para 323.
122 Case T-25/95 Cimenteries CBR v Commission, ECLI:EU:T:2000:77 para 49. 123 European Commission (n 120). 124 Ibid para 86. 125 Ibid. 126 Ibid para 89. 127 Ibid para 90. 128 Ibid para 90.
129 According to paragraph 86 Guidelines, this includes information relating to prices, customer lists, production costs, quantities, turnovers, sales, qualities, marketing plans, risks, investments, technologies, research and development programmes, and their results.
130 European Commission (n 120) para 60-63. 131 Schrepel (n 109) 132.
The Bananas-case 132 considered a concerted practice relating to pre-pricing communications for setting quotation prices for bananas to be sold on the market in Northern Europe.133 The Court held that 'exchange of information between competitors is to be incompatible with the competition rules if it reduces or removes the degree of uncertainty as to the operation of the market in question, with the result that competition between undertakings is restricted.'134 Also, the Court noted that a concerted practice might be a restriction by object, even if there is no direct connection between the practice and consumer prices.135
The facts of Bananas show ample evidence of collusion, as the undertakings, as part of the bilateral pre-pricing communications, discussed their quotation prices and specific price trends.136 This information was relevant for internal pricing, and employees of the concerned undertakings that were involved in the pre-pricing communications participated in the internal pricing meetings.137 The CJEU found that 'the pre-pricing communications had the object of creating conditions of competition that do not respond to the normal conditions on the market.'138 Exchange of information regarding price may, therefore, restrict competition under Article 101 TFEU if there is evidence of collusion and if a causal link between the practices is established.
Although Article 101 TFEU prohibits collusion that distorts competition, 'it does not deprive economic operators of the right to adapt themselves intelligently to the existing and anticipated conduct of their competitors.'139 Decoding data (of competitors) on the blockchain is therefore not to be viewed as communication or as a facilitating practice, but rather as a single observation.140
132 Case C-286/13 P Dole Food & Dole Fresh Fruit Europe v. Commission, ECLI:EU:C:2015:184. 133 Kristan Hugmark and My Becher, 'Exchange of Information, Between Competitors, on Price-Related Parameters' (2015) 6 Journal of European Competition Law & Practice 652.
134 Dole Food & Dole Fresh Fruit v. Commission (n 132) para 121. 135 Ibid para 123.
136 Ibid para 129. 137 Ibid para 130. 138 Ibid para 130.
139 Joined Cases C-89/95, C-104/84, C-114/85, C-116/85, C-117/85 & C-125/85 to C-129/85 Ahlström Osakeyhtiö
and Others, ECLI:EU:C:1993:120 para 71.
140 Ariel Ezrachi & Maurice E. Stucke, 'Sustainable and Unchallenged Algorithmic Tacit Collusion' (2019) 16 Oxford Legal Studies Paper 29.
In the Eturas141-judgement, which will be elaborated upon in more detail in a later stage, the CJEU held had (passive) participation in anticompetitive meetings is an indication of collusion if the undertaking does not publicly distance itself or does not report the event to the competent authority.142 However, does joining the blockchain amount to consenting to share sensitive information on the blockchain? Is a participant of the blockchain required to leave the blockchain as soon as it finds out sensitive information is shared?
Moreover, in its analysis, the CJEU evaluated the market on which the competitors operated.143 Determining the ordinary conditions of the market in the context of blockchain requires an in-depth analysis of the intrinsic characteristics of the market by the Commission and the CJEU and the possibilities it offers to unmask its pseudonymous users.144
3.2.3 Smart contracts and liability issues
Next to the sharing of strategic data, collusive behaviour on the blockchain may also occur through the use of smart contracts, in particular, if this were paired with algorithms that are adjusted automatically when the conditions of a smart contract are satisfied.
3.2.3.1 Smart contracts easing collusion
Smart contracts (1.3) are pieces of code stored on the blockchain that are automatically performed once deployed, enhancing the trust and security of the blockchain network. Members of a cartel may be connected through a smart contract, which could condition the automatic release of 'guarantee' if conditions concerning the deviation from the cartelised price are identified by one of the members of the cartel. This guarantee would be paid in cryptocurrencies by each of the members of a cartel and kept in an 'escrow account'145 at one of the 'off-blockchain' digital wallets. Algorithms relying on off-blockchain data harvested by oracles (person or group of
141 Case C-74/14 Eturas UAB and Others v. Lietuvos Respublikos konkurencijos taryba, ECLI:EU:C:2016:42. 142 Ibid para 28.
143 Ibid para 117. 144 Lianos (n 47).
145 An escrow account is an account where funds are stored when two or more parties complete a transaction. A trusted party will secure the funds in a trust account.
person that feeds the software relevant information, and verify whether an execution-relevant event occurred) would then ensure the execution of the smart contract.146
Pricing algorithms play an essential role in detecting deviant behaviour,147 resulting in high price transparency and high frequency trading that allows participants on the blockchain to react fast and aggressively.148 Because of the knowledge of actual data in the blockchain, sellers can detect deviations quicker in a collusive equilibrium. Hence, the ability to punish as a collective group is better coordinated through means as smart contracts.149 A smart contract could then correct deviance of the equilibrium
by putting in place targeted and automated punishments. For example, the participation fee in a blockchain may be regulated following the behaviour of each colluder, making it more expensive for individual members to deviate from the agreement. Ideally, the punishment will be explicitly directed to the deviated member.150
As a result, the use of algorithms in the blockchain, such as smart contracts, makes collusive strategies on the blockchain stable. It follows that competition law issues arise if smart contracts are used for the execution of collusion through detecting deviant behaviour, and punishing users who do not act according to the behaviour that the cartel prescribes. Such behaviour will constitute an infringement by object of Article 101 TFEU.
3.2.3.2 Ambiguity in regards to liability
Consequently, if collusive behaviour is facilitated on the blockchain, the next mandatory step is to determine who is to be held liable. The judgement of AC-Treuhand AG v Commission151 is relevant in this analysis because it may apply to any type of intermediary, including online platforms such as a blockchain. In AC-Treuhand, a consultancy firm acting as intermediary for the anticompetitive conduct was held liable. This firm was not active on the relevant market or related markets,
146 Lianos (n 47).
147 Ezrachi & Stucke (n 140) 8. 148 OECD (n 14).
149 Lin William Cong, Zhiguo He and Jingtao Zheng, 'Blockchain disruption and smart contracts' (2017) SSRN 5 <https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2985764> accessed 4 January 2020.
150 Schrepel (n 109) 148.
