BIG DATA AS AN ESSENTIAL FACILITY: JUSTIFICATIONS FOR REFUSAL OF ACCESS
Emmi Tunturi
e-mail: emmi.tunturi@hotmail.com student nr.: 12203793
International and European Law: European Competition Law and Regulation
Supervisor: Jan Broulík
ABSTRACT
This thesis explores the interplay and possible issues between the essential facilities doctrine, found under the EU competition law, and the objective justifications for the refusal to share the essential facility, in a case when the facility is big data. The purpose is to analyse the current situation of big data as an essential facility. Moreover, the research questions concern whether dominant undertakings can be forced to grant access to their data sets for their competitors, and whether the undertaking imposed the duty to deal can make an objective justification for refuse to share its data set, based on either economic or data protection justifications. The research methods used for this thesis are descriptive and explanatory methodology.
According to EU competition law, there must be an abuse of a dominant position for the prohibitions of EU competition law article 102 TFEU to apply. This thesis will seek to define whether big data can qualify an essential facility and therefore, become in certain circumstances a tool for abuse. The answer to the above question depends on the nature and uses of the particular data set at hand, besides the market structure and possible entry barriers present in that market. Afterwards, it will be defined when an objective justification is present, either economical or data protection justification. In practice, any dominant undertaking which is accused of abusive behaviour may try to invoke an objective justification to justify its otherwise abusive conduct. If an objective justification is successfully invoked, the Art. 102 TFEU prohibition is not applicable.
This thesis ends by concluding that big data can, under certain circumstances, qualify as an essential facility, found under the EU competition law. However, this is a case-specific issue and cannot be assumed automatically. What comes to the objective justifications, there are common conditions to be fulfilled before there can be any objective justifications present. Besides the common conditions, every different type of defence gives additional conditions to be met. For example, in data protection defence when the data concerned includes personal data of natural persons, the provisions of the General Data Protection Regulation should be complied with.
TABLE OF CONTENTS
ABBREVIATIONS ... 3
INTRODUCTION ... 4
1. BIG DATA AND THE ESSENTIAL FACILITIES DOCTRINE ... 6
1.1. Big Data ... 6
1.1.1. Benefits and Issues Related to Big Data ... 7
1.1.2. Big Data and Competition Law ... 8
1.2. Essential Facilities Doctrine ... 9
1.2.1. Requirements for Essential Facility ... 10
1.3. Application of Essential Facilities Doctrine to Big Data ... 12
1.3.1. The Requirements Which Big Data has to Fulfil ... 13
1.3.2. Application in Practice ... 15
2. ECONOMIC JUSTIFICATIONS FOR REFUSAL OF ACCESS ... 17
2.1. An Objective Justification ... 17
2.1.1. Legal Framework ... 18
2.1.2. An Objective Justification in Practice ... 19
2.2. Efficiency Defence ... 20
2.2.1. Benefits for Consumers and Specific Conduct ... 21
2.2.2. Necessity - No Less Anticompetitive Alternatives Available ... 21
2.2.3. Outweighing Competition and Consumer Harm ... 21
2.2.4. Maintaining Effective Competition ... 22
2.3. Public Interest Defence ... 22
2.4. Own Commercial Interest Defence ... 23
3. DATA PROTECTION JUSTIFICATION FOR REFUSAL OF ACCESS ... 25
3.1. Big Data, Competition Law and Data Protection ... 25
3.1.1. Data Protection ... 26
3.1.2. Big Data and Data Protection ... 27
3.2. Data Protection and Forced Sharing of Data ... 27
3.2.1. Forced Sharing of Personal Data ... 28
3.2.2. To Avoid Additional Requirements Imposed by Data Protection Provisions ... 29
3.3. Data Protection Defence ... 30
3.3.2. Problems of Data Protection Defence ... 32 CONCLUSION ... 34 BIBLIOGRAPHY ... 36 I. Case Law ... 36 II. Legislation ... 37 III. Literature ... 38
ABBREVIATIONS
AG Advocate General
ECHR Charter of Fundamental Rights of the European Union ENISA The European Union Agency for Network Security
EU The European Union
FRAND Fair, Reasonable and Non-Discriminatory
GC The General Court
GDPR General Data Protection Regulation (Regulation 2016/679)
IPRs Intellectual Property Rights
NCAs National Competition Authorities
OECD The Organisation for Economic Cooperation and Development
OJ Official Journal
TEU Treaty on the European Union
TFEU Treaty on the Functioning of the European Union
INTRODUCTION
In recent years, the importance of data in the economy has become increasingly apparent.1 More powerful computers, the growth of networks, and advances such as machine learning have led to an explosion in the use of data. Furthermore, in today’s digitalised world, every single click on a website, every purchase online, every song listened to, and every uploaded picture can and often is being turned into data and analysed.2 Everything done online is increasingly leaving a digital trace (data), which can be used and analysed.3 This analysed data is creating competitive strengths for businesses, especially to online businesses whose competitive strength is most often determined by the variety, amount and quality of the data they possess.4 The growth and success of an undertaking’s business are becoming more and more dependent on its abilities to process, obtain and develop data. The development in the technology, more precisely in information technology, and digital data processing has increased the need to understand and analyse the true impact of big data.5
This thesis analyses the relationship between EU competition law and big data. The object of this thesis is to evaluate the issues raised by the accumulation of big data to undertakings, and the possible implications from the point of view of competition law. A particular emphasis is on the doctrine of essential facilities found in the competition law, under which an undertaking’s facility can be made subject to forced access by competitors. This thesis intends to explore the possibility of big data to qualify as an essential facility and on the other side the possibility of a dominant undertaking to invoke an objective justification for refusal to share its facility (big data) to its competitors. What makes this thesis unique and current is that it tries to make clarifications and deepen the knowledge on the ongoing debates among academics, and contradictions in the existing literature, whether big data could be considered as an essential facility and be subject to forced access.6 The disagreement is based on the extraordinary character of data, and whether the data can satisfy the conditions for essential facilities doctrine. This thesis argues that when a facility is able to fulfil all the conditions for the essential facility, it should be treated as such. Moreover, this research and thesis hypothesis
1 Thomas Hoeren, ‘Big data and the ownership in data: recent developments in Europe’ (Vol. 36(12), European
Intellectual Property Review 2014)
2 Commission, ‘Guidance Towards a Data-Driven Economy’ COM (2014) 442
3 Thomas Hoeren and Barbara Kolany-Raiser, Big Data in Context: Legal, Social and Technological Insights (Springer
International Publishing AG 2018) 13-14
4 Inge Graef, ‘Data as Essential Facility: Competition and Innovation on Online Platforms’ (DPhil thesis, KU Leuven:
Faculty of Law 2016) 7
5 Maurice Stucke and Ariel Ezrachi, ’The curious case of competition and quality’ (2015) Journal of Antitrust
Enforcement <https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2494656> accessed 20 May 2019
6 Giuseppe Colangelo and Mariteresa Maggiolino, ’Big data as misleading facilities’ (2017) Vol. 13 European
that data (more precisely big data) can under certain circumstances amount an essential facility, without which other undertakings are not capable of surviving on the specific market.
The research questions which this thesis aims to answer concern what is the current situation of big data as an essential facility under EU competition law. This border question includes sub-questions such as can the accumulation of big data to a dominant undertaking result in such data assets being held as an essential facility under the EU competition law, whether a dominant undertaking could be forced to grant access to its data under the essential facilities doctrine, and whether there are any legally acceptable justifications for the concerned undertaking to refuse to give the access. In addition to the traditional economical justifications, in this thesis, it will be analysed whether as a result of forced access, the undertaking could be violating the applicable data protection regulation and whether that could be a ground for justification for refusal of access. To establish well-based research and conclusion, the research method used will be a descriptive and explanatory methodology.7
The main area of law of the research and this thesis is competition law. Nevertheless, others such as data protection laws will also be discussed. To provide answer to the main issues of this thesis, whether the competition law doctrine of essential facilities, where dominant undertakings can be forced to allow competitors access to their facility, is compatible with big data as facility and what can the dominant undertaking issue as its defence not to share its facility, a comprehensive analysis is conducted of the legal concepts and rules surrounding the essential facilities doctrine and big data. This thesis is structured to explain how big data can qualify as an essential facility and how an objective justification can justify undertaking refusal to comply with the obligation to share its facility. The first chapter will analyse the concept of big data and essential facilities doctrine and provide an overview of how big data can qualify as an essential facility. The second and third chapter will analyse the possible objective justifications invoked by the refusing undertaking. Moreover, the evaluation is divided between justifications based on economical grounds and data protection justification.
7 Ari Hirvonen, ’Mitkä metodit? Opas oikeustieteen metodologiaan’ (In English: What methods? Guide to the
methodology of jurisprudence, Yleisen oikeustieteen julkaisuja 2011/17) 36-49; Robert Cryer and others, Research Methodologies in EU and International Law (Hart Publishing 2011)
1. BIG DATA AND THE ESSENTIAL FACILITIES DOCTRINE
Big data – extremely enormous data sets that might be analysed computationally to reveal patterns, associations, and trends relating to human behaviour and their interactions. The first part of this thesis focuses on the concept of big data as well as to the doctrine of essential facilities. Furthermore, the objective is to analyse the issues raised by the accumulation of big data to undertakings and the possible implication from the point of view of EU competition law. In previous literature, it has been recognised that big data can help to improve an undertaking’s services or products. Nevertheless, at the same time, it is possible for it to create problems and competitive concerns in the relevant market.8 In this thesis, a particular emphasis is on the doctrine of essential facilities found in the competition law, which establishes that a dominant undertaking’s facility can be made subject to forced access by competitors.9 This thesis begins by defining and examining the concept of big data and essential facility. Afterwards, an analysis will be conducted whether it would be possible for big data to qualify as an essential facility, found in the competition law, in today’s technologically driven world.
1.1. Big Data
The task of laying down the definition for the term ‘big data’ is not a simple one, as the meaning of the term is still evolving, and there is no consensus for the legal definition of the term. In its simplest, big data has been understood as encompassing a massive volume of structured and unstructured data which is so vast that it is impossible, or extremely hard, to process and manage by traditional database or software techniques.10 Gartner, an information technology research and advisory company, has provided a technical and widely accepted definition for big data. By this definition, big data is
“high-volume, high-velocity and high-variety information assets that demand cost-effective, innovative forms of information processing for enhanced insight and decision making”.11 In other words, the concept of big data refers to large volumes of a variety of data which can be collected at high velocity and is then processed to produce unique data sets which can have significant commercial value. Big data refers to the data which is collected and analysed and later the ability to use this analysed data.
8 Marcelo Corrales, Mark Fenwick and Nikolaus Forgo, New Technology, Big Data and the Law (Springer 2017) 18-20 9 Allison Jones and Brenda Sufrin, EU Competition Law: Text, Cases, and Materials (6th edn, Oxford University Press
2016) 496-497
10 Jens Heidrich, Adam Trendowicz and Christof Ebert, Exploiting Big Data’s Benefits (IEEE Software 2016) 111-114 11Gartner IT Glossary: ‘Big Data’, (Gartner IT Glossary) <http://www.gartner.com/it-glossary/big-data/> accessed 06
Furthermore, literature often refers to the specific characteristics of big data as three, or sometimes four or more ‘Vs’. Namely: volume (the amount of data available), velocity (the rate at which new data are generated and analysed), variety (the differences in types of data used and the increasing complexity of data analysis), and variability (the different interpretations of data analysis and the extent to which data is consolidated, cleaned and consistent).12 In the context of market competition, big data can be understood as the ability to collect, process and analyse huge volumes and a wide variety of data in a sufficient period of time to obtain information giving the owner the ability to establish business strategies and, in some situations possibly comprising an essential facility, market power and even a dominant position to unilaterally interfere in the functioning of the market.13
1.1.1. Benefits and Issues Related to Big Data
Big data is widely considered to be able to provide significant benefits to the modern economy, and to both businesses and consumers who operate in that specific economy. It is a valuable tool, and often undertakings which are failing to recognise and capture its benefits are risking to fall behind.14 Also, many social networks providers, such as search engines and e-commerce platforms, need and are dependent on the acquisition of big data to be able to work effectively and efficiently in the market.15 The reason why big data is beneficial relates to its unique character. It can add value to almost any type of business, and it is seen as a non-rivalrous good, which can be used simultaneously by more than one undertaking. The non-rivalrous character of big data means, at a technical level, that data is not depleted through use.16 Most of the goods and products at the market are rival, for example, if natural person purchases three apples or an hour of lawyer’s time, some resources with a positive opportunity cost are used up. In contrast, existing big data can be used by any number of undertakings simultaneously, without being diminished. The value of big data does not reduce even the data is used or processed.
Big data can be considered as a vital element of an undertaking’s business strategy to compete more effectively in the market. It might increase operative efficiency, mitigate information asymmetries and contribute to more efficient sales transactions. On the one hand, big data may in many ways be
12 Maurice Stucke and Allen Grunes, Big Data and Competition Policy (Oxford University Press 2016) 15-28 13 ibid
14 Rodney Heisterberg and Alakh Verma, Creating Business Agility: How Convergence of Cloud, Social, Mobile, Video,
and Big Data Enables Competitive Advantage (John Wiley & Sons, Inc. 2014) 230
15 Inge Graef (n 4) 127
16 Charles Jones and Christopher Tonetti, ‘Nonrivalry and the Economics of Data’ (Version 0.6, Stanford GSB and
conducive to greater competition in the relevant market.17 For example, by making information more accessible, it can lower search costs, diminish barriers to prospective entrants, and create new ways for entry and expansion. On the other hand, big data is presenting new challenges to competitive markets. It could drive network effects, raise the barriers to entry, and lead towards platform and market foreclosure by denying potential competitor undertakings access to the data.18 Besides these problems recognised under competition law, there are other problems related to big data, such as problems considering the privacy of the consumers and users of online platforms and services.
It is essential to keep in mind that the opportunities which big data affords to undertakings to develop a competitive edge are the very same features through which competition concerns may arise. For example, the use of big data is problematic as it grows continuously, and undertakings often fail to capture the opportunities and catch actionable data. Moreover, the massive explosion in the data available exceeds the amount of data which can be stored and computed.19 Therefore, it can be stated that it is not problematic to find useful data; the management of the big data creates the issues. Other issues are the lack of technology and tools needed for the processing of big data and data security. Most of the undertakings wishing to process and use the big data catered are not able to do so as they lack the knowledge, technologies or the tools to do so.20 Finally, the issue of data security relates to the fact that the big data that comes into undertakings are made available from a wide range of sources, some of which cannot be trusted to be secure.21
1.1.2. Big Data and Competition Law
Competition law aims towards the protection of the process of competition. The primary source of EU competition law is TFEU,22 which includes all the main articles governing competition law in the EU. For this thesis, the main focus will be on Art. 102 TFEU, which concerns abuses by dominant undertakings.23 For an undertaking to violate this article, it must possess a dominant position in the market, the undertaking has to be able to abuse its market position, and this abuse must affect trade between the Member States. Possessing a dominant position is not abusive behaviour; the abuse of
17 Paul Lugard and Lee Roach, ‘The Era of “Big Data” and EU/U.S. Divergence for Refusal to Deal’ (Antitrust Vol. 31
American Bar Association 2017)
18 Maurice Stucke and Allen Grunes (n 12) 157-165
19 Grünter Fahrnberger and others, Distributing Computing and Internet Technology (Springer 2019) 295-300 20 ibid
21 ibid
22 Consolidated version of the Treaty on the Functioning of the European Union [2012] OJ C326 (hereinafter TFEU) 23 ibid, Art. 102 TFEU
this position is. Case law has added that in many instances, the dominant undertaking has a special responsibility not to abuse its dominant position, not to engage in conduct that can impair competition.24 When considering big data from a competition law perspective, it should be evaluated whether the undertaking, possessing the big data, is dominant in the specific market. In case it is dominant, an analysis should be conducted whether this undertaking is abusing its market position, for example, by refusing to deal or dealing in a discriminative manner. In other words, evaluation should be conducted whether the big data at hand can become a tool for abuse.
At its simplest, big data is just a product and the same competition law analysis can be applied to it as is applied to any other product. The problem is that data is not finite in the same way as most products. Moreover, it is often replicable. Nevertheless, large data sets can be challenging to replicate, first-mover advantage can be significant, and data can reinforce barriers to entry. Also, the value of big data is linked to the undertaking’s ability to process it.25 Pure data may be worthless if it cannot be analysed. The accumulation of big data is not in itself problematic under EU competition law. The key concerns related to relations of big data and competition law includes problems such as: could owning a significant data set create a dominant undertaking, can a large data set reinforce dominance in another market, or could dominant undertaking be able to accumulate an extensive data set use it to exploit other markets?26 The most straightforward answer to these questions is that it depends on the type of data in question, the market in which the undertaking operates, and the undertaking’s ability to process the data.
1.2. Essential Facilities Doctrine
When an undertaking is condemned to have a dominant position in the market where it is operating, a refusal to give access of its facility to other undertakings on non-discriminatory terms may constitute an abuse of the dominant position.27 The problem of dominant undertaking refusing to give access was solved by introducing a new competition law doctrine, known as the essential facilities doctrine. Under this doctrine, a dominant undertaking’s facility can be made subject to forced access by competitors. Moreover, this doctrine may create a legal obligation for a dominant undertaking to share its facility with its competitors. The doctrine specifies when the undertaking possessing the
24 Case C-322/81 NV Nederlandsche Banden Industrie Michelin v The Commission [1983] EU:C:1983:313, para. 57 25 Maurice Stucke and Allen Grunes (n 12) 36-41
26 ibid
(essential) facility must give access to that concerned facility, at a reasonable price (it is not demanded without adequate compensation).28 The essential facilities doctrine can be seen problematic as it goes against the general principle (freedom of contract)29 recognising that everyone should have the right to choose their trading partners and dispose of their property freely.30 Undertakings may not counter the allegations of anticompetitive activity by arguing on the freedom to conduct business and the right to property because these principles are also subject to public-interest restrictions.31
Nowadays, the essential facilities doctrine is considered as a mandatory access remedy under EU competition law.32 It is used to deal with exceptional cases where a dominant undertaking takes advantages of its monopoly power by refusing to deal and therefore denying access to the essential product or service (this conduct could be able to foreclose the competition in the relevant market). In essence, competition authorities can impose an obligation on a dominant undertaking to grant access to its facility to make sure that markets are not foreclosed.33 To do so, enforcers need to conduct an analysis and comply with a legal test which courts have crafted through case law. Nevertheless, it has been widely accepted that this possible remedy should only be imposed in limited circumstances.34
1.2.1. Requirements for Essential Facility
An undertaking which has a dominant position and possesses essential facility and does not grant access to this facility without objective justification or grants access to competitors only on terms less favourable than those which it gives to its subsidiaries infringes competition law.35 In a landmark case, Commercial Solvents the Court established a broad rule whereby it could be inferred that an undertaking holding a dominant position had a duty to deal with another undertaking operating in a downstream market.36 Afterwards, the EU case law evolved and broadened the possible application
28 OECD, ‘The Essential Facilities Concept’ (Organisation for Economic Co-Operation and Development 1996) 91 29 Nestor Duch-Brown, Bert Martens and Frank Mueller-Langer, ‘The economics of ownership, access and trade in
digital data’ (JRC Technical Reports, JRC Digital Economy Working Paper 2017)
30 Case C-7/97 Oscar Bronner GmbH & Co. KG v Me- diaprint Zeitungs- und Zeitschriftenverlag GmbH & Co. KG and
Others [1998] EU:C:1998:264, Opinion of AG Jacobs, para. 56
31 Case C-308/04 P SGL Carbon v Commission [2006] EU:C:2006:433, para. 108
32 Jere Lehtioksa, ‘Big Data as an Essential Facility: The Possible Implications for Data Privacy’ (Master’s thesis,
University of Helsinki 2018) 44-47
33 Richard Wish and David Bailey (n 27) 715-725
34 European Commission, ‘Fines for breaking EU Competition Law’ (2011)
<http://ec.europa.eu/competition/cartels/overview/factsheet_fines_en.pdf> accessed 19 May 2019
35 Case COMP IV/34.689 Sea Containers v Stena Sealink [1994] L 15/8, para. 66
of essential facilities doctrine. Newer cases such as Magill,37 Oscar Bronner,38 IMS39 and Microsoft40 are the primary cases through which courts have crafted the so-called essential facilities analyse to determine whether a competition law breach exists when an undertaking refuses to deal.
In these cases, the Court clarified the requirements for the essential facilities doctrine’s application.41 In order to impose a duty to deal, under the essential facilities doctrine, the following circumstances need to exist. Firstly, the service or product to which access is being denied must be indispensable to be able to compete in the downstream market.42 Secondly, the refusal would lead to the elimination of effective competition in the downstream market. Thirdly, the refusal would lead to the prevention of the introduction of a new product. Fourthly, there is no objective justification for the refusal.43 Enforcers have to carry out analysis before imposing any remedy based on the essential facilities doctrine. In other words, there is no presumption that the dominant undertaking has a duty to deal with competitors.
To the requirements established by the case-law of the Court, the Commission has stated that it will consider these requirements when determining whether an obligation to deal should be imposed on an undertaking or not. However, it adds that it will also look at whether the specific refusal is likely to lead to consumer harm. Commission’s argumentation that consumer welfare should be taken into account is based on the underlying objectives of competition law, the protection of consumer welfare.44 Besides, to the requirements of essential facility doctrine, it is important to note that the duty to deal is not automatically imposed. Even in the case where the conditions for the essential facilities doctrine are met, the Commission and courts are looking at other possible factors and evaluating whether there needs to be forced access or not.45 In other words, a case-by-case analysis is required before imposing a duty to deal on undertaking possessing the essential facility.
37 Joined cases C-241/91P and C-242/91 P Radio Telefis Eireann (RTE) and others v Commission [1995] EU:C:1995:98 38 Case C-7/97 Oscar Bronner (n 30)
39 Case T-184/01 R IMS Health Inc. v Commission [2006] EU:T:2005:95 40 Case T-201/04 Microsoft v Commission [2007] EU:T:289
41 For example, Case C-7/97 Oscar Bronner, Opinion of AG Jacobs, [1998] ECR I-07794, para 35 42 Case C-7/97 Oscar Bronner (n 30) para 41
43 Robert O’Donoghue and Jorge Padilla, The Law and Economics of Article 82 EC (Hart Publishing 2006) 421-434 44 Commission, ‘Guidance on the Commission’s enforcement priorities in applying Article 82 of the EC Treaty to
abusive exclusionary conduct by dominant undertakings’ (Communication)’ COM (2009) OJ C45/7, para. 81
1.3. Application of Essential Facilities Doctrine to Big Data
The previous case-law of the Court illustrates that the essential facilities doctrine has been successfully applied in cases where the essential facility is physical infrastructure, for example, ports or gas pipelines.46 Later it was also established that the essential facilities doctrine could be applied towards facilities which have been protected by IPRs. For now, it will shortly be examined the application of essential facilities doctrine towards IP rights as the application of the doctrine towards them opened the door for the wider application of the essential facilities doctrine (even towards products which are not physical infrastructures). Generally, IPRs are characterised by immateriality. They are aiming to protect the creative intellectual human production in the sectors of science, industry, art and literature.47 Furthermore, they do not protect the specific physical property. They protect the creativity or innovation that is enshrined in a specific product. However, there is still one common feature with the physical facilities and IPRs, namely the power of the owner to exclude everybody else from their use and commercial exploitation.48
The application of the essential facilities doctrine towards IPRs, in the Court’s case-law, shows that even the product/facility at hand is not physical infrastructure the doctrine can be applied in circumstances where the four cumulative requirements for it are met. Moreover, the IPRs and big data have some common aspects such as the fact that intellectual property does not diminish in use (as does not big data). The owner of the specific intellectual property can continue to exploit it also after licensing it for third parties. In other words, both intellectual properties and big data are non-rivalrous.49 Therefore, it seems improbable that the Court would in the future deny the application of the doctrine towards data, as the Court has already applied the doctrine towards IPRs. From the foregoing, it can be stated that in case big data can fulfil the requirements for an essential facility, it should be treated as such, as was the case with IPRs.50
More precisely, whether essential facilities doctrine could be applied to big data depends on the characteristics of the data at hand. One argument to support the application of the doctrine to big data is that, if the big data would be genuinely ubiquitous, low cost and widely available, then undertakings
46 Allison Jones and Brenda Sufrin (n 9) 500
47 Annette Kur and Thomas Dreier, European Intellectual Property Law: Text, Cases and Materials (Edward Elgar
Publishing 2013) 2-3
48 ibid
49 Tuomas Mylly, Intellectual Property and European Economic Constitutional Law: The Trouble of Private and
Informational Law (IPR University Center 4 2009) 515-527
would not waste money, resources and time, for example, by offering “free” services in order to gain the personal data of their users.51 At the national level, it has been noted that the collection and how data are used can be considered a factor able to raise entry barriers and be a source of market power. When considering big data, the barriers might be technological, legal, behavioural, or a combination of the previously mentioned possible barriers.52 As big data can raise barriers, and it is offering more value to the business conduct of the possessing undertaking, it seems unreasonable to deny the application of essential facilities doctrine towards big data.
In this part of this thesis, it is considered, in detail, whether big data can fulfil the conditions for the essential facility, and how the application of this doctrine to big data would work in practice. The application of the doctrine to big data is not a simple task and is problematic to demonstrate as data creates new kinds of challenges to this doctrine and its application. For example, certain entry barriers are unique to big data, such as the storage of large volumes of data. The specific character of the data necessary for a particular market in which the data serve as an input can influence the intensity and type of entry barriers.53 It might be required to take different types of analysis into account when determining the possibility of big data to qualify as an essential facility.54
1.3.1. The Requirements Which Big Data has to Fulfil
For big data to qualify as an essential facility, it has to fulfil the requirements for essential facilities doctrine. These requirements, when drafted, were intended to be applied towards physical infrastructures, at that time it did not become apparent to anybody that in the future there might be need to apply the doctrine toward IPRs or data. 55 Therefore, the application of the doctrine towards big data is challenging and requires considerable evaluation of whether, big data, a modern and technical product can fulfil the old requirements.
The Indispensability of Data – When deciding whether or not a facility is essential, the Court applies
a so-called indispensability test. The concept of this test is that there is no actual or potential substitute for the facility, on which the competitors can rely on, and that the facility must be objectively
51 Maurice Stucke and Allen Grunes (n 12) 42
52 Autorité de la Concurrence and Bundeskartellamt (2016) 11,
<http://www.autoritedelaconcurrence.fr/doc/reportcompetitionlawanddatafinal.pdf> accessed 30 May 2019
53 Daniel Rubinfeld and Michal Gal, ’Access Barriers to Big Data’ (Vol. 59:339, Arizona Law Review 2017) 349 54 Inge Graef (n 4) 245
necessary for competitors to be able to compete on the market.56 Data can qualify as indispensable only in specific situations.57 There have been cases where the Court ruled that the data was not indispensable as the data concerned was already available to competitors in the market.58 Despite these cases, there have been other cases at the national level where data has been considered to fulfil the conditions for indispensability.59 In these cases, it was considered that another undertaking in the market could not have reproduced the data in question at reasonable financial conditions and within a reasonable period of time. Therefore, the data was seen as indispensable. The data in question has to be truly unique, and there can be no reasonable ways for competitors to create, use or storage similar data on their own. Therefore, big data depending on the characteristics of the concerned data may fulfil the indispensability requirement in certain situations.
The Risk of Exclusion of All Effective Competition on the Downstream Market – In the previous case
law the Court has held that if the holder of the requested input is already active on the downstream market and reserves this market to itself by refusing to deal, this is affecting competition negatively and is therefore prohibited.60 However, in case the owner of the request input is only active in one market and is not trying to foreclose the competition, there is no anticompetitive behaviour. Therefore, an undertaking (which possesses the data) cannot be forced to give access to its data to an access seeker which needs the data to start competing in a market in which the provider is not present.61 It would be possible for the owner of the data concerned legitimately prevent access seekers from opening up a new market in which it is not active itself by refusing to give access to its data. Moreover, in the Microsoft case, the Court clarified that there is no need to establish that competition has been eliminated.62 As there is no requirement to prove elimination, only a risk of elimination, it seems that there should not be a problem to state that data as input can fulfil this requirement.
56 Arutyun Arutyunyan, ‘Intellectual Property vs. Essential Facility Doctrine: Microsoft vs. Commission’ (2004)
186-188
57 Richard Wish and David Bailey (n 27) 719
58 Case No COMP/M.4731 Google/DoubleClick [2008], paras. 365-366; Case No COMP/M.6314, Telefonica
UK/Vodafone UK/Everything Everywhere/JV [2012], paras. 539-558
59 Autorité́ de la concurrence, Décision n° 14-MC-02 2014 relative à une demande de mesures conservatoires présentée
par la société́ Direct Energie dans les secteurs du gaz et de l’électricité́, paras 169-174, 290-296, 285; Belgian Competition Authority, Beslissing n° BMA-2015-P/K-27-AUD 2015, Zaken nr. MEDE-P/K-13/0012 en CONC-P/K-13/0013, Stanleybet Belgium NV/Stanley International Betting Ltd en Sagevas S.A./World Football Association S.P.R.L./Samenwerkende Nevenmaatschappij Belgische PMU S.C.R.L. t. Nationale Loterij NV, paras 69-70
60 Allison Jones and Brenda Sufrin (n 9) 497-511 61 Inge Graef (n 4) 268
The Prevention of the Introduction of a New Product – When considering whether the new product
requirement should be applied to cases where big data is an alleged essential facility, the possible competitive advantages for the owner of the data and the potential entry barriers for competitors have to be analysed.63 In case the refusal to give access of particular data can create entry barriers, and its efficiencies to consumers are not significant, it is likely that the refusal will be considered to create competition law concerns and forced access could be a possibility to solve these concerns.64 Moreover, big data can create entry barriers and can prevent the introduction of a new product to the market.
The Absence of an Objective Justification – In cases where a dominant undertaking possessing the
asked essential facility refuses to give access to it, without any apparent and justified reason, the undertaking could be considered as acting against competition law and be forced to give access to its data under the essential facilities doctrine.65 However, in some situations, it is possible for the undertaking concerned to make an objective justification, justify the refusal. When the Court accepts the objective justification, the undertaking cannot be made subject to forced sharing. When the Court establishes that there is no objective justification or the undertaking is not able to provide adequate evidence to prove its justification, the facility can be made the subject of forced sharing.66 The objective justification will be analysed more deeply in the upcoming chapters of this thesis.
1.3.2. Application in Practice
As it has been stated, the application of essential facilities doctrine to big data can be seen problematic, because of the extraordinary character of the data. This character makes it different from other facilities, which can be made accessible to competing undertakings by forced access.67 For example, data can be seen as a non-rivalrous good, which means that it can be used simultaneously by more than one undertaking. In other words, in case one undertaking has collected and is using a specific type of data, it does not preclude other undertakings of gathering the same data. The non-rivalrous nature can impact the application of essential facilities doctrine as if the data in question is purely non-rivalrous, it cannot be found indispensable under the first condition of the essential
63 Inge Graef (n 4) 268-270 64 Arutyun Arutyunyan (n 56) 188 65 Inge Graef (n 4) 270-271 66 Arutyun Arutyunyan (n 56) 189 67 Inge Graef (n 4) 720
facilities doctrine.68 However, a case-by-case assessment should be conducted whether the indispensability requirement is fulfilled by the big data concerned. It is clear that all types of data are not able to justify this requirement, but it does not preclude the possibility of some data being able to fulfil the requirement. Besides of extraordinary character, the application of the doctrine to big data depends on whether the big data in question is capable of being an asset without which the competitors, of the owner undertaking, cannot conduct their businesses.69
When the Court is asked to determine whether the big data could be made subject to forced access under the essential facilities doctrine, it will start by defining whether the protected big data is indispensable for carrying on a particular business conducted by the competing undertaking. The party seeking to have access should demonstrate that the conditions for the essential facility are met. 70 The Court will determine whether the undertaking, by refusing to give access, could prevent the emergence of a new product. After this, the Court will evaluate the reasons for the justification. The undertaking making the refusal should illustrate justifications to back up its grounds for the refusal. Finally, the Court will examine whether the refusal can exclude all competition in the secondary market.71
How the access should be granted to the requested input is a different question. In case Microsoft,72 the undertaking was made obligated to make the necessary interoperability information available on the basis of reasonable and non-discriminatory terms. It can be concluded from the Microsoft case that the essential facility holder is obliged to share its data with its competitors by reasonable and non-discriminatory conditions. However, it is still on the owner of the essential facility to set the rate.73 In other words, in case the big data to which access is mandated has been developed by the undertaking itself, the undertaking should be entitled to demand for a reasonable remuneration rate.
68 Inge Graef (n 4) 262
69 Jere Lehtioksa (n 32) 22
70 Robert O’Donoghue and Jorge Padilla (n 43) 447 71 Arutyun Arutyunyan (n 56) 182
72 Case COMP/C-3/37.792 Microsoft v Commission (n 62) 73 ibid, paras. 1005-1008
2. ECONOMIC JUSTIFICATIONS FOR REFUSAL OF ACCESS
The second part of this thesis will analyse the possible exceptions for forced sharing under the essential facilities doctrine. It will be studied when the conditions for objective justification for refusal of access are met. The duty to deal, imposed under the essential facilities doctrine, is not automatically imposed. Not even in the case where the requirements for the essential facilities doctrine are met. Access to one’s facility should not be granted without a case-by-case assessment.74 The evaluation of the factors has to be thorough, and when undertakings can state clear and acceptable justification for why not to share their facility with competitors, the essential facilities doctrine should not be applied.75 The exception rule to the duty to deal under the essential facility doctrine was developed because the duty to deal (even when it aims to protect competition) can create competitive problems to the whole market, for example, by reducing the incentive towards investments. Therefore, in some specific situations, an undertaking is able to argue that the forced access, when granted, would have an adverse effect, and it should not be conducted.
2.1. An Objective Justification
An objective justification creates a legal reason to treat someone, or a certain situation, differently than others. Almost every prohibition in EU law is paired with an exception stating that, under certain circumstances, the rules may be set aside.76 A duty to deal, under the essential facilities doctrine, is no different. A dominant undertaking which is accused of abusive conduct, such as refusing to give access to its facility, may try to invoke an objective justification. If objective justification is successfully invoked, the Art. 102 TFEU prohibition does not apply in the case at hand.77 The objective justification plea is useful to provide more legal certainty and coherence as to the application of Art. 102 TFEU. Former Director-General of DG Competition Philip Lowe recognised three types of economical objective justifications, which could be applied in cases to justify refusal of access. Namely; undertaking’s commercial interest defence, legitimate public interest, and efficiency defence.78 These three defences will be analysed more precisely later in this chapter.
74 Commission Communication (n 44) para. 85 75 Robert O’Donoghue and Jorge Padilla (n 43) 447
76 Vijver Tjarda van Der, ’Objective Justification and Article 102 TFEU’ (Vol.35(1) World Competition 2012) 55-76 77 TFEU (n 22) Art. 102
78 Phillip Lowe, ’DG Competition’s Review of the Policy on Abuse of Dominance’, In International Antitrust and
When the dominant undertaking is claiming an objective justification to justify its refusal to grant access, it has to ensure that the conditions for an objective justification are met. Some of the conditions are general conditions, meaning that they have to be met before any objective justification can be considered to exist. These are the primary intent of the refusing undertaking, the refusal must be proportional, the refusal must fulfil the necessity test, and the effect of the refusal cannot be anticompetitive.79 However, the relevance of these elements depends on the type of justification that has been invoked. Besides, to these general conditions for all objective justifications, there are additional conditions for every defence type. The NCAs and courts must conduct a case-by-case analysis, whether the justification made by the undertaking concerned could be accepted.80
2.1.1. Legal Framework
Commission has issued Communication which has provisions focusing on refusal to supply.81 More precisely, it is dealing with the refusal to grant access to an essential facility or a network.82 These provisions guide how the essential facilities should be analysed, and when the refusal could be justified. It is stated that it could be possible for an undertaking to refuse to give access to its facility when the refusal “…is necessary to allow the undertaking to realise an adequate return on the
investment required to develop its input business, thus generating incentives to continue to invest in the future, taking the risk of failed projects into account.”83 Also, the Commission states that it will accept justifications based on the fact that the undertakings innovation towards its facility is, or will probably be, negatively affected by the obligation to share the concerned facility, and also when the refusal can create enough efficiencies to overweight its adverse effects.84
Moreover, a refusal to give access to an essential facility can be justified if the refusal is considered as being “reasonable”.85 Also, the refusal in question has to comply with the proportionality principle.86 Along with purely economic justification to the refusal of access, the facility in question
79 Vijver Tjarda van Der (n 76) 69-75
80 Thomas Graf, ‘How Indispensable is Indispensability?’ (Kluwer Competition Law Blog 18 April 2011)
<http://competitionlawblog.kluwercompetitionlaw.com/2011/04/18/how-indispensable-is-indispensability/> accessed 05 May 2019 81 Commission Communication (n 44) 82 ibid, para. 78 83 ibid, para. 89 84 ibid
85 John Lang, ‘The Principle of Essential Facilities in European Community Competition Law – The Position since
Bronner’ (2000) Journal of Network Industries <https://doi.org/10.1177/178359170000100401> accessed 05 May 2019
might possess limited physical or technological capacity (as it could be with big data), making the sharing of the facility impossible, as the rivalling undertakings would lack the technology and knowledge how to process the data/facility.87 Therefore, it can be stated that refusals based on technical or technological justifications can be accepted. The burden of proof is one the owner of the facility.88
2.1.2. An Objective Justification in Practice
In practice, when claiming an objective justification for refusal of access, the burden of proof is on the party who is seeking to claim objective justification. For example, the Microsoft case establishes that, “… it is for the dominant undertaking concerned, and not for the Commission […] to raise any
plea of objective justification and to support it with arguments and evidence…”89 Therefore, if the undertaking concerned is not making the plea towards objective justification, or it is not able to provide adequate evidence the Commission is not obligated under the law to investigate the possibility of justification for the refusal of access. In case the Court or NCA establishes that there is no adequate proof or the justification is otherwise not acceptable, the justification is not valid. Furthermore, the exception to the duty to deal is not applicable, and the undertaking can be made subject to forced access.90
To give a concrete example, in the Microsoft case, the General Court (hereinafter GC) concluded that the Commission had correctly dismissed Microsoft’s argument for objective justification for not to give access to its facility. In this case, Microsoft was arguing that its refusal to share its facility was justified as in the absence of that refusal there would be considerably lower incentives to innovate.91 Even this justification would theoretically be able to meet the conditions for an objective justification, (in this case) Microsoft was not able to sufficiently establish the negative impact on its innovation incentives but merely put forward vague argumentation.92 The burden of proof, and also the burden to provide adequate proof, is on the undertaking making the justification claim.
87 John Lang (n 85) 15-16
88 ibid
89 Case T-201/04 Microsoft (n 40) para. 688 90 Commission Communication (n 44) 91 Inge Graef (n 4) 173
2.2. Efficiency Defence
Efficiency defence aims to illustrate that the prima facie abuse of the dominant undertaking has no net harm effect on consumers.93 There are two types of efficiency defences, ex-post efficiency defence and ex-ante efficiency defence. Ex-post means ‘after the event’ and ex-ante means ‘before the event’.94 The ex-post efficiency defence is based on already established facts and efficiencies.95 To be able to rely on ex-post efficiency defence, the act has to be already committed, and the efficiencies resulted from the act has to be present. The opposite, ex-ante defence means that the action which could be considered anti-competitive is defended before the impact of the conduct is felt. In practice, the claim for an objective justification, under the ex-ante efficiency defence, is made before the efficiencies have born, when the outcome is still uncertain.96
In Post Danmark 1 case,97 the Court set out four additional conditions just for the efficiency defence. It must be illustrated that the efficiencies are likely to overweight any likely adverse effects on competition and consumer’s welfare, the efficiencies have been (or are likely to be) arisen as a result of the specific conduct in question, the conduct is necessary for the creation of the efficiencies, and the conduct does not eliminate effective competition, by removing all or most existing sources of actual or potential competition. Despite the possibility of efficiency claims, those are rare as they are usually dismissed by the Court. The problems of efficiency defences include the following. Firstly, the requirements for the defence are stringent.98 Secondly, the requirements are subject to a high standard of proof. Thirdly, the fourth condition in Post Danmark 1,99 that the conduct does not eliminate effective competition is often hard to prove. Fourthly, the formulation of efficiency defence appears to be narrow and focused on the specific transaction.100 Because of the mentioned difficulties of efficiency defences, there is yet to be any Art. 102 case in which the conduct has been saved from infringing Art. 102 solely because of efficiencies.
93 Commission Communication (n 44) para. 30
94 Marie-Anne Frison-Roche, ‘Ex ante/Ex post’ (Compliance and Regulation Law Bilingual Dictionary)
<http://mafr.fr/en/article/ex-ante-ex-post/> accessed 06 May 2019
95 Marc Fleurbaey, ‘A defence of the ex-post evaluation of risk’ (2010)
<https://www.law.upenn.edu/institutes/cerl/conferences/prioritarianism_papers/AdditionalPapers/FleurbaeyNotesAdler. pdf> accessed 06 May 2019
96 ibid
97 Case C-209/10 Post Danmark A/S v Konkurrencerådet (hereinafter Post Danmark 1) [2012] ECLI:EU:C:2012:172 98 The dominant undertaking in question has to ensure that there is no net harm to competition nor to the consumer
welfare, this ‘balancing exercise’ is notoriously difficult and usually the undertaking needs to make the assessment before it implements its commercial policy
99 Case C- 209/10 Post Danmark 1 (n 97) para. 42
2.2.1. Benefits for Consumers and Specific Conduct
In the Commission’s Communication, it is recognised that the benefits which the conduct (refusal in this case) can provide have to be felt at the consumer level. Therefore, the first condition for an objective efficiency justification is that the conduct in question can provide efficiencies, and these efficiencies born have to be also enjoyed by the consumers in the relevant market.101 These benefits could be, for example, lower price of the final product or service, a better quality of the products, or more services offered.102 Also, it is important that these benefits are ascended from one specific conduct, the same conduct which is seeking to be justified. In case of refusal to give access, a dominant undertaking refusing to give access to its facility has to illustrate that this refusal can offer efficiencies to the consumers. The benefit cannot be a result of a combination of different conducts. It has to be born from specific conduct.103
2.2.2. Necessity - No Less Anticompetitive Alternatives Available
By the necessity condition, the Commission and the courts are illustrating that the restricting of competition (by refusing to give access) should not take place when the refusal is not necessary. However, when the refusal is considered to be necessary, the conduct (refusal) should be as limited as possible.104 The objective justification is only possible if the refusal is highly critical towards, for example, the protection of the business of the undertaking, and there is no other possible way, which would be less burdensome to the competition and the relevant market, to gain the same outcome. How to decide which conduct is necessary is on the NCAs and the courts to determinate.105
2.2.3. Outweighing Competition and Consumer Harm
Outweighing competition and consumer harm condition could be called the balancing condition, as here the courts and NCAs should take into account, and compare the anticompetitive effects of the conduct at hand (the refusal) and the efficiencies which it can create.106 When the efficiencies, born from the refusal, have “bigger” positive effect or value than the competition concerns born in the
101 Case T-203/01 Michellin 2 (n 100)
102 Keith Hylton, Antitrust Law and Economics (2nd edn, Edward Elgar 2010) 52
103 COMESCA Competition Commission ‘A Guide to Anti-competitive Business Practices’
<https://www.comesacompetition.org/?page_id=498> accessed 06 May 2019
104Commission Communication (n 44) 105 ibid
106 Case C-209/10 Post Denmark v Konkurrencerådet ECLI:EU:C:2012:172, para. 42; Case C-95/04 P British Airways
same conduct, the refusal might be justifiable under this condition. The efficiencies created by the refusal to give access to the facility can outweigh the negative impacts of the same refusal.107
2.2.4. Maintaining Effective Competition
The last condition for the efficiency-based objective justification concerns the possibility of the refusal (the conduct which is sought to be justified) to limit competition in the relevant market. It is required that the conduct is not limiting the whole competition.108 Nevertheless, it does not disclose the possibility to limit competition. It only states that effective competition cannot be totally limited.109 Therefore, to meet this condition the dominant undertaking refusing to share its facility has to illustrate that the rival undertakings are still, after the refusal, able to operate profitably in the relevant market without the access to the concerned facility possessed by the dominant undertaking. Moreover, their business operations and profit-making capabilities are not wholly dependent on the facility of the refusing undertaking.
2.3. Public Interest Defence
The second type of economic objective justification is the public interest defence. This defence is based on the Court’s opinion that Art. 102 TFEU must be interpreted in light of the principles and objectives of the Union.110 For example, Art. 3(3) of the Treaty on European Union (hereinafter TEU) states that the Union strives for sustainable development, social progress, and a high level of protection and improvement of the quality of the environment.111 These goals can and are influencing to the application of Art. 102 TFEU. Moreover, well-known cases where parties of the dispute have used the public interest defence are Hilti112 and Tetra Pak113. However, in both of these cases, the
plea for public interest was denied, as the Court considered that the evidence to back up these pleas were not efficient enough.
107 Armando Marrapodi, ‘Freedom to Conduct Business and Abusive Refusal to Deal: Do dominant undertakings enjoy
limited contractual freedom?’ (Master thesis, Lund University 2018) 32
108 Commission Communication (n 44) paras. 5-6 109 Armando Marrapodi (n 107) 27
110 Case C-6/71 Continental Can v Commission [1973] ECR 215, para. 25
111 Consolidated version of the Treaty on the European Union [2012] OJ C326, Art. 3(3)
112 Case T-30/89 Hilti v Commission [1990] ECR II-163, para. 118. For the ECJ appeal Case C-53/92 P Hilti v
Commission, [1994] ECR I-667
The Commission has dealt with public interest defences in various Art. 102 TFEU cases. For example, in the Port of Genoa case, the Commission established that environmental concerns could be subsumed under an objective justification,114 and in GVG/FS, the Commission stated that public safety considerations could also constitute an objective justification. Nevertheless, in these cases, the Commission rejected the public interest pleas, by stating that the reasons for rejection were not a matter of law, but a matter of the facts of the case. The objective justification pleas, under the public interest defence, seem to have been unsuccessful as the conducts at play have not been able to meet the general conditions for all objective justification, namely necessity and proportionality conditions. From the foregoing, it can be concluded that it does not matter whether the public interest defence would be accepted if the case is not able to fulfil the four common conditions for all objective justifications.
2.4. Own Commercial Interest Defence
Art. 102 TFEU does not prohibit competition based on merits.115 Competition law should not intervene if a dominant undertaking can outperform its competitors simply as a result of superior industry or innovation. Commercial interest has been described as an interest consistent with the rational profit-maximizing behaviour of an undertaking.116 Protection of the undertaking’s own commercial interests’ as a defence was first accepted in the landmark case United Brands.117 There the Court established that Art. 102 TFEU does not prevent an undertaking from protecting its own commercial interest when they are attacked by rival undertakings. Nevertheless, the principle is not applicable when the undertaking’s real intent is to strengthen its dominant position, or when it is abusing its dominant position.118 The response of the dominant undertaking concerned has to be reasonable as well as proportional. Case Tetra Pak confirms this by establishing that: “… the
undertaking in a dominant position may act in a profit-oriented way, strive through its efforts to improve its market position and pursue its legitimate interest. But in so doing it may employ only such methods as are necessary to pursue those legitimate aims. In particular it may not act in a way which, foreseeably, will limit competition more than is necessary.”119
114 Commission Decision 97/745/EF Port of Genoa [1997] L 301/27, para. 21
115 Case C-87/76 Hoffmann-La Roche [1979] ECR 461, para. 91; Case C-280/08 P Deutsche Telekom v Commission
[2010] ECR I-346, para. 177
116 Allison Jones and Brenda Sufrin (n 9) 375-376
117 Case C-27/76 United Brands [1978] ECLI:EU:C:1978:22, paras. 189-190 118 Richard Wish and David Bailey (n 27) 713-726
Furthermore, the protecting legitimate commercial interest defence has especially been used in cases concerning refusal to supply.120 For example, in the United Brands case, the Court stated that the refusal to supply was not abusive conduct in situations where the refusal is conducted to protect the commercial interest of the undertaking.121 This ruling clarifies that an undertaking is entitled to take steps which it deems necessary to protect its commercial interest. Commercial interest defence is used by undertakings to justify their refusal to deal more often than the previously dealt efficiency and public interest defences. Reason for this is that the commercial interest is much easier and simpler to illustrate by the undertaking than the other defences, as their conditions are stricter and harder to fulfil by the dominant undertaking which is seeking to justify its refusal.
When considering big data as an essential facility, it might be seen unreasonable to force a dominant undertaking to share its facility in a situation where the undertaking has legally obtained its advantages (by innovation, research and technology developments).122 The competing undertakings might try to take advantage of the essential facilities doctrine and only ask/demand access to gain knowledge of the technology, data or research without wasting their own time and money to do the work required by themselves. The Court has accepted the commercial interest as an objective justification form, as in certain situations undertakings must be able to protect their facilities against hostile undertakings which might be seeking access on unreasonable grounds.
120 For example, in landmark cases United Brands (n 112) and BBI/Boosey & Hawkes [1987] OJ L282/36 121 Case United Brands (n 117) para. 189
3. DATA PROTECTION JUSTIFICATION FOR REFUSAL OF ACCESS
In this final chapter, it is aimed to combine the previously discussed topics and move towards the newest possibility for dominant undertaking to justify its refusal of access to its facility under the essential facilities doctrine. Namely, the possibility of data protection to qualify as an objective justification form. The aim is to evaluate the interaction between the forced sharing of big data and data protection rules.123 As previously mentioned, big data is being implemented and used to create value in many different settings, but what happens when data includes personal information of natural persons and is protected under data protection laws?124 How is the effective competition maintained while at the same time protecting the data subjects? This chapter is aiming to clarify these questions and explain how would the data protection work as an objective justification invoked by a dominant undertaking when refusing to share its facility with other undertakings in the relevant market.
3.1. Big Data, Competition Law and Data Protection
Competition and data protection laws interact in terms of the substance of their respective obligations. To be able to analyse how they interact, it is instructive to first look at the most critical data protection principles.125 These principles illustrate that the nature of data is vital, as whenever the data in question includes personal data of natural persons the data protection law provisions are applicable and should be taken into account.126 In other words, before analysing the possible justifications based on data protection rules, it should be made clear what is data protection, when does big data fall under the data protection rules, and how dominant undertakings might be able to use data protection rules to defend their refusal to grant access to their facilities.
123 Paul Lugard and Lee Roach (n 17)
124 Article 29 Data Protection Working Party, ‘Opinion 1/2008 on data protection issues related to search engines’
(2008) 00737/EN WP 148 <http://194.242.234.211/documents/10160/10704/WP148+-
+Opinion+on+data+protection+issues+related+to+search+engines> accessed 05 June 2019; Tim Cowen, ‘Big Data as a Competition Issue: Should the EU Commission’s Approach Be More Careful’ (Vol. 4 European Networks Law and Regulation 2016)
125 European Data Protection Supervisor, Preliminary Opinion: Privacy and competitiveness in the age of big data: the
interplay between data protection, competition law and consumer protection in the Digital Economy (2014) <https://edps.europa.eu/sites/edp/files/publication/14-03-26_competitition_law_big_data_en.pdf> accessed 05 June 2019
126 Marina Lao, ‘Search, Essential Facilities, and the Antitrust Duty to Deal’ (Vol. 11 Northwest Journal of Technology
3.1.1. Data Protection
Data protection is legislation created to protect personal, commercial, or governmental data from unwanted and unauthorised access, alteration, destruction, or use. It aims to pounce a balance between individual privacy rights and at the same time, allow data to be used for business purposes.127 The terms ‘data protection’ and ‘privacy’ are often used interchangeably. However, there is a difference between these two terms. The scope of the right to data protection is more limited,128 it only comes into play if personal data is processed. The right to privacy, on the other hand, protects the private sphere of individuals irrespective of whether any processing of personal data is involved.129 This thesis focuses on the first one (data protection). In the EU, data protection enjoys special protection, meaning that its protection is recognised under the ECHR130, Charter131 and TFEU132. Besides the primary law, secondary law also provides data protection legislation. Most important of them is the General Data Protection Regulation (hereinafter GDPR), which came into force in May 2018.133
Under the GDPR, personal data is protected.134 Personal data can encompass any information relating to an identified or identifiable natural person who might be identified, directly or indirectly, in particular by reference to an identifier such as a name, location data, an identification number, an online identifier or to one or more factors specified to the physical, physical, genetic, mental, economic, cultural or social identity of that person.135 The definition of personal data is broad. Besides to the limitation, that the GDPR only applies to personal data, the GDPR states conditions for the lawful processing and use of personal data. It is lawful to process the data if the data subject has given its consent for the processing, the processing is necessary for the performance of a contract, the processing is necessary for compliance with a legal obligation, or the processing is necessary to be able to protect the interest of the data subject or another natural person.136 The underlying idea of the GDPR is to lay down guidelines on how the processing of data can be conducted while still protecting the privacy of natural persons.
127 Ann Jones ‘Data Protection Policy’ (Heriot-Watt University 2018)
128 Peter Blume, ‘Data Protection and Privacy – Basic Concepts in a Changing World’ (Scandinavian Studies in Law,
2015) 153-154
129 The European Union Agency for Network and Information Security (hereinafter ENISA), ‘Privacy by design in big
data – an overview of privacy enhancing technologies in the era of big data analytics’ (2015)
130 The European Convention on Human Rights [2010] (hereinafter ECHR) Art. 8 131 Charter of Fundamental Rights of the European Union [2000] C 364/01, Arts. 7-8 132 TFEU (n 22) Art. 16
133 Council Regulation (EU) 2016/679 of 4 May 2016, The General Data Protection Regulation (hereinafter GDPR)
[2018] OJ L127; Another secondary legislation example could be the Directive 96/9/EC of the European Parliament and of the Council [1996] OJ L77
134 GDPR (n 133) Art. 1 135 ibid, Art. 4
