Big data is a term which was associated with multiple meanings in the recent years. In general, this is a technological trend of increasing importance, enveloping the gathering of data flows (mostly personal) in various formats (video, photos, text, etc.) and having diverse origin for the purpose of a more objective view on certain phenomena and improvement of the process of decision making. In the basic of the concept stands the fact that almost everything we do in our daily lives, leaves digital footprint - listening to music, shopping, working, social interaction, even driving. On the other hand, the emergence of ubiquitous computing and the Internet of Things proved that the computer is not only a unit which is inseparable from the desk and gave this technology’s development an extra boost.
At the end of September this year, Brussels hosted a conference, organized by the European Data Protection Supervisor and the European Consumer Organization. At the center of discussion was the topic of Big Data through the prism of the individual European citizen and their rights as a consumer. One of the most interesting speeches of the event came from Margrethe Vestager herself, the European Commissioner responsible for the competition in the EU, who commented on the problem of the intersection of “big data,” competition and privacy.
The topic has long lasting presence in the public domain, the first time the discussion being raised was in the preliminary opinion of the European Data Protection Supervisor from 2014. The topics commented in the document are quickly pick up by both academics and institutions. Nevertheless, the European institutions seem to still refrain from taking a hard line on the issue, with the exception of occasional public appearances of representatives of the European Commission, which periodically put the topic on the agenda.
In her speech, the Commissioner focused on several key cornerstones being important for the understanding of the issues that are to be discussed in detail in this article.
Big Data as a resource and market force source
On the first place, Margrethe Vestager unveils the role of the data as an intangible economic resource, beneficial for the enterprises that have it. Evidences in that regard can be found in the success of the ‘’data markets,’’ and even the emergence of a new profession – data broker. We should also mention the frequent tendency of mergers and acquisitions, [1] primarily motivated by the willingness of increase and diversification of the data catalog acquired by the successor. Big Data allow businesses to create products and services that are more attractive to their customers, improve productivity and reduce business risk. A good example of this is the search engines - a large number of users who use the service, combined with the ability to monitor their behavior after looking for certain keywords, means more information which is the key to improving the service and search refinement of its algorithms. On the other hand, Big Data help companies in creating innovations, which would reveal unfamiliar markets and attract new customers. A typical example of this trend is the so called two-sided platforms such as social networks and search engines. By offering free goods or services, these companies compete for users who, taking advantage of the free good offer, generate personal data of economic value to another group of users of the platform – advertisers, for example. The more detailed the information about a particular user is, the higher the price the advertiser is willing to pay for it. In this way, a manufacturer of golf equipment, for instance, would have directed their advertisement towards persons who have previously used the search services to find out about events or topics related to golf or who are members of golf fan groups in social networks.
In pursuit of this profitable model, completed business strategies based on “big data” as a key source of enterprise value become more frequent. Moreover, companies that traditionally have not been active in the digital industry, also use Big Data to improve their services and to monitor more effectively market trends, on which their business operates. One of these examples is the accounting and consulting house PricewaterhouseCoopers, which through the use of analytical algorithms successfully improved internal corporate process of decision-making and service to its customers.
It is evident that the progress of technology is used both by businesses and consumers, as it ensures better goods and services, and often, more affordable prices. According to the Organization for Economic Cooperation and Development “big data is a key economic asset that can provide significant competitive advantages for companies...” In a more disturbing continuation of that statement, the European Data Protection Supervisor warns that “extracting value from big data has become a powerful source of supremacy for the biggest players on the Internet market.”
Change in the paradigm
The specifics of Big Data make businesses in the digital sector highly dependent on their accumulation, which can easily turn harming for the consumer, as well as turn into an opportunity for dishonest behavior among competitors. Hypothetically, in the context of Big Data and competition, such behavior could take two basic forms - when the company collects large data flows through anticompetitive means or where a certain undertaking blocks the attempts of others to gather their own information resource and thus creating insurmountable barriers to access or adequate competition in a particular market.
The first hypothesis can be classified as an abuse of dominant position, expressed in imposing general terms and conditions which are disadvantageous for consumers’ protection of personal data. A leadership in a particular sector enables an enterprise to make consumers face a choice of the type „take advantage of the service as it is, or not use it at all.“ This choice will be further complicated by the so-called Network Effect, typical for digital platforms or also called a snowball effect - the more people use a service, the more customers it will attract. The chance of a user to permanently stop using Facebook because of a unfavorable policy platform will significantly decrease if all their social networking contacts continue to use it. Thus, established dominant position would allow an undertaking to „bend“ the rules in their favor, a privilege that its competitors could not afford. Moreover, this practice will lead to more extensive data collection, which will further cements the leadership of the first company.
This scenario actually does not remain solely to the realm of possibilities. In March 2016 the German authority for protection of competition (Bundeskartellamt) has initiated proceedings against Facebook for abuse of dominant position under Article 102 TFEU, resulting in violation of European legislation on data protection by insufficiently explicit terms and conditions which mislead consumers about the aims their information was going to be used for.
A “comprehensive” look, combining the areas of competition law, data protection and consumer law, so far is unknown to the European and national competition protection authorities. However, we can say that in the future the effectiveness of such approach in analyzing the specific characteristics and defects of the digital market should not be underestimated. The methodology combining these seemingly unrelated fields of law, seems quite logical, if we consider that in the digital economy the subject of personal data collection is often a consumer as well. On the other hand, it must be remembered that the ultimate objective of competition law has always been to protect the interests of the end user, rather than punish those who violate the law per se.
Despite this, the European Commission seems to be still skeptical regarding such change in the regulation of the competition law, especially in the field of the control of the merger of undertakings. [2] In two separate decisions, examining the acquisition of Whatsapp by Facebook and of DoubleClick by Google, the Commission clearly recognized the potential elimination effects that these acquisitions might have. It concluded that combining data which previously belonged to the individual enterprises would lead to the creation of a competitive advantage that cannot be overcome by other market players. However, in terms of protection of personal data in its Facebook / Whatsapp decision the Commission said that “any privacy-related concerns flowing from the increased concentration of data within the control of Facebook as a result of the Transaction do not fall within the scope of the EU competition law rules.” [3] Despite being inline with the letter of the law, such a claim limits the scope of the assessment made by the regulator, as it deprives it from the opportunity to consider all outcomes that such a concentration would have had on the welfare of consumers and on the competition. In fact, two years after its adoption, in September 2016, the German authorities for data protection issued an order for Facebook to cease the use of personal data of German users of Whatsapp, for lack of any legal basis for it in within the meaning of Directive 95/46 on data protection. The case illustrates a classic method called Function Creep, in which the personal data given by users for specific purposes, is used for a completely different purpose.
What has been said up to this point clearly proves that merger of undertakings have the potential to lead to competition issues as well as issues in the field of data protection. Let us imagine a situation in which two companies providing similar services in the same market segment, competing in the protection of personal data of their clients. Logically, if the two companies decide to merge, forming a new market leader, the need to maintain the “friendly” consumer policy will be abandoned. This will lead to collection of more users’ data, which in turn can be seen as a raise the “price” of the service; and by price here we mean the personal data that users “pay” to use the service, but seen from a different angle - a decline in its quality. Thus, in theory, such concentration may cause harm to the end users according to the competition terminology, disregarding the fact that the damage is also a violation in the area of data protection. This theory, however, will only be valid in areas where privacy is a key element in taking end-user decision. Meanwhile, consumers’ perceptions of privacy and data protection are highly subjective and vary depending on the type of services provided (especially if they are free), the age of the subject, social status and a number of other contextual specifics. This makes the establishment of general and generally applicable rule for this type of situations extremely difficult.
Indeed, not every violation of the European legislation on data protection by a company is a breach in the field of competition law. However, it cannot be denied that the most severe violations of the right to privacy of Internet users were committed precisely by the most influential actors on the digital scene. Therefore, the essence of the topic lies in the study of the functional relationship between these two circumstances. When such behavior is directly related to the market position of the offender, then the situation should become a competition law problem. Moreover, similar multidisciplinary approach was perceived by the European Court of Justice in Case C-457/10 P of 12/06/2012. In the judgment of that case the Court concluded that the company AstraZeneca had abused its dominant position (Article 102 TFEU), violating provisions in the field of intellectual property law, in particular committing ‘abuse of the patent system and the system for authorization of medicines,’ preventing or delaying the entry of competing medicines on the market. Indeed, intellectual property law initially has some connection with the competition area, mainly due to the institute of exclusive rights to industrial property. However, it is quite logical to find similar correlation between competition law and data protection.
“Essential facility” or minor restriction?
The collection of Big Data allows businesses to converge significantly the quality of goods and/or services to the needs of its users. Moreover, Big Data serve as a catalyst for innovation, enabling discovery of unfamiliar markets and increasing the customer base. An example of this is Google, which first mapped the world through Google Street View, and then used what they had learned about traffic and urban planning, to become the first pioneer in the field of autonomous vehicles, followed closely by the electric giant Tesla. In a similar way, Big Data, combined with artificial intelligence, laid the foundations of the market for the so-called digital assistants, such as Apple Siri, Microsoft Cortana, Google Now and Alexa from Echo device by Amazon.
The common thing between the autonomous cars and the digital assistants is that they are presented in a market environment in which business participation is largely determined by the access to large volumes of data, and the necessary for their collection and processing infrastructure. This feature exists partly due to the growing scope of the technology of artificial intelligence and machine learning that function more smoothly, the greater the information base from which it “learn” is. For example, behind the voice of Siri, sophisticated neural networks function enabling the service itself to improve after every conversation that it has with the consumers. In practice, the more customers use Siri, the more precise it becomes in guessing their needs.
The aforementioned gives to a number of experts a reason to argue that in such heavily dependent on digital information industries Big Data will likely take on the role of “essential facility”, a key object that falls within the exclusive property of a particular holder, depriving all of the other smaller enterprises from participation in the relevant market, thus damaging the production, the technological development and the interests of the end user. The doctrine of the “essential facilities” appeared for the first time in the USA, but was quickly accepted by the European competition regulators, today being mostly applied in with respect to Article 102, point b TFEU. Although in its initial application theory focuses mainly on power and rail infrastructure, in the European Commission’s most recent decisions in cases such as IMS Health and Microsoft it regards as an essential facility the intellectual property objects. For example, in the case of Microsoft, the corporation is required to provide licensing access to information related to the operating system, which is crucial for creating applications interoperable with the Windows software.
The doctrine of the essential facilities is subject to much criticisms because of the door that it opens for great interference in the freedom of contractual relations and its destructive influence on incentives for innovation. Therefore, in practice, the European institutions have always tried hard to narrow the optimal application of its range, introducing a number of mandatory requirements that must be fulfilled, such as: can an essential facility be recreated solely by competing companies; are there real alternatives to the object and is it indeed “essential” for participation in the relevant market; to what extent is a refusal by the dominant undertaking for access to an essential facility objectively justified; and others. This extensive catalog of criteria makes the doctrine relatively difficult to and rarely applied in practice. That said, the author believes that the theory of “essential facility” is hard to find an application in terms of Big Data for the following reasons.
First, the data are inexhaustible resource. The fact that an undertaking has a huge amount of information does not prevent other players to compete against it. They can build their own information resource or if they do not have the necessary infrastructure to do so to obtain information from third parties through data brokers for example. Last but not least, as Commissioner Vestager pointed out, undertakings can overcome this barrier by merging their data resources through cooperative actions such as consortia, licensing agreements and through participation in open source and open data initiatives.
Secondly, in the center of the Big Data does not stand quantity as much as quality and variety of data. Owning a lot of data does not apriori constitute a competitive advantage. The real potential behind the technology lies in finding unsuspected correlations in the data and properties of heterogeneous character, and this is more a matter of good analytical algorithms, human resources and effectively established business models.
Although it is unlikely the Big Data to become “essential facility” it is quite possible that they become the reason for other unfair practices such as discrimination in relations with trading partners under Article 102 TFEU. The development of the industry will prove to what extent this can really happen in practice.
Instead of a conclusion
Regardless of the outcome of the debate over whether the topic of data protection can become part of the Competition dogmatics, one thing remains clear - Big Data become more and more important in matters falling within the field of competition. The way in which the data affect the behavior of enterprises have yet to be the subject of study in the doctrine and the practice. The results of this debate should lead to the creation of more effective tools to ensure that as far as the Big Data is concerned, the behavior of the enterprises will be in line with the rules and principles of fair competition.
Going back again to the acquisition of Whatsapp by Facebook, it should be mentioned that it fell below the minimum thresholds established by Regulation (EC) 139/2004, over which the operation must be considered by the Commission as having a Community dimension. The reason for this is that the thresholds are based on fixed quantitative indicators such as total annual turnover. Thus, thresholds are not taken into account for acquisitions of small start-up enterprises, possessing valuable resource of data that has not yet found its monetary equivalency on the market. The economic importance of this factor is the fact that Facebook paid 19 billion dollars for a company with only 10 million euros in annual revenue. At the end of his speech, Commissioner Vestager left this discussion open, promising that if the Commission deems it necessary, new legislative steps are to be taken in a year to fill these gaps at EU level.
Until then the situation will be mitigated by the effect of tools that address both issues in the field of competition law and the data protection. Such tool is the right to data portability, determined in Art. 20 of the new General Regulation on Data Protection (GDPR). Its content allows each subject to request all the personal information regarding them from an administrator in a “structured, commonly used and machine-readable format” for the purpose of transferring such data to another platform. In addition, to ensure extensive control over personal data, the right allows users to “choose” between alternative platforms considering their specific needs for privacy, and thus motivate businesses to compete in providing services optimally respecting the interests of their users. It must not be forgotten, however, that the scope of GDPR and its instruments is going to be limited to only that portion of Big Data that qualifies as personal data [4] as well.
It can be assumed with great certainty that Big Data is going to continue to be a key part of the business strategies of the undertakings. What should not be forgotten is that the usage of the benefits of this new technology should not be done in such a way as to violate the basic rights of consumers. After all Big Data exist because of them.
This publication is originally written in Bulgarian language and most of the sources lead to texts in Bulgarian. In case of any questions on the English translation of the publication or the cited sources, do not hesitate to contact the team of LIBRe Foundation at:
office@libreresearchgroup.org.
All publications and comments published on the LIBRe Stories platform can be quoted in other websites or in articles in the press subject to the condition of identifying the author, the publishing date and their source (URL address of the publication in question).
References
[1] Exampled of such practices are the mergers between TomTom and Tele Atlas; Google and DoubleClick; Bazaarvoice and PowerReviews Inc. etc.
[2] The control over the merger among undertakings is part of the Commission’s competence under Regulation No 139/ 2004 of the Council, 20 January 2004, promulgated in the OJEU, section L 24/1 on 29.1.2004 http://eur-lex.europa.eu/legal-content/en/ALL/?uri=CELEX%3A32004R0139
[3] §164 of the decision.
[4] According to the legal deffintion in art. 4 of GDPR
