This entry to our annual Sir Henry Brooke Student Essay Prize by Benjamin Evans from the University of East Anglia was highly commended by the judges. A second highly commended entry will be published next week.
Our entrants were asked to answer this question: “At the start of the 2010s, two billion people used the Internet, MySpace rivalled Facebook as the most popular social network, iPads did not exist and few people had swapped their trusty Nokias for iPhones. Peer to peer networking was seen as an existential threat to copyright industries and net neutrality was not yet the law anywhere, while cloud computing was unknown to the general population. The future was unpredictable.” What, where, how, and when will the greatest regulatory challenges for the Internet of 2030 be?’
The Internet permeates almost every conceivable sector in international commerce, and it underpins some of the twenty-first century’s most innovative companies whose disruption of traditional markets led many to believe that “the future was unpredictable.” Yet, as the proverbial dust has started to settle, it is apparent that these disruptors are merely harnessing the potential of the Internet in order to ply the same broad set of goods and services to consumers. This essay posits that although regulatory challenges for the Internet of 2030 can readily be identified in a number of discrete areas of law, such as the law of privacy governing social networking or the law of Intellectual Property relating to peer to peer networks, the greatest regulatory challenges exist at the macro level of supra-national competition law. Indeed, whilst few academics and policy makers would dispute the potentially positive long term impact of the Internet on human welfare, there is considerable debate concerning the actual, immediate term impact digitisation is having on consumer welfare. The European Union (EU) competition authority is the largest supra-national competition regulator and its competition law is designed to protect the welfare of the greatest consumer population in the world. By critically assessing the fitness and efficacy of EU competition law in light of significant changes in the landscape of competitive markets, focussing primarily on the recent approach of the European Commission (EC), this essay will explore the principle challenges that the rise of the Internet and the emergence of new digital business models pose to EU competition law, and in so doing draw attention to areas of competition law that may need to be reformed to regulate the Internet of 2030.
Abuse of Dominance by Digital Platforms
Since 2017 the EC has been in pursuit of the digital platform Google and the sparring match has taken the form of three landmark cases under Article 102 TFEU concerning abuse of dominance.1 In the first of these cases Google Search (Shopping) it was held that the technology firm had systematically abused its dominant position by favouring its own price comparison results over the results of competitors in the price comparison marketplace in product searches by users of Google Search.2 The language of the EC reflected the scale of the €2.42 billion fine it imposed: “Google's practices have stifled competition on the merits in comparison shopping markets, depriving European consumers of genuine choice and innovation.”3 In addition to the fine, Google was required to amend its algorithm. The EC’s actions resulted in a change in the appearance of the Google search user interface, and an “auction system” was introduced to allow competitors to Google to bid for the opportunity to appear in the Google Search advertisement banners at the top of the interface. Although on the surface these remedial measures may be judged as significant, in fact they have been the subject of weighty criticism, not least as the “auction system” permits Google Shopping to engage in what is effectively internalised bidding alongside other competitors to Google. Indeed, critics continue to assert that the problem will persist because the search result slots are populated by auction and not by relevance; a toxic system that leads to bidders sacrificing a substantial percentage of their future profits in the attempt to gain traction in an oversaturated, artificial marketplace.4
Most recently, in Google Search (AdSense) the EC turned its attention to restrictive clauses in the company’s agreements with third parties, which had the effect of “preventing rivals from competing in the online search advertising intermediation market.”5 By the time a fine of €1.49 billion was imposed on Google by the EC, it had already ceased using such clauses in its agreements; therefore, upon finding that no efficiencies could be proffered by Google to justify restrictive clauses ongoing, it was simply prescribed that the company must “refrain from any measure that has the same or equivalent object or effect.”6 It is perhaps not surprising that detractors claim that far from offering a robust solution of any permanence, this simply creates an environment where there is a marginally heavier threat of periodical scrutiny by the EC. Indeed, it could quite reasonably be posited that the scale of the fine and the scale of deterrent are disproportionately modest, and perhaps lenient, towards Google taking into consideration the effect of the abusive behaviour in the competitive marketplace.
Looking ahead, digital platforms will continue to be a major priority and the EC’s 2019 report Competition policy for the digital era critically assesses these platforms in two distinct ways. Firstly, the report examines competition “for” the market. Due to network externalities and returns to scale, it is natural that even in a highly competitive marketplace there can exist only a small number of rival digital platforms.7 Hence, the report strongly recommends that the competition authority should take action to ensure that this small number of dominant digital platforms are unable to erect barriers to entry including: Most Favoured Nation clauses or best price clauses, and prevention or restriction in regard to multi-homing, switching services and complimentary services.8 Indeed, the report highlights that a fit and efficacious EU competition law regime should scrutinise any activities by dominant platforms that appear to disadvantage and discourage rivals either from garnering a critical mass of users, or raising their costs.9 Secondly, the report considers competition “on” the platform, which it recognises “in many cases might be the same as protecting competition “in” the market”.10 Accordingly, the report recommends that the competition authority should be vigilant and take action against digital platforms that appear to be engaged in “abusive self-preferencing”, or to be actively selling “monopoly positions” to their users.11 In fact, the report posits that digital platforms should be mandated to design and uphold rules that “do not impede free, undistorted, and vigorous competition without objective justification.”12 This high standard, which is in line with the longstanding competition law concept of special responsibilities, is of particular importance given the almost regulatory role that dominant digital platforms are increasingly playing.13
Of course, the EC needs to apply its theories to case law and its ongoing investigations into the conduct of the technology giant Amazon stand to provide the most thorough and far-reaching insight into the fitness of competition law. Upon unveiling the EC’s 2019 report at the G7 Conference, Margrethe Vestager stated that: “We need to be mindful of new strategies that can harm competition, so we’re ready to deal with them when we meet them in practice, when the evidence shows that competition could be harmed, even when it is hard to put an exact figure on the cost to consumers … It may even be necessary for governments to reassert control of parts of the digital world, when they find that commercial interests alone don’t provide the services we need.”14 And with the Platform to Business Regulation in effect as of 12th July 2020, it is clear that EU competition law is evolving new and advanced tools for dealing with certain types of anti- competitive behaviour. However, whilst restorative remedies are hinted at by Vestager and in the report, in particular in cases where self-preferencing can be shown to have conferred a substantive competitive advantage in terms of comparative market power, it may be somewhat optimistic for Amazon detractors to expect the EC to flex its muscles to a truly radical extent.15
Turning attention back to the EC’s landmark Google abuse of dominance cases, arguably the most significant debate arising therefrom is concerned not with the specific anti-competitive activities and remedies, but with the overarching issue of market definition and more specifically the competition authority’s apparent struggle to apply traditional market definitions to the digital economy.16 In particular, in Google Search (Shopping) the EC defined the market as “consumer shopping services” and, whilst this included such websites as Yahoo! Shopping, it in fact failed to include digital marketplaces including Amazon, Alibaba, eBay and Facebook Marketplace.17 Critics have asserted that the shortcomings arising from this market definition are evidence that traditional concepts are inadequate and being inefficaciously applied by the competition authority, which is not sufficiently mindful of the realities of the digital marketplace in which differentiated offerings compete within the same market boundaries for consumer attention.18
Data Rich Mergers and Acquisitions
The EC has been active in the area of Merger Control in the digital economy in recent years, with increasing focus on data rich mergers and acquisitions. In one of its earliest landmark decisions Facebook/WhatsApp, the competition authority has been heavily criticised for relying upon evidence supplied by the social network giant to reach a conclusion that the limited data integration potential in the merger was sufficient to negate concerns voiced by detractors relating to network effects.19 Upon discovering that the evidence supplied by Facebook in relation to integration had been misleading, the EC issued a fine of €110 million. On the one hand, it could be held that the EC demonstrated its ability to levy fines in the event of a breach under merger control; on the other hand, as critics are hasty to point out, the 2014 decision is evidence that the competition authority is not suitably versed in dealing with cases involving companies operating in the diverse digital economy. Regardless, the case provided an important opportunity for the competition authority to start to develop a framework for applying traditional and proven analytical approaches to marketplaces distinguished by starkly different characteristics.
The most recent decision by the EC relating to a data rich acquisition is Apple/Shazam, a case that indicates a clear shift in approach.20 Two key areas were addressed in the decision: firstly, the extent to which Shazam data would serve as a tool for Apple to attract additional customers to its streaming business. In this regard, the EC held that the music streaming marketplace is rapidly expanding and that direct acquisition of customers should be the focus.21 Secondly, the EC assessed how unique the data held by Shazam was in respect of user preferences. Whilst the data was clearly not without commercial value and utility, the competition authority concluded that acquisition of this type of data is not difficult and therefore it does not represent a major competitive advantage.22 Perhaps most importantly this case demonstrates how the EC has matured in its critical assessment of data advantage and, in so doing, refined its definition of precisely what “big data” might mean as a concept. In a framework referred to as the “four Vs”, four parameters are used to analyse the commercial and competitive edge derived from different types of data sets: variety, velocity, volume and value.23 Applying this framework to Apple/Shazam it is clear from the competition authority’s analysis that the data held by Shazam was not sufficiently various to afford Apple an anti- competitive level of data advantage; equally, the rate at which Shazam collected data was deemed to be comparatively low.24 With regard to volume, it was found that Shazam’s data set was far smaller than many other sets that are available to music streaming apps and, crucially, the data held by Shazam was deemed to be of low value given its lack of uniqueness.25
Looking ahead, the recent announcement by Google relating to its acquisition of FitBit is noteworthy as it provides the EC with a timely platform for clarifying and applying its approach to data rich acquisitions under the evolving merger control framework.26 Indeed, the EC is due to deliver its decision on the merger on 4th August 2020, extended from 20th July 2020, following submission of a proposal by Google that aims to avoid a full-scale probe into the matter.
Theories of Harm
One of the greatest challenges facing the current competition law regime relates to theories of harm. The EC’s 2019 report highlights how difficulties arise when applying traditional theories of harm relating to innovation in the context of the digital economy, and this is fundamentally because these theories were developed in response to innovation in the agrochemical and pharmaceutical sectors, both of which are markedly different to digital innovation.27 Whereas innovation in these well-established sectors tends to be long term, asset driven and based around Intellectual Property rights, digital innovation would appear to be quite the opposite.28 Accordingly, the litmus test for efficacious competition policy has shifted and the competition authority should be judged on the basis of whether it is capable of preventing barriers and delays to rapid innovation, the foreclosure of innovative businesses and, critically, understanding where boundaries exist in order to properly define the appropriate competitive marketplace in any given case.29 Moreover, it is apparent that theories of harm specifically within the context of merger control will need to be reconsidered in light of “killer acquisitions.”30 In the report it is proposed that the most efficient means by which to deal with such anti-competitive acquisitions is not simply to analyse the impact of access to data as an input within the context of rivalrous incumbents, but to apply some aspects of horizontal analysis to the conglomerate theories of harm.31
In order to regulate the Internet of 2030, and thereby address the challenges that the rise of digital technologies and the emergence of new business models pose, the EC should be careful not to enact changes that destabilise or undermine the foundations of EU competition law. Adopting a broad view, it would appear that the EC is open minded to reform, whilst striving to strike a careful balance between intervention and caution. However, is must not be overlooked that the EC’s 2019 report contains a suggestion that due to the highly concentrated nature of markets, which are characterised by strong network effects and high barriers to entry, the competition authority should consider a reversal of the burden of proof on the incumbent to show that conduct is pro- competitive.32 In fact, the report points to cases where this reversal should be applied “even where consumer harm cannot be precisely measured,” positing that “strategies employed by dominant platforms aimed at reducing the competitive pressure they face should be forbidden in the absence of clearly documented consumer welfare gains.”33 Whilst clearly progressive in many ways, and borne out of a legitimate concern over the “error cost” of enforcement in digital competition law cases, this may be a step in a very dangerous direction. Cases heard by the courts concerning abuse of dominance under Article 102 TFEU have historically been held as “quasi-criminal”, and as such a shift in the burden of proof to the defendant would appear to run counter to the presumption of innocence.34 By way of conclusion, this essay recommends that in the years ahead the EC heed the words of Johannes Laitenberger: “the fundamental EU competition rules as embodied in the treaties have shown over their first 60 years of existence that they can capture changing realities and new phenomena. The treaty rules have been crafted to capture a film, not just a photograph.”35
Benjamin Evans is reading for an LLM in IT & IP Law at the University of East Anglia, and is currently writing a Dissertation that considers the merits of creating a new property right in Data. He is interested in the intersection between IT, IP and Competition Law and his academic work draws upon his experience in the management and commercialisation of IP.
Notes & sources
1 Case AT.40411 Google Search (AdSense) ; Case AT.40099 Google Android ; Case AT.39740 Google Search (Shopping) .
2 Case AT.39740 Google Search (Shopping) .
3 EC Press Release, 27 June 2017 https://ec.europa.eu/commission/presscorner/detail/en/IP_17_1784.
4 H. Hobbelen and others, ‘Report on the Status of the Regulation of the Digital Economy in the EU’ (2019), 16
5 Case AT.40411 Google Search (AdSense) ; EC Press Release, 20 March 2019 .
6 EC Press Release, 20 March 2019 .
7 J. Crémer and others, ‘Competition policy for the digital era’ (2019), 5, 36
8 ibid, 5 - 6.
9 ibid, 5, 55.
10 ibid, 5.
11 ibid, 6.
12 ibid, 6.
13 ibid, 6.
14 M. Vestager, ‘Competition and the digital economy’ (2019) .
15 J. Crémer and others, ‘Competition policy for the digital era’ (2019), 7.
16 H. Hobbelen and others, ‘Report on the Status of the Regulation of the Digital Economy in the EU’ (2019), 15.
17 Case AT.39740 Google Search (Shopping) ;
18 H. Hobbelen and others, ‘Report on the Status of the Regulation of the Digital Economy in the EU’ (2019), 15.
19 Case M.7217 Facebook/WhatsApp ; E. Ocello and others, EC Competition Merger Brief, ‘What's Up with Merger Control in the Digital Sector?’, February 2015, 4 - 5 .
20 Case M.8788 Apple/Shazam .
21 H. Hobbelen and others, ‘Report on the Status of the Regulation of the Digital Economy in the EU’ (2019), 11.
22 ibid, 12.
23 G. Murray, ‘Apple/Shazam: 5 observations on the data-related aspects’ (2018) .
26 A. Rosen, ‘Large tech merger provides an early opportunity to test the fitness of the competition regime in 2020’ (2019) .
27 H. Hobbelen and others, ‘Report on the Status of the Regulation of the Digital Economy in the EU’ (2019), 12.
28 ibid, 12.
29 H. Hobbelen and others, ‘Report on the Status of the Regulation of the Digital Economy in the EU’ (2019), 12.
30 ibid, 9.
31 J. Crémer and others, ‘Competition policy for the digital era’ (2019), 11.
32 ibid, 4.
33 ibid, 3.
34 H. Hobbelen and others, ‘Report on the Status of the Regulation of the Digital Economy in the EU’ (2019), 15.
35 J. Laitenberger, Speech at “Shaping competition policy in the era of digitisation” Conference (2019) .