ISSN 1745-638X (Online)

THE COMPETITION LAW REVIEW

Volume 11 Issue 1 pp 109-131 July 2015

Google’s Anti-Competitive and Unfair Practices in Digital Leisure Markets
Anca D Chirita*

The purpose of this article is to reflect on the critical use of commitments in the Google case
and to analyse and review the matrix of facts that have been highlighted in the academic and
practitioner literature. Therefore, the core areas of reflection in this contribution are: relevant
markets; barriers to entry; network and lock-in effects; dominance; and, potential anti-
competitive, as well as unfair practices as regards commercial advertisements. The analysis of
the online search-engine market is complemented by the comparative insights offered by the US
class action against Google’s Android mobile applications. In the EU, a similar trend is
noticeable in the complaining tone of Google’s competitors. When this is coupled with the
transitional period of the mandate of the newly appointed Commissioner for Competition and
the political sensitivity over the potential to misuse search-engine users’ personal data to serve
commercial purposes, such as boosting its advertising revenues, the giant Google swims in
uncertain waters.

I. THE MAIN CONTROVERSIAL ISSUES OF THE GOOGLE CASE

One of the most controversially debated investigations, 1 by practitioners, academics,
and even politicians, 2 is that of Google: a saga of long negotiated and re-negotiated
commitments.3 This saga has been paved with public criticism for the lack of any

* Lecturer in competition law, Durham Law School, England, UK. This paper was presented on 26 September
2014 at the CLaSF Workshop at the IE Law School in Madrid, Spain. I would like to thank the organisers
and participants of the wonderful Competition Law Scholars Forum XXIII Workshop ‘Competition Law in
Leisure Markets’ for many insightful suggestions, comments and questions, in particular, Angus McCulloch,
Alan Riley, Barry Rodger, Francisco Marcos, Ben van Rompuy, Fernando Diez, Cristina Volpin and
Christian Hocepied.
1 The EC started its investigation against Google almost four years ago, see EC, ‘Antitrust: Commission
probes allegations of antitrust violations by Google’, IP/10/1624, 30 November 2010; Cases COMP/C-
3/39740, Foundem v Google; COMP/C-3/39775, Ciao v Google, and COMP/C-3/39768, 1PlusC v Google Inc.
(Google).
2 Apart from the EU Commissioner for Competition, J Almunia, see e.g., the German Minister of Justice, H
Maas, ‘Das letzte Mittel ist die Entflechtung von Google’, Frankfurter Allgemeine Zeitung, 27 June 2014,
who considered a possible divesture of Google as a remedy of last resort; cf. that a lack of internet regulation
does not allow competition authorities to divest Google, see, very recently the President of the German
competition authority, A Mundt, ‘Kartellamt lehnt Zerschlagung von Google ab’, WallStreet: online, 8
October 2014; for an economist’s similar view, see e.g. J Haucap, ‘Eine Zerschlagung von Google würde
wenig bringen’, Frankfurter Allgemeine Zeitung, 26 April 2014; for the political opinion that divesting
Google is impossible because Google is established elsewhere in the US, see e.g. the German Minister for
Internal Affairs, Thomas de Maizière, who is in sharp contrast to the socialist Ministry of Economics, Sigmar
Gabriel, The Wall Street Journal Deutschland, ‘Gabriels großer Google-Bluff’, 16 May 2014; Reuters,
‘German vice chancellor urges stricter rules for Google and peers’, Berlin 14 October 2014, Fortunately, UK
ministers do not seem to care about Google’s fate. Otherwise, DG COMP would soon start pooling political
opinions.
3 Generally on commitments procedure, see e.g. EC, ‘To Commit or not to Commit?’ 3 Comp Policy brief
(2014); on the EC’s wide margin of discretion after Alrosa, see e.g. M Messina and JC Alexandre Ho, ‘Re-
Google’s Anti-Competitive and Unfair Practices

antitrust fine being imposed on Google, for its allegedly anti-competitive business
conduct, with academic criticism for a continued lack of new ‘precedents’ under Article
102 TFEU, and with an uncertain political climate inspired by the announcement of the
transition to another mandate of the new Commissioner for Competition. 4 What, then,
will happen to Google? Recent political statements, such as that by Commissioner
Almunia, have suggested that Google could even end up involved in a case bigger than
that of Microsoft. 5
Arguments for and against intervention have already divided the academic arena with
robust and pertinent argumentation on both sides. 6 The previous approach by the US
Federal Trade Commission (FTC) 7 demonstrates that choosing not to pursue a case
against Google 8 is equally litigious, as consumer associations reacted with a class action
introduced before the US District Court of California. 9 This is a very welcome
development, and one that is inspiring a comparative precedent for the European
Commission (EC). In contrast, the EC’s decision to agree on negotiated
commitments,10 instead of fining Google, is by no means a better deal. On the one

establishing the Orthodoxy of Commitment Decisions under Article 9 of Regulation 1/2003: Comment on
Commission v Alrosa’ 36 Eur L Rev (2011), 747; F Wagner-von Papp, ‘Best and Even Better Practices in
Commitment Procedures after Alrosa: The Dangers of Abandoning the ‘Struggle for Competition Law’’, 49
Common Market L Rev (2012), 941; on the ‘policitization’ of commitments, see e.g., N Dunne, ‘Commitment
Decisions in EU Competition Law’ 10 J of Comp L & Ec 2 (2014), 440.
4 See the incoming Commissioner, Margrethe Vestager, on the use of settlements that ‘enable cases to be
closed rather than to be dragged over years and years’, in Reuters, FY Chee and A Macdonald, ‘New EU
antitrust head not swayed by anti-Americanism bullies’, 23 September 2014. See also Euractiv, ‘Vestager
dodges questions on Google probe, Irish tax loophole’, 3 October 2014.
5 Reuters, ‘EU’s Almunia says may probe Google’s non-search services’, 23 September 2014.
6 See as dismissive of Google’s alleged biased search results, in particular, RH Bork and JG Sidak, ‘ What Does
the Chicago School Teach about Internet Search and the Antitrust Treatment of Google’ 8 J of Comp L & Ec
663 (2012); GA Manne and JD Wright, ‘Google and the Limits of Antitrust: The Case Against the Antitrust
Case Against Google’ Harvard J of L & Public Policy 34 (2011) 1; in favour of intervention, see e.g. I Lianos
and E Motchenkova, ‘Market Dominance and Search Quality in the Search Engine Market’ J of Comp L & Ec
(2013); MA Carrier, ‘Google and Antitrust: Five Approaches to an Evolving Issue’ Harvard J of L &
Technology Occasional Paper Series, July 2013; MR Patterson, ‘Google and Search-Engine Market Power’
Harvard J of L & Technology Occasional Paper Series, July 2013; R Burguet, R Caminal and M Ellman, ‘In
Google We Trust?’ Barcelona GSE Working Paper Series no 717 (2013); N Zingales, ‘Product Market Definition
in Online Search and Advertising’ 9 Comp L Rev 1 (2013).
7 The FTC acknowledged that alterations by Google of its algorithm deprived competing advertisement sites
of traffic; see e.g., AA Foer and S Vaheesan, ‘Google: the unique case of monopolistic search engine’, 24
June 2013, available at http://www.blog.oup.com/2013/06/google-monopoly-search/.
8 Two commentators attributed non-intervention to the influential views of Bork and Sidak, ‘What Does the
Chicago School Teach about Internet Search and the Antitrust Treatment of Google?’ 8 J of Comp L & Ec
(2012) 663, see e.g. F Pasquale and S Vaidhyanathan, ‘Borking Antitrust: Google Secures Its Monopoly
Dissent’, Dissent, 4 January 2013, available on http://www.dissentmagazine.org/blog/borking-antitrust-
google-secures-its-monopoly .
9 US Class Action Complaint 010437-11 683086 VI, para 74. Previously, the District Court of California ruled
that there is no separate online market for search queries; see District Court of the Northern District of
California, Kinderstart.com, LLC v Googl Inc., C 06-2057 JF, 2007 WL 831806, 16 March 2007.
10 EC, ‘Antitrust: Commission obtains from Google comparable display of specialised search rivals’,
MEMO/14/87, 5 February 2014; EC, Commitments in Case COMP/C-3/39740, 3 April 20133; A
Lamadrid de Pablo, ‘EU Google Commitments’, J of Eur Comp L & Practice (2013); A Leyden and M
Dolmans, ‘The Google Commitments: Now with a Cherry on Top’ J of Eur Comp L & Practice (2014).

110 (2015) 11(1) CompLRev

 A Anca D Chirita

hand, this sends across the Atlantic a positive signal of convergence with the FTC’s
philosophy of not protecting competitors 11 under the Sherman Act, which twists the
EC’s approach to one of protecting competition irrespective of what competitors or
intermediary consumers say about their rivals. However, this is a critical use of
commitment decisions that were intended to be used only as an exceptional remedy. 12
The previous EU Commissioner for Competition acknowledged at his hearing before
the European Parliament that ‘antitrust enforcement is about consumer welfare,
innovation and choice, not about protecting competitors’. 13 On the other hand, despite
not yet having the privilege of a similar consumers’ class action, the lack of any clear
reaction on the part of the EC against Google can only upset the legitimate
expectations of Google’s competitors; including Microsoft, Expedia, Yelp,
TripAdvisor, 14 the European Consumer Organisation, 15 European publishers, an
association of picture industries and photo libraries, a telecom operator, and an
advertising platform. 16 The huge number of competitors with complaints against
Google shows that this case is very different from the situation faced by Microsoft.
Recent EU news demonstrates that the Google saga is far from over, 17 with suggestions
of possible investigations, inter alia, into mobile applications and social networks. 18
There is an emerging trend of expressed dissatisfaction with the outcome reached by
competition enforcers. This comes from both intermediary and final users of Google’s
search engine and additional tied services.
Finally, in recent years, the EC has all too often 19 used commitments in previous cases,
such as Rambus,20 Microsoft, 21 IBM, 22Apple E-books, 23 Samsung, 24 and, most notably, in

11 See e.g. JD Ratliff and DL Rubinfeld, ‘Is There a Market for Search Engine Results and Can Their
Manipulation Give Rise to Antitrust Liability’ J of Comp L & Ec (2014), 4, 13.
12 On the use of commitments as ‘unusual and rare’, see e.g. J Temple Lang, ‘Commitment Decisions and
Settlements with Antitrust Authorities and Private Parties under European Antitrust Law’ in BE Hauwk (ed),
International Antitrust Law and Policy (Fordham Corporate Law Institute, 2005) 265.
13 See EC, J Almunia, ‘The Google antitrust case: what is at stake?’, European Parliament hearing, Brussels, 1
October 2013, Speech/13/768.
14 Financial Times, ‘EU: Yelp, TripAdvisor launch new anti-Google campaign’, 30 September 2014.
15 See a recent complaint by Hot Maps on the EC’s monitoring of commitments undertaking by Google, e.g. K
Fiveash, ‘Google will ‘pre-select’ an ‘independent’ competition inspector in EU search case’, The Register, 18
June 2014; that the EC receive ‘very, very negative’ feedback from Microsoft and Expedia, see e.g. A White
and F Rotondi, ‘Google Concessions Sought by EU to Rescue Antitrust Pact’, Bloomberg Businessweek, 8
September 2014; D Meyer, ‘Yelp piles into Google EU antitrust case with formal complaint about local
search’, 9 July 2014, gigaom.com.
16 See Reuters, n 5.
17 See e.g. ‘EU: Google rethinks concessions as Commission’s case continues’, 22 September 2014, Competition
Policy International.
18 J Kanter, ‘Google’s European antitrust woes are far from over’ Economic Times, 23 June 2014. For a similar
approach see e.g. the U.S. Federal Trade Commission, press release, ‘Google Agrees to Change its Business
Practices to Resolve FTC Competition Concerns in the Markets for Devices Like Smart Phones, Games and
Tablets, and in Online Search’, 3 January 2013.
19 There have been at least eighteen commitment decisions during the last ten years and, more significantly,
almost 40% of these commitments fall under the scope of Article 102 TFEU.
20 EC, decision, Rambus, OJ C30/2010; Case COMP/38636, Rambus, 9 December 2009, OJ C30/2009.

(2015) 11(1) CompLRev 111

Google’s Anti-Competitive and Unfair Practices

the energy sector, 25 which has caused alarm bells to ring. One has to question why the
decision to commit is better in highly litigious cases of unfair competition against rivals,
but not in other cases, such as Google’s, where the decline in the quality of the search
engine affects all of its users, not only competitors’ inability to match the efficacy of
this innovative search-engine. In certain instances, negotiated commitments can only
restore competition for the future, as is the case, for example, with the commitment to
refrain from entering into de facto exclusivity agreements. 26 Thus, the above
commitment cannot undo the anti-competitive harm already being caused and
subsequently, no civil claims can be made on the same grounds. The better alternative
remains the imposition of a fine. In particular, there is a persistent perception of
preferential ranking of Google’s own services vis-à-vis those of its rivals. 27 While a
separate display of ‘generic’ and ‘specialised’, i.e., advertised, search results will no
doubt improve the user experience, the additional requirement of listing three vertical
links to rivals’ specialised searches partially addresses the competitors’ rather than the
generic users’ concerns. The EC continues its investigation concerning Google’s alleged
abuse of dominance in relation to Android.

II. GETTING THE FACTS RIGHT ON GOOGLE’S BUSINESS MODEL

This is probably the most daunting task to be faced; no one knows all the facts, yet
everyone seems to have an opinion on Google’s fate. Therefore, before getting into the
matrix of facts, one has first to understand Google’s business model.
A. A Few Contentious Issues: Relevant Markets, Network Effects, Barriers To
Entry And The Lock-In Effect
Initially, Google’s founders were not sympathetic to an advertisement-funded, i.e.,
biased, business model for Google’s search engine. 28 As licensing the search engine was
not a viable option, the founders had to compromise; it was then that they started
placing sponsored ‘links’. In the primary relevant market, Google offers its users, that
is, both intermediary and final consumers, a free horizontal search engine 29 that
provides general information available on its online platform. 30 In the secondary

21 See EC, press release, IP/09/1941.

22 EC, decision, IBM, OJ C18/2012; EC, press release, IP/11/1539.
23 EC, decision, E-Books, OJ C 73/2013; Case COMP/39847 – E-Books (Penguin), C 4750 final, 25 July 2013;
EC, press release, IP/12/1367.
24 Case COMP/39939 –Samsung/Enforcement of UMTS standards essential patents.
25 Around nine commitment decisions applicable in the electricity and gas sectors.
26 See Commitments in Case COMP/C-3/39740 – Foundem and others, 3 April 2013.
27 iComp, Press Publishers’ View in Google’s Second Set of Commitment Proposals, 2013, available at
www.icomp.org.
28 On the history of Google’s search engine, see e.g., AD Vanberg, ‘From Archie to Google – Search engine
providers and emergent challenges in relation to EU competition law’ 3 Eur J of L & Technology 1 (2013), 3.
29 For the distinction between horizontal and vertical search engine markets, see e.g., A Langford, ‘gMonopoly:
Does Search Bias Warrant Antitrust or Regulatory Intervention?’ 88 Indiana L J, 1564.
30 See e.g. DF Spulber, ‘The Map of Commerce: Internet Search, Competition, and the Circular Flow of
Information’ 5 J of Comp L & Ec 4 (2009), 635; generally on search engines, see e.g., A Milstein, ‘Search
112 (2015) 11(1) CompLRev
 A Anca D Chirita

relevant market, Google charges online traders for displaying ‘AdWords’ in its vertical
search engine, which ranks differently price comparisons for shopping or travelling.31
Google’s business model is simple: a commercial tie-in agreement serving the marketing
purposes of online traders subsidises the internet experience of its users surfing
through its engine for online content. Thus, not all content is organically generated by
Google’s engine. Similar to unwanted and very annoying TV or newspaper advertising,
a multitude of keywords, which are relevant to both horizontal searches and vertical
advertisements, are inserted into Google’s engine and then displayed to the user for
free, while an intermediate beneficiary of the keywords ‘Ad’ pays Google revenues each
time a user clicks on it. In this way, Google can offer horizontal searches sponsored by
its advertising revenues. This ‘magic’ circle is Google-centric: the more content is
indexed in the search engine, the more users it attracts, so the likelihood of clicking on
commercial ads increases and, consequently, Google’s revenues increase, too. One has
to overcome the bone of contention on the dual purpose of high-technology markets
such as Google’s. 32 They are economically two-sided markets 33 that generate mutual
benefits. For example, the search engine ‘network’ produces direct benefits to its users
as more users discover online content via horizontal searches and indirect benefits
through the advertising rents charged by Google to intermediate users of its vertically
integrated online market place, 34 where bidders meet auctioneers of commercial Ads. A
word of caution is needed on the potentially knock-out effect that vertical commercial
ads can have when interfering with generated horizontal search results: final users, such
as TV viewers, generally dislike advertising. Therefore, the more Ads that are shown,
the more likely it is that users will eventually switch to alternative search engines that
offer less or no advertising. Nonetheless, maintaining a non-irritating degree of

Engine Bias as Rechtsproblem’ 11 Computer und Recht (2013); J Kühling and N Gauß, ‘Suchmaschinen-eine
Gefahr für den Informationszugang und die Informationsvielfalt’ 51 Zeitschrift für Urheber und Medienrecht 12
(2007); T Höppner, ‘Das Verhältnis von Suchmachinen zu Inhalteanbitern an der Schnittstelle von Urheber-
und Kartellrecht’ 6 Wettbewerb in Recht und Praxis (2012); J Haucap and C Kehder, ‘Suchmaschinen zwischen
Wettbewerb und Monopol: Der Fall Google’, 44 DICE Ordnungspolitische Perspektiven (2013).
31 For the contrary opinion that general and specialised searches are two separate markets, see e.g., Bork and
Sidak, n 6.
32 Although implausible, the contrary opinion goes on to contradict that Google is a two-sided market as
operating the two transactions is a ‘business strategy, not a structural feature of the market’; see e.g. G
Luchetta, ‘Is Google Platform a Two-Sided Market’ J of Comp L & Ec (2013), 9. In contrast, Patterson argued
that as price, quality, and output are inter-related, one cannot consider Google’s search and advertisement
markets in isolation; see e.g. Patterson, n 6, 16. On the economics of the two-sided search engine market
model, see e.g. Lianos and Motchenkova, n 6, 27.
33 See e.g. E Engelhardt, A Freytag and V Köllman, ‘Competition policy and vertical integration in internet-
based two-sided markets: the Google case’ MPRA Paper no 43326 Munich (2012); N Zingales, cited above; F
Thépot, ‘Market Power in Online Search and Social Networking: A Matter of Two-Sided Markets’ 36 World
Comp L & Ec Rev 2 (2013), 195; M Gal, ‘Antitrust in High-Technology Industries: A Symposium
Introduction’ 8 J of Comp L & Economics 3 (2012); M Armstrong, ‘Competition in Two-Sided Markets’, 37
RAND J of Ec 3 (2006), 668; L Filistrucchi, D Geradin and E van Damme, ‘Identifying Two-Sided Markets’
36 World Comp L & Ec Rev (2013), 33; M Armstrong and J Wright, ‘ Two-Sided markets, Competitive
Bottlenecks and Exclusive Contracts’ 32 Economic Theory 2 (2007) 353; on the lack of empirical analysis, see
e.g. A Goldfarb and C Tucker, ‘Substitution Between Offline and Online Advertising Markets’ 7 J of Comp L
& Ec (2011), 37.
34 For insightful suggestions on ‘search’ and ‘display’ advertising, see Burguet et al, n 6.

(2015) 11(1) CompLRev 113

Google’s Anti-Competitive and Unfair Practices

commercial advertising cannot work against Google. This raises one particular concern
since even if switching costs are zero or very low, 35 switching or ‘multi-homing’ 36 is less
likely to occur 37 and, therefore, Google’s strategic business model locks-in all existing
users, without the latter’s discovery of another rival search engine. Competition is ‘a
process of discovery’ but, similar to being in a locked-in job, users often do not venture
to discover new avenues. Therefore, the evaluation that Google competes on the basis
of the merits of its search engine alone is wrong given the extent to which its users do
not use any alternative engines. Furthermore, the efficacy of its horizontal engine could
be attributed to its internal algorithmic metrics, which insert as many commercial ads as
is subjectively and psychologically acceptable for TV viewers not to switch to another
channel. 38 In fact, Google’s algorithmic metric adjusts the ads and costs to the
advertisers based on the relevance of the link to the search query and the quality of the
web pages. 39 In other words, advertisers pay Google if users click on the ads, which
could turn into another profitable business if Google were to employ people to click on
such ads. Leaving aside the maverick that follows from an internal manipulation by
Google of its listings of ads based on auctioneered ranking, behavioural economics40
could offer some further explanation of the decline of quality 41 due to the insertion of
more ads. Another assumption is that a better search quality actually decreases the
likelihood that users will click on sponsored ads. 42 Therefore, from this perspective,
Google would be better off not improving the quality of its organic searches.
These telling facts should never encourage an optimistic expectation that the quasi-
monopolist should first do something wrong to lose clientele, in other words, sink its
own boat, 43 before competition authorities do something to prevent the anti-
competitive lock-in effects on Google’s users and its rivals being kept out of the
market. Unfortunately, this is not all. Google offers Google Places for hotels,

35 See e.g. Bork and Sidak, n 6; and, R Pollock, ‘Is Google the Next Microsoft? Competition, Welfare and
Regulation in Internet Search’ Cambridge Working Paper (2009), 26.
36 For the contrary opinion, see e.g. DA Crane, ‘Search Neutrality and Referral Dominance’ J of Comp L & Ec
(2012), 5.
37 According to Performics, 89% of users would switch to a different search engine if they could not find the
information they were looking for; see e.g. Performics, press release, ‘ Search Engine Usage Study: 92
Percent of Searchers Click on Sponsored Results’, 28 September 2010, available at
http://www.performics.com/news-room/press-release/Search-Engine-Usage-Study-92-Percent/1422.
38 Similarly, in mergers, the EC, COMP/M5727 – Microsoft/Yahoo! Search, 18 February 2010, para 144-69, where
the EC assumed that consumers could detect the degradation in quality, see e.g. A Ezrachi and ME Stucke,
‘The Curious Case of Competition and Quality’ SSRN Working Paper (2014), 8.
39 See e.g. GA Manne and JD Wright, ‘Google and the Limits of Antitrust: The Case Against the Antitrust Case
Against Google’ 34 Harvard J of L & Public Policy (2011) 1, 23.
40 Critically, an intervention to ‘regulate’ Google’s search algorithm could be ‘unconstitutional’; see e.g. A
Candeub, ‘Behavioural Economics, Internet Search, and Antitrust’ 9 J of L & Policy 3 (2014).
41 It is difficult for users to assess the quality of the search results they receive; see e.g., MR Patterson, ‘ Google
and Search-Engine Market Power’, Harvard J of L & Technology (2013) 8.
42 See e.g. Pollock, n 35, 36.
43 See e.g. Bork & Sidak’s, n 6, scenario where Google’s overinvestment in advertisement will trigger a
subsequent decline in users’ experience.

114 (2015) 11(1) CompLRev

 A Anca D Chirita

restaurants and travel destinations, Google Travel for flights, 44 Google Product Search
for product information and price comparisons, Google Maps for location and
direction information, Google Chrome for browsing, Google Translate for translations,
Google Scholar for citations of scholarly articles and books and, specifically for
entertainment, YouTube for video content. 45 Similarly, Yahoo offers various categories,
such as services, finance, real estate, jobs, movies, music, sports, personal, and yellow
pages, 46 while Amazon offers books, movies, music, games, Kindles and so on.
One can also consider that having indexed a larger amount of information offers
Google the competitive advantage of a larger search engine 47 that can produce better
results. Any new entry into the market will incur significant investments in building its
own search-engine platform, i.e., crawling pages, indexing them, and processing queries,
and additional sunk costs to make the online platform known to users. A similar view
suggested in the economic literature48 is that the effect of Google’s monopoly in the
search engine market is to demand exponentially higher investments, i.e., sunk costs, on
the part of its incumbents to rival the performance of its search engine. These costs
represent actual barriers to market entry.
At this point, one cannot say that there are insurmountable barriers to market entry,49
since there are quite a few competitors, albeit with insignificant shares of the market.
The problem is that the competitive pressure on Google is either ineffective or non-
existent. It could therefore be questioned whether effective competition is the result of
massive investments in innovation, thereby improving Google’s search engine. If these
investments were gained from the other side of its business, namely, sponsored ads,
then this is something that Google’s competitors will have to adapt to as a business
model. Applying the SSNIP (Small but Significant and Non-Transitory Increase in
Price) test is not really helpful here; in the horizontal market, users are not charged at
all, while in the vertical market, no uniform pricing is detectable.
In conclusion, a zero-priced search-engine is a mathematical delusion when
approaching a near monopoly position with ‘inefficiently low search quality’. 50 The real
price to pay will be the decline in quality as a result of more or aggressive advertising.

44 For concerns about Google favouring its own travel products, see e.g., M Williams, ‘Expedia is Worried
About Google/ITA Deal’, Inside Google, 12 July 2010, available at
http://www.insidegoogle.com/2010/07/expedia-is-worried-about-googleita-deal/.
45 See e.g. Ratliff and Rubinfeld, n 11.
46 Spulber, n 30, 641.
47 See generally B Pal, ‘Immaterialgüter, Internetmonopole und Kartellrecht’ GRUR-Beilage (2014), 69; M Rato
and N Petit, ‘Abuse of Dominance in Technology-Enabled Markets: Established Standards Reconsidered’ 9
Eur Comp J 1 (2013); J Verhaert, ‘The Challenges involved with the application of article 102 TFEU to the
new economy: a case study of Google’ 35 Eur Comp L Rev 6 (2014).
48 See e.g. Pollock, n 35, 28; Langford, n 29, 1575.
49 In contrast, see e.g. I Lianos and E Motchenkova, ‘Market Dominance and Search Quality in the Search
Engine Market’ J of Comp L & Ec (2013), 10, where it is assumed that network effects and the fixed costs
related to R&D or the development and maintenance of service infrastructure are barriers to entry into the
search-engine market.
50 See e.g. Pollock, n 35, 41.

(2015) 11(1) CompLRev 115

Google’s Anti-Competitive and Unfair Practices

For rivals to catch up with Google, there are certain barriers associated with market
entry or with Google’s competitive pressure, in particular, building an immense data
infrastructure, which could amount to significant fixed costs, 51 and developing both a
search and ranking algorithm. The latter is essentially ‘learning-by-doing’ and, therefore,
could take years to develop. However, given that the information is a trade secret, a
possible disclosure of Google’s ranking algorithm could give rise to similar criticism as
in the Microsoft case. Therefore, so far, the monitoring of Google’s behaviour has been
the better option adopted by the EC.
Figure 1

Dominance in
More Users More Revenue Vertical Listing of
Ads

Horizontal Search
More Content More Ads
Dominance

B. Dominance
Dominating both primary and secondary markets is crucial to maintaining the clientele
from both sides of the market, namely, online users who navigate Google’s search
engine and traders who pay to have commercial ads inserted vertically into it.
In 2010, the OECD suggested that the global search engine market is highly
concentrated in the hands of only five major companies, 52 who account for over 90%
of the market. 53 It is no secret that Google dominates the European online search-
engine market with nearly 90% of the market share. 54 In the US, recent data acclaims
Google as an incontestable leader of general searches with 81.87% as of March 2014.55
This ever-growing monopoly has been assessed against the Microsoft/Yahoo! merger56
where the EC considered it unlikely that the parties could degrade the quality of the

51 See e.g. Pollock, n 35, 39.

52 According to comScore data in 2009, the five leaders were Google with a 64% market share, Yahoo with
14%, the Chinese Baidu with 13%, Microsoft with 4%, and the South Korean Naver with a 2% market
share.
53 OECD, Report on ‘The Economic and Social Role of Internet Intermediaries’, DSTI/ICCP (2009)
9/FINAL, April 2010, 28, available at http://www.oecd.org/dataoecd/49/4/44949023.pdf. Recently, the
same figure of 90% is being maintained in the EU; see e.g. M Scott, ‘E.U. Official Urges Google to Offer
Greater Concessions in Antitrust Inquiry’, New York Times, 20 January 2015.
54 See e.g. C Argeton and J Prüfer, ‘Search Engine Competition with Network Externalities’ 8 J of Comp L &
Ec 1 (2012).
55 Statcounter, available at http://www.gs.statcounter.com#desktop+mobile+tablet-search_engine-US-
monthly-201401-201403.
56 Microsoft/Yahoo!, n 38.

116 (2015) 11(1) CompLRev

 A Anca D Chirita

search engine results due to Google’s presence. When it entered the market in 1998,
Google was competing only with Microsoft’s MSN (later Live Search), and thereafter,
with many others such as AltaVista, Yahoo, Infoseek, Lykos, 57 or GoTo (later re-
named Overture). The latter introduced the first auctions of keywords for the listing of
ads, followed by Google’s AdWords and its settlement with Overture in the patent
lawsuit. 58
Google Google Ads Google Google per Google
Online revenue usage country
vertical integration
search-
engine
market
90% EU $36.5 bn 81.57% (2009) 95% GoogleMaps +50%
(2012) (2011) Germany
MapQuest -20%
>66% US
$15-18 bn
(2012)
US Ads
70% US
(2010)
81.87% Facebook Yahoo YouTube >80%
10.07%
US (2014) $3.2 billion Photobucket <3%
MSN 2.97%
Google Images 50%
AOL
Yahoo Images >7%
MLS
Ask
AltaVista
all <1%

Based on the revenues gained from advertisements, anecdotal evidence from 2011
suggests Google leads the market with $36.5 billion compared to its competitor,
Facebook, with only $3.2 billion. Recent US estimates indicate that Google’s US
AdWords revenues are ‘somewhere in the range of $15 to 18 billion annually’.59
Facebook remains a potential contender in the advertisement market for entertainment.
The same can be said about the strategic alliance in the Bing/Yelp deal, following

57 See e.g. S Lawrence and CL Giles, ‘Accessibility of Information on the Web’, 400 Nature 107 (1999).
58 Spulber, n 30, 645.
59 See e.g., http://arstechnica.com/tech-policy/2014/01/court-orders-google-to-pay-1-36-of-adwords-revenue
-for-infringing-patents/.
(2015) 11(1) CompLRev 117
Google’s Anti-Competitive and Unfair Practices

which reviews of Yelp restaurants could be featured prominently in Bing’s search

engine. 60
Based on online usage, in 2009, Google was the market leader with 81.57% of the
market share, followed by Yahoo with 10.07%, MSN with 2.97% and other
competitors, including AOL, MLS, Ask, and AltaVista, all with less than 1% of the
market share. 61 However, in 2012, Google held over 95% of the market share in
Germany and over 66% in the US. 62 Another telling fact is that since its free vertical
integration in Google’s search-engine, Google Maps has gained 50% of the market
share, while its rival, MapQuest, has lost 20% of its market share. 63 The same applies to
Google-owned YouTube, which has doubled its market share of the video market to
over 80%, and is now followed by Photobucket with less than 3% and MySpace with
less than 1% in the US. 64 Google Images enjoys 50% of the market share compared to
Yahoo Images with over 7% and other competitors with less than 10% of the market
share. 65 In the search advertising market, Google was clearly dominant with over 70%
of the US market in 2010. 66
The period from 1998 to 2014, throughout which Google sought to consolidate its
prominent and thereafter dominant position in the online search-engine market, cannot
be considered as merely temporary, and the resultant monopoly should not be under-
estimated as ‘a little monopoly’. 67 Things may be different in the market for digital
devices, but solely on the basis of the period in question. However, the intention to
maintain Google’s prominent position in the online search-engine market prevails over
purely innovative pursuits of a seemingly unfortunate number of thirteen Google
applications, namely, Google Chrome, Google Maps, Google Drive, YouTube, Gmail,
Google+, Google Play Music, Google Play Movies, Google Play Books, Google Play
Newsstand, Google Play Games, Google+ Photos, and Google+ Hangouts.

60 See e.g. M Lao, ‘Search, Essential Facilities, and the Antitrust Duty to Deal’ 11 Northwestern J of Technology &
Intellectual Property 5 (2013), 300.
61 Spulber, n 30, 646.
62 See Comscore, press release, ‘comScore Releases March 2012 U.S. Search Engine Rankings’, available at
http://www.comscore.com/Press_Events/Press_Releases/2012/4/comScore_Releases_March_2012_U.S._
Search_Engine_Rankings.
63 See e.g. T Körber, ‘Google im Fokus des Kartellrechts’ 7 Wettbewerb in Recht und Praxis (2012), 768; Vanberg,
cited above, 6.
64 Consumer Watchdog’s Inside Google, ‘Traffic Report: How Google is Squeezing Out Competitors and
Muscling Into New Markets’, 2 June 2010, available on http://www.consumerwatchdog.org/
resources/TrafficStudy-Google.pdf.
65 Ibid, 5.
66 Ibid, 7.
67 See e.g. in the words of J Haucap ‘Monopölchen’, n 2. In contrast to temporary market power, practices that
facilitate the acquisition of a monopoly should not be immune from intervention. On the effects of social
networking, see e.g. SW Waller, ‘Antitrust and social networking’ 90 North Carolina L Rev (2012), 1802.

118 (2015) 11(1) CompLRev

 A Anca D Chirita

Google mobile Google

Android Vertical integration

78.4% smart-phones Google Maps 65.9% (2nd after Facebook)

61.9% tablets (2012)

Google Play 54.4%

Google Search 53.5%

Gmail 47.6%

YouTube 46.4%

Google also transplanted its successful business model from the world of PCs to
mobile devices and tablets. In particular, Google’s open source application, Android,
which is offered freely to owners of mobiles and tablets, is leading the market with 80%
of the market share in the EU. 68 Statistics indicate that 78.4% of the smartphones and
61.9% of the tablets sold are running Android. 69 In 2012, in the US, Google Maps on
mobiles came second after Facebook with 65.9%, followed by Google Play with 54.4%,
Google Search with 53.5%, Gmail with 47.6%, and YouTube with 46.4%.70
Android promotes not only Google’s famous search-engine, but also its other free
services, such as Google Maps and Google Play. The business strategy disguised by
Google’s gratuitous Android offer seeks to maintain its incontestable online dominance
on both static and mobile search-engine markets, and, with this in mind, maintains its
sponsored links plus traffic-based revenues from advertising. Otherwise, the offering
enhances consumers’ satisfaction and, through the tying of mobile devices to pre-
installed software applications, it avoids further fragmentation, too. However, the
shortcoming is, again, a missed opportunity for Google’s competitors to impose their
own search-engine on digital devices.
C. The Concept of Abuse and Potential Anti-Competitive and/or Unfair
Practices
The contentious issue is that Article 102 TFEU has traditionally left outside its scope
the monitoring of marketing strategies of this kind, e.g., through advertising, under the
misleading and comparative advertisement directive and unfair commercial practices
directive. 71 In contrast to the Microsoft-tying case, the borderline between technological

68 See FY Chee and A Oreskovic, ‘European regulators training sights on Google’s mobile software’, Reuters, 30
July 2014.
69 See T Körber, ‘Let’s Talk About Android – Observations on Competition in the Field of Mobile Operating
Systems’, http://ssrn.com/abstract=2462393, 14.
70 See http://www.comscore.com/Insights/Press-Releases/2013/2/comScore-Releases-the-2013-U.S.-Digital-
Future-in-Focus-Report.
71 The Unfair Commercial Practices Directive 2005/29, 11 May 2005, OJ L 149-22/2005.

(2015) 11(1) CompLRev 119

Google’s Anti-Competitive and Unfair Practices

and contractual tying is clearly defined since Google engages in non-technological tying,
irrespective of whether there is a ‘written’ commercial agreement to tie advertising to a
selected AdWord. What matters is that traders bid online with other auctioneers for
their Ad to be placed on a ‘priority’ listing. This can only affect the quality of the
generated search results offered ‘freely’ by Google to its users. 72 In other words,
commercial advertising in this secondary market interferes with free competition in the
primary market as it has the potential to distort and limit the natural listing of search
results through the insertion of these paid ads. This contradicts the finding that the
priority listing of organic searches is not interfered with by Google itself through its
own algorithmic sequence. The quality of the organic searches will depend on the
volume of available advertisements and the suitability of a key word to generate ‘unfair
competition’.
The issue of dealing with the protection of competitors and the allegation that Google’s
business model is set up to favour its own commercial interests to the detriment of
those of its competitors may go beyond the scope of Article 102 TFEU. For example,
one recent complaint from Google’s competitors was that it displays specialised search
queries, such as hotels, restaurants, or flights, more favourably than it displays
competing services. 73 Therefore, while downgrading vertical search services offered by
Google’s competitors 74 could pose the risk of secondary-line discrimination and
preferential treatment, this commercial practice appears more suitable to being dealt
with under, or in conjunction with, the unfair competition rules on advertising rather
than being captured by Article 102(c) alone. The first option, namely, a suit on unfair
competition, should be considered where there are only an insignificant number of
competitors at stake and no negative effects on final users. Otherwise, a general claim
that Google engages in deceptive conduct against its users 75 could be substantiated on
the basis of comparative and misleading advertising, but not on the basis of a
promotion of its own brand. Such claims could be brought before the EC, irrespective
of Google’s dominance on the secondary market since the practice is behavioural, and
it affects competitors directly. This approach was recently applied by a lower court,
Landesgericht, in Hamburg, which ruled that Google displayed its own content
prominently to ‘increase the overall attractiveness of its search engine’, 76 for which
Google enjoys discretion. As such, Google is not required ‘to limit itself to a neutral

72 For the opinion that ‘what appears nowadays in Google search results pages is increasingly paid-for content’,
see e.g. Euractiv, ‘Online consumer choice stifled by lack of competition’, 11 April 2014; for the view that
‘when the producer primarily earns its profits from one side of the market (such as advertising), its incentive
to degrade quality (below levels that consumers prefer) on the other side of the market can increase’, see e.g.
Ezrachi and Stucke, n 38, 21.
73 See MEMO/14/87, n 10.
74 EC, press release, ‘Antitrust: Commission probes allegations of antitrust violations by Google’, 30 November
2010; ‘Antitrust: Commission seeks feedback on commitments offered by Google to address competition
concerns’, 25 April 2013.
75 This is similar to the US FTC’s Act that empowers the FTC to prevent unfair methods of competition as
well as unfair or deceptive acts or practices that are likely to mislead a reasonable consumer.
76 LG Hamburg, Verband Deutscher Wetterdienstleister e.V. v. Google, no 408 HKO 36/13, 11 April 2013.

120 (2015) 11(1) CompLRev

 A Anca D Chirita

presentation of the results of its search algorithm’. 77 In a previous ruling, the High
Regional Court in Frankfurt held that the use of online ads does not misuse a
competitor’s trademark as long as inserting the ad into the search engine does not result
in ‘clear and unambiguous’ advertising.78 In this case, users searching for the registered
trademark received automatically an advertising link to this competitor’s web site.
However, the demand that third party web publishers purchase ‘all or most’ of their
online advertisement requirements from Google, made conditional upon purchase to
guarantee content indexing in Google’s horizontal search engine, 79 should be subject to
scrutiny under the prohibition of contractual tying under Article 102(d) and (b) on
exclusive arrangements. This commercial practice affects not only competitors through
its foreclosure effect,80 but it directly affects the search-engine’s users. By offering
consensual commitments, the EC prevented intervention in the advertising market by
requiring Google to display comparatively specialised searches offered by its rivals, 81
instead of clarifying the law, namely, whether or not Google had breached Article 102
TFEU and/or the harmonising directives on unfair competition. As has already been
argued elsewhere, the EC made critical use of commitments in investigations such as
Google’s, which ‘raise novel legal questions or rest upon less-established theories of
harm’. 82
It is clear that the EU approach to dominance remains unfortunately too conservative
on the concept of abuse as an ‘objective’ one. 83 The role of subjective indicators, such
as the intention to free-ride on original publishers’ content 84 or to downgrade
competing bids in the search-engine results, are not taken as prerequisites for a
monopolist’s anti-competitive conduct, i.e., deception, though such subjective
indicators could build a strong case against the monopolist’s deception in
advertisements. However, Google is immune from any accusation of free-riding since it

77 Ibid.
78 HRC Frankfurt, 6W17/08, 26 February 2008. See also the reference for a preliminary ruling in Joined Cases
CC-236/08 to C-238/08, Google France SARL, Google Inc v Louis Vuitton Malletier SA; Google France SARL v
Viaticum SA, Luteciel SARL; Google SARL v CNRRH, 23 March 2010, in AD Chiriță, The German and
Romanian Abuse of Market Dominance in the Light of Article 102 TFEU (Baden-Baden, Nomos, 2011), 366.
79 See the EC’s MEMO/14/87, n 10.
80 See e.g. Lianos and Motchenkova, n 6, 12.
81 EC, press release, ‘Antitrust: Commission obtains from Google comparable display of specialised search
rivals’, 5 February 2014.
82 Y Botteman and A Patsa, ‘Towards a more sustainable use of commitment decision in Article 102 TFEU
cases’, 1 J of Antitrust Enforcement 2 (2013), 348. In the same vein, see very recently Professor R Whish’s
Editorial, ‘Motorola and Samsung: An Effective Use of Article 7 and Article 9 of Regulation 1/2003’ J of Eur
Comp L & Practice (3 September 2014).
83 See H Schröter, T Jakob, R klotz and W Mederer, Europäisches Wettbewerbsrecht (Baden-Baden, Nomos, 2nd
ed, 2014), 840.
84 See e.g. EC, press release, ‘Antitrust: Commission obtains from Google comparable display of specialised
search rivals’, 5 February 2014; EC Memo, ‘Commission seeks feedback on commitments offered by Google
to address competition concerns – questions and answers’, 25 April 2013.

(2015) 11(1) CompLRev 121

Google’s Anti-Competitive and Unfair Practices

imposed contractual restrictions on the use of its commercial Ads by rival search-
engines. 85
As has already been mentioned, Google’s atypical non-pricing model to its end-users
must extract revenues from elsewhere, such as auction bids for Ad placements, or even
free-riding from web publishers of original content. For example, for indexing
purposes, Google effectively uses other web publishers’ content, such as text snippets
and thumbnails or preview pictures. While this is, indeed, beneficial to the average user,
who is made aware of the original content, there is also a shortcoming to this business
model. Once the search-engine becomes a famous brand, publishers will certainly claim
some revenues for their own inclusion. Then, Google could extract its lost profit from
more advertising, which, in turn, will inevitably worsen the quality of the search-engine
experience.
Google’s successful business model of acquiring incontestable dominance in the search-
engine market has been made possible by a simultaneous gain in prominence in vertical
advertisements. It is a sine qua non condition to improve online presence so that traders’
ads will be linked to Google’s horizontal platform, which in economics is tantamount
to leveraging dominance from one side of the market to the other. By displaying its
own commercial Ads more prominently, Google cannot remain silent on its own
altering of the natural occurrence of the search-engine results. This manipulative
technique can not only discriminate against rivals; 86 but it can also affect the quality of
organic searches offered to all users.
Nonetheless, regarding Google’s search engine as an essential facility, i.e., a universal
search engine that is ‘an indispensable distribution tool,’ 87 is quite hazardous. 88 First,
there are functionally viable alternative search-engines, and while competitors could
incur significant costs to reach the monopolist’s storage of information 89 so as to make
the engine more attractive, this has already been replicated, though unsuccessfully to
date. The ranking methodology, such as search listing, can only be considered as
proprietary trade secret information rather than an essential input. However, one
powerful argument in favour of applying the essential facilities doctrine is the current
EU loophole on the protection of trade secrets. This is due to change following an EU
proposal that regulates trade secrets 90 as information which is ‘generally known among
or readily accessible to persons within the circles that normally deal with the kind of
information in question, has commercial value because it is secret, and has been subject

85 However, free-ride on Google is not acceptable see e.g. Speech by J Almunia, ‘Public policies in digital
markets: reflections from competition enforcement’, 30 June 2014.
86 See e.g. Pal, n 47, 73.
87 Lianos and Motchenkova, n 6, 16.
88 For the suggestion that the listing of results be considered as an essential facility, see e.g. Lao, n 60, 302; on
the inapplicability of the essential facilities doctrine in the case against Google, see Bork and Sidak, n 6, 13.
89 For the view that duplication of search engines is clearly impossible, see e.g. C Argenton and J Prüfer,
‘Search Engine Competition with Network Externalities’ 8 J of Comp L & Ec (2012), 73, 97.
90
Proposal for a Directive of the European Parliament and of the Council on the protection of undisclosed
know-how and business information (trade secrets) against their unlawful acquisition, use and disclosure,
Brussels, 28 November 2013, COM (2013) 813 final.

122 (2015) 11(1) CompLRev

 A Anca D Chirita

to reasonable steps under the circumstances, by the person lawfully in control of the
information, to keep it secret’. 91
The ‘economically viable’ test developed under the EU essential facilities doctrine is
also critical to the extent to which it could afford competitors’ protection, rather than
to address real competition law issues. The link to a commercial refusal to deal or
supply is entirely missing in the Google case. 92 It is, therefore, clear that a case against
Google can be instrumental in creating a new ‘precedent’ applicable to instantaneous
bids for online advertisements targeted at dynamic searchers rather than following old-
fashioned doctrines where the input is statically physical and not virtually dynamic.
Finally, the most powerful argument against Google’s anti-competitive conduct that
harms final consumers directly is their online monitoring and gathering of personal
information, 93 such as server logs from users’ browsers, which is passed on to
advertisers and used for commercial purposes.94 Similar to Microsoft’s problem of the
lack of interoperability of its Windows Operating System with non-competing systems,
the Achilles’ heel that demands vigorous scrutiny and intervention is not the existing
competition on its rather insignificant (Windows Media Player or search-engine)
market, but is the general public interest in ensuring that a privately-owned corporation
does not successfully expand over so many tiny, inoffensive, and innovative markets in
order to enjoy the monopolistic power over its users’ personal data, which could be
passed on to third parties for commercial advertising. This is similar to the loyalty cards
offered by supermarkets, which monitor buyers’ preferences and eventually offer them
discounts on their next shopping.
However, Google’s quasi-dominance on the search-engine market could raise legitimate
national and EU security concerns, which explains the political sensitivity of this case.
This is why the Bundeskartellamt proposed recently that Google should be regulated as
‘utilities’. 95 In sharp contrast, a policy proposal has been put forward by economists
according to which ‘all search engines should be required to share their anonymized
data on clicking behaviour of users following previous research queries’. 96 While
allowing competing search engines access to the users’ behavioural data makes perfect

91 See e.g. the timely contribution of N Sousa e Silva, ‘What exactly is a trade secret under the proposed
directive’, J of Intellectual Property Law and Practice (2014).
92 This time, the scope of the doctrine is different from Cases 6-7/73, Commercial Solvents, [1974] ECR 223; Case
238/87, Volvo v Veng, [1988] ECR 6211; Case C-7/97, Oscar Bronner, [1998], ECR I-7791; Case C-418/01,
IMS Health, [2004], ECR I-5039; Case C-52/09, TeliaSonera, [2011] ECR I-527; Case C-209/10, Post Danmark,
ECLI:EU:C:2012:172.
93 On the ‘right to be forgotten’, see C-131/12, Google v AEPD and Costeja, ECLI:EU:C:2014:317; H Crowther,
‘Google v Spain: is there now a ‘right to be forgotten’?’ J of Intellectual Property L & Practice (2014). However,
free-ride on Google is not acceptable; see e.g. Speech by J Almunia, ‘Public policies in digital markets:
reflections from competition enforcement’, 30 June 2014.
94 See the concern expressed by the German Monopolies Commission on the use of private data by Google
and Facebook, Monopolkommission, press release, ‘Google, Facebook and Co. – eine Herausforderung für
Wettbewerbspolitik’, Bonn/Berlin, 9 July 2014; see e.g. A Gebicka and A Heinemann, ‘Social Media and
Competition Law’ 37 World Comp L & Ec Rev 2 (2014).
95 PC Tech, N Kamanzi, ‘Germany Cartel Offices Says Google Could be Regulated as Utilities’, 14 July 2014.
96 Argenton and Prüfer, n 89, 77.

(2015) 11(1) CompLRev 123

Google’s Anti-Competitive and Unfair Practices

economic sense, the EU rules governing privacy 97 and security would not allow this to
happen, even if the data were anonymised.

III. GOOGLE’S ANTI-COMPETITIVE PRACTICES ON ADJACENT LEISURE

MARKETS
In the US, the consumer class action 98 submitted that Google made its most popular
applications, such as mobile and tablet Android applications, including the Google
Phone-top Search app, 99 YouTube, Google Maps, Gmail, and Google Play, all of which
enable mobile customers to buy music, movies, and books from the Google Play store,
subject to an unlawful tie-in secret agreement called ‘Mobile Application Distribution
Agreement’. 100 The latter required manufacturers of mobile phones and tablets to make
Google the default search engine on all of their devices sold to consumers worldwide.
For example, it is submitted that ‘Google uses its popular apps to coerce manufacturers
into making it the default search engine provider on handheld devices’. 101 All of the
above agreements are ‘contracts in restraint of trade’ 102 and fall under the scope of the
US Sherman Act. Thus, the plaintiffs mention various other acts dealing with unfair
methods of competition, such as the Clayton Act or California’s Unfair Competition
Act.
‘Google has conditioned the rights to pre-load any application from a suite of
Google applications, including the YouTube app or the Google Play client, on the
manufacturer’s mandatory acceptance and installation of Google search, or so-
called Google Phone-top Search, as the default search engine on that device’. 103
This means that if a smartphone or tablet manufacturer wishes to pre-load the popular
YouTube app on a given Android device, then it has to make Google the default search
engine, too. 104 In other words, ‘a manufacturer would struggle to offer a phone without
a pre-installed YouTube app’, 105 which makes mandatory the adoption of the default
engine. The same technique applies to other rivals, for example, were AOL to pre-load
its MapQuest onto Android mobile devices. 106 Furthermore, the plaintiffs submitted
that Google had paid Apple hundreds of millions of dollars to act as the default engine

97 See e.g. EU Data Protection Directive 95/46/EC of the European Parliament and of the Council of 24
October 1995 on the protection of individuals with regard to the processing of personal data and on the free
movement of such data, OJ [2005] L281, and the proposal for a new EU Data Protection Regulation,
available at http://ec.europa.eu/justice/data-protection/.
98 Initiated in the US by B Edelman, see e.g. ‘Secret Ties in Google’s ‘Open’ Android’, available on

http://www.benedelman.org/news/021314-1.html. Many thanks to Benjamin Edelman for the pointer.

99 A widget for conducting web searches via Google’s search engine.

100 US Class Action, para 74.

101 US Class Action, para 32.

102 US Class Action, para 7.

103 US Class Action Complaint 010437-11 683086 VI, para 74.

104 Ibid, para 33.

105 Ibid, para 40.

106 Ibid, para 53.

124 (2015) 11(1) CompLRev

 A Anca D Chirita

on Apple iPhones, iPads, and iPods. 107 This pre-installing trend is noticeable on
Samsung, too.
A mitigating factor is that, unlike in the Microsoft tying case, 108 where users could not
remove the Windows Media Player, mobile devices entail Google’s default search-
engine is set up by manufacturers; thus, users are able to download other applications,
interoperate, and/or change the settings. 109 Of course, as in the Microsoft-Internet Explorer
tying case, 110 users may lack basic IT skills while downloading and installing an internet
browser other than the default one, as well as inertia. Google’s business model is the
same: offering Android for mobile devices free of charge. This means that should a
user decide to install competing applications, there is no guarantee that they will also be
offered for free. A flexible option to switching may still require extra costs and operate
as a potential barrier to market entry. This makes any pre-loading and pre-installation
not particularly user friendly both to intermediary consumers, i.e., competitors, and to
final consumers, i.e. ordinary users. 111 Therefore, combining both tying cases, the
answer is that intervention against Google could proceed. This is unlike the seemingly
inconsistent, shifting approach to network effects pronounced in recent merger
proceedings. 112 In Cisco Systems and Messagenet,113 the General Court (GC) departed from
the previous Microsoft cases as there were no technical or economic constraints that
could have prevented intermediary consumers, active on a narrower market for
business communications, from downloading competing services on their device. The
merged entity between Windows Live Messenger and Skype raised no entry barriers as
the communications software had been offered for free, being made easy to be
downloaded by corporate clients and occupying a limited space on their PC’s hard
drive. In this scenario, switching was found to be relatively easy and multi-homing
possible. However, the business model operated by the merged entity is slightly
different from Google’s in that the free offering of communications to final consumers
is made possible by the revenues extracted from the placement of commercial
advertisements, while corporate clients are in no way subsidised by the final users.
Rather, such intermediary users themselves have to pay for their enterprise
communications services. However, what makes the previous approach to network
effects inconsistently applied is precisely the much narrower scope for the definition of
the relevant market for businesses’ communications services, as opposed to those of
final users.
Unfortunately, the EU merger policy and the prohibition of abuse of dominance may
bring about contradicting outcomes. The recent approval by the EC of the acquisition

107 Ibid, para 62.

108 Case COMP/C-3/37792 – Microsoft (WMP), OJ L 32-23/2007; Case T-201/2004, Microsoft, [2007] ECR II-
3601.
109 See e.g. Körber, n 69, 2.

110 See Case COMP/C-3/39530 – Microsoft (Internet Explorer).

111 For the view that the storage memory of the Google suit of apps is negligible, see e.g. Körber n 69, 42.

112 For an insightful analysis, see e.g. I Graef, ‘Sneak preview of the future application of European competition

law on the Internet?: Cisco and Messagenet’ 51 CML Rev (2014), 1263-1280.
113 Case T-79/12, Cisco Systems Inc and Messagenet SpA v Commission, ECLI:EU:T:2013:635, paras 79-81.

(2015) 11(1) CompLRev 125

Google’s Anti-Competitive and Unfair Practices

by Facebook of WhatsApp was justified due to the flexible switching by customers

from one mobile application to another, the popularity of the network amongst users,
and the fast growing market for consumer communications apps, in particular, its ‘short
innovation cycles’. 114 The initial premise for allowing a merger appears to be that an
emergence of dominance ex post is to be ignored a priori because of the unknown pace
of an innovation cycle. This merger-friendly approach can only sit awkwardly when,
under Article 102, it is thereafter suspected that Facebook may be dominant on the
social networking platforms. It highlights how following preceding evaluations of
network effects applied in the context of allowing mergers is simply a false friend
should it be followed in another area.
While the quantification and recovery of the research and innovation pursuits by
Google could be extremely difficult to measure, the revenues extracted from its
commercial advertising represents a valid indicator that Google does not compete only
on the basis of its zero-priced and innovative search-engine.
A final argument in support of intervention has to consider the user’s experience,
which has to be balanced somewhere in the middle to avoid a panoply of pre-installed
applications, such as Desk Clock, Browser, Calendar, Contacts, Gallery, Global Search,
Launcher, Music, Google Talk, and Settings, against the danger of fragmentation.
Would the average user be better off without pre-installing these additional features?
Based on the current consumers’ class action, the balance seems to be tipped in favour
of a ‘yes’ answer.
The plaintiffs questioned primarily the existence of a mandatory secret tie-in agreement,
whether Google unlawfully monopolized or attempted to monopolize the search
engine market in respect of the Android apps, and whether consumer harm can be
established from consumers having paid more for mobile devices 115 with such pre-
installed features than they could otherwise have paid had more choice been
available. 116 In other words, this raises another pertinent question of whether when
buying the mobile hard-core device, any increment will also be paid towards the
running of software applications. Since Google has paid manufacturers of mobile
devices for inclusion, this could be a competitive advantage to enhance consumers’
experience. However, as no such apps’ costs are passed on to consumers, the pre-
installing of Google features could be used as an artificial means to maintain its present
dominance, rather than as a natural enhancement of the hard-core device. In the long
run, the effects on consumers will be a loss of choice and possibly a lack of innovation.
The above practice is similar in scope to Article 102(d) TFEU, which prohibits ‘making
the conclusion of contracts subject to acceptance by the other parties of supplementary
obligations (…) which have no connection with the subject of such contracts’. In
particular, the above contractual tie-in agreement could eventually oblige manufacturers

114 See recently, EC, press release, ‘Mergers: Commission approves acquisition of WhatsApp by Facebook’,
IP/14/1088, Brussels, 3 October 2014.
115 For the contrary opinion that Google’s offering of Android as open source software has actually lowered the

price of mobile devices, see Körber, n 69, 4.

116 See Cisco Systems, n 113, paras 79 a, c, and e.

126 (2015) 11(1) CompLRev

 A Anca D Chirita

of digital services to pre-install, predominantly or exclusively, Google’s Android

applications content and to accept its online search engine by default. 117 This
commercial practice, obviously, pre-empts the choice of individual users, 118 namely,
using or downloading other competing applications and/or search engines. The
practice is anti-competitive as it encourages final users of smartphones and tablets to
experience only Google’s pre-installed applications. The purpose of this business
strategy has to be understood in the wider context of maintaining Google’s reputation
and dominance in the horizontal and vertical search-engine markets. Due to such sunk
costs, for example, the offering of Android applications as open source software,
Google’s rivals are practically denied an effective entry into the market for digital
devices, which are mostly used for personal entertainment. In addition, YouTube and
Google Play installation is also made conditional upon the manufacturer’s pre-
instalment of Google Phone-top search.
FairSearch’s complaint against Google 119 is similar to the criticism of the US’s plaintiff,
i.e., that ‘Android phone makers who want to include must-have Google apps such as
Maps, YouTube or Play are required to pre-load an entire suite of Google mobile
services and to give them prominent default placement on the phone’. This obviously
places Google at a competitive advantage vis-à-vis its competitors.
One can sum up from the above that Google’s ubiquitous presence 120 has imposed
itself not only on PCs but also on all other digital devices. In the light of these recent
developments, similar to the booking of cheap or last minute flights or hotels,121 the
previous allegations of ‘biased’ or ‘manipulated’ search results122 in the advertisement
market have gained more weight. It is plausible to think of Google’s search engine
giving priority to a good number of commercial ads over others, which will remove the

117 On Android as a ‘Trojan Horse’ see e.g. Chillin’ Competition, ‘Some thoughts on the new anti-Google
(Android) complaint’ available at http://chillingcompetition.com/2013/09/09/some-thoughts-on-the-new-
anti-google-android-com.
118 US Class Action, para 64, where the plaintiffs emphasised the lack of choice and the stifling of innovations

as a result of Google’s monopoly.

119 See http://www.fairsearcheurope.eu/wp-content/uploads/2013/04/FairSearch-Announces-EU-Complaint-

on-Google-Mobile-Strategy-9-April-2013.pdf.
120 Recently it has been suggested that Android forced access to a ‘collection’ of thirteen Google applications,

namely Google Chrome, Google Maps, Google Drive, YouTube, Gmail, Google+, Google Play Music,
Google Play Movies, Google Play Books, Google Play Newsstand, Google Play Games, Google+ Photos
and Google+ Hangouts, in Quartz, ‘Google is ‘tightening the screws’ on Android to keep control over the
web’, 26 September 2014.
121 See J Hazan, ‘Stop Being Evil: A Proposal for Unbiased Google Search’ 111 Michigan L Rev (2013) 789, 805.

122 See e.g. B Edelman, ‘Hard-Coding Bias in Google ‘Algorithmic’ Search Results’, 15 November 2010,

available at http://www.benedelman.org/hardcoding/; Hazan, ibid, 789; for the view that the appearance of
Google’s products in search results is three times higher than for non-Google products, see e.g., B Edelman
and B Lockwood, ‘Measuring Bias in ‘Organic’ Web Search’, available at
http://www.bededelman.org/searchbias/; BG Edelman, ‘Google’s Dominance and What to Do About It’ 2
J of Law, ACS Blog ‘Debate on Antitrust Scrutiny of Google: Benjamin G. Edelman vs. Joshua D. Wright’,
453: ‘It’s handy to have a single Google password providing access to personalized search, finance, videos,
and more. But this misses the serious harms of Google’s ever-broadening panoply of services’; for the view
that it is possible to manipulate or trick algorithm results, see e.g. L Introna and H Nissenbaum, ‘Shaping the
Web: Why the Politics of Search Engine Matters’ 16 The Information Society 3 (2000), 141.

(2015) 11(1) CompLRev 127

Google’s Anti-Competitive and Unfair Practices

expected ‘neutrality’123 when generating results in the primary search-engine market.

For example, airline companies were found to have weighted in their booking system
carrier-specific factors, to have inflated the flight times of competitors, to have blocked
flights with lower fares, and so on. 124
A dangerous option is allowing cross-subsidisation from the free primary market to
expand over the secondary digital market. Then, Google’s brand and popularity in the
primary market will leverage the secondary to the detriment of other competing
applications. Recently, a Portuguese competitor, Aptoide, submitted to the EC that
Google is ‘leveraging’ its dominant position in the Android market to control the
market for such applications. 125
The litigious aspect of intervention could follow suit under the umbrella of unfair
competition; thus, the crucial element is that the likelihood of exclusionary conduct
exists. Furthermore, it is difficult to see how users or final consumers, who know little
about how the search itself operates, could be alerted to the decline in the quality of the
search engine. The secrecy of Google’s algorithmic metric has already been referred to
as creating a ‘black box’ effect 126 with the consequence that consumers lack the
knowledge to realise they are being misled. As long as the results do not become merely
advertising, it is the task of the competition authorities to monitor how the search
engine operates and to require manufacturers to include competing applications on
digital devices. Pre-installing certain features can work only in Google’s favour to gain
an anti-competitive advantage and not to compete exclusively on the merit of its
products. Limiting ab initio consumer choice can only have a ‘lock-in’ effect on
consumers who cannot experience the quality of alternative products or services,
thereby foreclosing potential competition. Nonetheless, this approach is very much
consumer-welfare focused. Advocates of non-intervention appear somewhat malicious
to a consumer-welfare approach since in the long run, Google’s aggressive business
model conquers every single digital market. It cannot be that in the name of avoiding
fragmentation, all pre-installed features are made by Google.
From the conundrum of arguments for and against intervention in the case of Google,
one clear conclusion is that, unlike the US Sherman Act, having an open-textured
provision in Article 102 TFEU makes it easier to attack Google’s abuse in the leisure
digital market since exclusive and conditional agreements harm both intermediary and
final consumers, thereby maintaining Google’s dominant position. Anti-competitive
practices of this kind are to be fined rather than negotiated and finalised with light-
touch commitments. The monitoring of Google for the next five years and three
months will partially solve the existing problems. For example, Google will cease to

123 For an interesting contribution on neutrality, see M Thompson, ‘In Search of Alterity: on Google, neutrality,
and otherness’14 Tulane J of Technology & Intellectual Property (2011).
124 See e.g. Langford, n 29, 1587.

125 ‘Google hit with new antitrust complaint in Europe, officials confirm’, 23 June 2014, available at

http://www.pcworld.com/article/2364580/google-hit-with-new-antitrust-complaint-in-europe-officials-
confirm.html.
126 See O Bracha and F Pasquale, ‘Federal Search Commission? Fairness, Access and Accountability in the Law

of Search’ 93 Cornell L Rev (2008), 1149.

128 (2015) 11(1) CompLRev

 A Anca D Chirita

include ‘any unwritten obligations’ in future contracts, which could amount to a de facto
exclusivity agreement to source any commercial requirements for ads solely from
Google. 127 This behavioural remedy cannot address the past anti-competitive harm
caused to competitors. What appears as a daunting task for Google is to create a pool
of eligible Rival Vertical Search Sites according to objective criteria established by the
EC. The site domain should enjoy some popularity based on its usage data and may not
engage in harmful commercial practices, such as the deception of search engines,
including index gaming, sneaky redirects, keyword stuffing, link spamming, or any other
practices designed to deceive or manipulate legitimate site indexing and ranking. 128 The
EC also mentioned the deception of consumers through deceptive or frustrating
navigation, bait and switch advertising, deceptive billing practices, or other practices
that mislead consumers. The end effect of what is being implemented is a ‘regulation’
of online commerce aimed at prohibiting deceptive practices to users and competitors.
It is a welcome development that the specialised vertical market has been opened to
external competition from Google’s rivals. However, while this may be pleasing to
Google’s competitors, overall, these commitments do not touch upon the potential
manipulation of results by Google’s own algorithmic metrics. The orientation of
commitments is towards responding to competitors’ complaints, while the
commitments remain minimalistic in terms of dealing with wider concerns in the
horizontal search-engine market. Given that more vertical searches will be included,
this could have a negative impact on the users’ experience in the primary market.
At the EU level, the existing Electronic Commerce Directive 129 also provides for a
mechanism to apply for an injunction aimed at ‘the protection of the collective interests
of consumers’.130 This mechanism is devised to facilitate the free movement of services,
thereby ensuring ‘a high level of consumer protection’. Member States may only restrict
the free movement of services in exceptional cases of ‘public security, including the
safeguarding of national security and defence, and the protection of consumers’.131
Therefore, in the absence of a specialist regulation of online commerce, Article 6 of the
Electronic Commerce Directive could be helpful as it refers, in particular, to
commercial communications, i.e., advertising, including promotional offers, such as
discounts, premiums, gifts, and unsolicited advertisements. In the eyes of its end-users,
Google’s search-engine operates an online system of unsolicited advertisements.
Furthermore, the area of competition is not excluded from the scope of application of
this directive, as are, notably, tax, VAT, fiscal, or data protection issues.

127 See EC, Commitments, Case COMP/C-3/39740 – Foundem and others, 3 April 2013.
128 Ibid.
129 See e.g. the EC Directive 2000/31/EC of the European Parliament and of the Council of 8 June 2000 on

certain legal aspects of information society services, in particular electronic commerce, in the Internal Market
(Directive on electronic commerce), OJ [2000] 178/1.
130 See recital 53.

131 See Article 3 (4) (a) (i).

(2015) 11(1) CompLRev 129

Google’s Anti-Competitive and Unfair Practices

The cornerstone of the ongoing investigation against Google remains public policy, or
wider non-economic considerations, such as the online privacy of Google’s users as a
fundamental human right. 132

IV. CONCLUSIONS
The negotiated and re-negotiated commitments in the Google investigation are an
attempt to regulate online commercial advertising that is taking place when, on one
side, users search for keywords while on the other, advertisers bid for such keywords to
distract users’ attention. What once was essentially a process of searching through the
internet has evolved into an advertising engine. A giant processor of a vast amount of
information, Google has been under fire from competitors who felt downgraded in
commercial search results when bidding for a place in specialised ads. The EC chose
not to respond with a fine on the alleged anti-competitive conduct, as has traditionally
been the case, but to remedy a truly complicated situation. Unfortunately, the proposed
commitments fail to clearly address a wider competition concern over the possible
decline in the quality of the organic search results as a result of more commercial ads,
which will have an impact on Google’s final users/consumers.
The commitments accepted by Google are not only quite long, but are also ambiguous
pronouncements of the law applicable to Google’s future conduct in the market.
Therefore, the lack of an erga omnes effect of these commitments means they are limited
to Google only, instead of clarifying the law for the benefit of many other online
service providers, including advertisers. However, the most sensitive aspect of the
above commitments remains the fact that they are not being exposed to judicial review.
Therefore, unless they are successfully implemented and accepted by Google’s vocal
competitors and final users, such commitments may, indeed, weaken EU competition
enforcement, in particular, in the area of abuse of dominance where the number of
preceding prohibition decisions by the DG COMP has declined since the famous
Microsoft ruling. Probably, the widespread criticisms of the theories of harm and
network effects employed by the EC in the latter ruling discouraged the DG COMP
from taking a bolder action.
In contrast, the EU Parliament’s recent non-binding resolution, which passed with an
overwhelming majority of 384 votes, is to be welcomed since it suggested a possible
divesture of internet giants by ‘unbundling search engines from other commercial
services’. 133 Should this happen to Google in the near future, the resolution will also
apply, without discrimination, to all search-engines’ providers.

132 See recently the European Commission’s Statement 14/1646, ‘Statement by Commissioner Vestager on
Google antitrust investigations at the European Parliament (ECON committee meeting)’, Brussels, 11
November 2014.
133 See EU Parliament’s Plenary Session of 26 November 2014, ‘MEPs to vote on plans to separate internet

search engines from commercial activities’, available on http://www.europarl.europa.eu/news/en/news-

room/content/20141126STO80726/html/MEPs-to-vote-on-plans-to-separate-search-engines-from-
commercial-activities.

130 (2015) 11(1) CompLRev

 A Anca D Chirita

Nonetheless, one thing is clear: irrespective of whether there will be a case or not, there
will always be an angle of criticism for academics to lean on. However, the final
decision rests in the very capable hands of the DG COMP to review the matter giving
careful consideration of all the facts of the case, without any fear of further
consequences and outside pressures, coming notably from the colour of politicians who
also include the Commissioner for Competition, national ministers, and many more.

(2015) 11(1) CompLRev 131