“In the United States, the antitrust laws are enforced to protect consumers by protecting competition, not competitors. In the absence of demonstrable consumer harm, all companies, including dominant firms, are encouraged to compete vigorously. U.S. courts recognize the potential benefits to consumers when a company, including a dominant company, makes unilateral business decisions, for example to add features to its popular products or license its intellectual property to rivals, or to refuse to do so”.

The September, 17th 2007, Thomas O. Barnett, then Assistant Attorney General for the Department’s Antitrust Division, issued a now emblematic statement after the Court of First Instance of the European Communities (CFI) affirmed the substance of the European Commission’s (EC) March 2004 decision against Microsoft. 

The lawsuits against the dominant operators seemed to have very different fates on either side of the Atlantic. The differences could be explained by several factors. On the one side, the aim was to protect the competitive process for its own sake by placing a particular responsibility on the dominant operator. On the other side, it was a question of basing the enforcement of the competition rules on a single criterion: consumer welfare. In other words, the protection of competition is limited to the search for its result defined exclusively as allocative efficiency[1].

The divergences of views were also based on a rather different estimate of the cost of errors in antitrust decisions. A European view tends to consider that false negatives are the most costly because some markets, once characterised by a significant dominant position, cannot easily be “contestable” in a competition-law based perspective, owing to barriers to entry (financial, technical, informational, or related to network effects, etc.). The competition process may therefore encounter a structural failure against which no competitive remedy can be effective. Conversely, a hypothesis of self-regulating markets makes the cost of the false negative negligible. The monopoly, the roi fainéant(e.g. the monopolist enjoying its quiet life), will soon be the victim, if not of a palace revolution, at least of a technological and competitive disruption. Symmetrically, no market signal in the form of abnormal profitability can correct a false positive. Not only is the dominant company deprived of the fruits of its past investments and efforts, but the practice is prohibited for all market players knowing that it could only be implemented modulo a reversal of case law. 

These gaps materialised in the autumn of 2008 in a DoJ report on single firm conduct[2]. Preventing the risk of false positives and mistrust of potentially opportunistic lawsuits by competitors led to raising the standard of proof on the plaintiff to such an extent that lawsuits based on an alleged monopolization strategy seemed doomed to failure. Indeed, despite the withdrawal of this report in the spring of 2009, the activation of Section 2 of the Sherman Act remained staggered over the decade.

Two points are worth noting. The first relates to the decision-making practice of the courts and the second relates to the antitrust policy pursued by the US authorities themselves. 

Firstly, the courts’ decision-making has been very favourable to the defendants. A few data produced in 2010 by Judge Douglas Ginsburg are sufficient to understand the extent of the reversal of case law observed in the United States since the mid-1970s[3]. Between 1967 and 1976, defendants were successful in 36% of the judgments handed down by the Supreme Court in antitrust cases (16/44). Between 1987 and 1996, the rate was 50%. Finally, between 1997 and 2006, it reached 100%. Douglas Ginsburg also shows that the judgments were delivered with a super-majority (more than 6 of the 9 judges). Between 1967 and 1976, the plaintiff won in 55% of the cases with this configuration. Between 1997 and 2006, the defendant won under these conditions in 85% of the judgments. Similarly, in the case of a private enforcement action, the DoJ’s amicus curiae award went to the plaintiff in 86% of cases under the Carter administration, 55% under Reagan, 40% under George Bush, 33% under Clinton, and only 10% under George W. Bush. 

Second, the number of antitrust suits initiated by the Antitrust Division of the DoJ shows a dramatic decline in the number of suits initiated under Section 2 of the Sherman Act. The following table taken from the statistics published by the Antitrust Division for the fiscal years 2010 to 2019 shows that while activity in terms of initiating investigations has remained relatively steady in the areas of anticompetitive agreements (Section 1 of the Sherman Act) and merger control (Section 7 of the Clayton Act), this was not far off the case for Section 2 of the Sherman Act, including for the period corresponding to the Obama administration… 2019 therefore marks a turning point, of which the complaint against Google is a possible first step[4]. 

The litigation against Google, whether it is the first in a series that would mark a revival of public enforcement of US competition rules (based on Section 2 of the Sherman Act or Section 5 of the FTC Act) or whether it is an isolated case, is therefore unprecedented in view of the practice of the US authorities for more than forty years now. This is the first procedure opened on this basis since the Microsoft case[5]at the end of the 1990s, which was concluded as a settlement in November 2001. In other words, it is an extremely rare procedure for four decades now, whereas within the European Union there have been numerous proceedings based on Article 102 TFEU.

However, if this procedure was somewhat unexpected in that it was initiated by a Republican administration a few days before a presidential election[6]and that for the last four years the Democrats have been at the forefront of the fight against the Big Tech[7], it takes place in a context that is less and less favourable for the latter, a context that is in part a Tech-Lash but which leads to concrete legal and political effects. In addition to the enquiries launched in 2019, two important events should be highlighted.

The first event was the retrospective enquiry initiated by the FTC in February 2020 on the external growth operations carried out by Big Tech. Two points were already worth noting: firstly, the debate on killer acquisitions or at least consolidating acquisitions (of a dominant position or a pivotal (or keystone) role in an ecosystem) is finding meaning through an investigation initiated by the US administration; secondly, the companies targeted correspond to the GAFAM (Alphabet Inc, Amazon.com, Inc, Apple Inc, Facebook, Inc, Google Inc, and Microsoft Corp.). 

The second event is much more recent. It is the bipartisan report prepared for the Antitrust Subcommittee of the Judiciary Committee of the House of Representatives (investigation of competition in digital markets) and published on 4 October 2020. This report – whose neo-Brandeis impregnation was noticeable[8]– had two characteristics. A first characteristic, in line with the investigation launched by the FTC in February, is that it targets the behaviour of four specific companies: our GAFA (Microsoft not being specifically considered). A second characteristic is that it proposes to converge American antitrust practice towards European standards, at least, through an evolution of Section 2 of the Sherman Act[ix]: « Strengthening Section 2 of the Sherman Act, including by introducing a prohibition on abuse of dominance and clarifying prohibitions on monopoly leveraging, predatory pricing, denial of essential facilities, refusals to deal, tying, and anticompetitive self-preferencing and product design » (p.20). 

The complaint naturally relates to Section 2 of the Sherman Act as enacted in 1890 and as interpreted in the Supreme Court’s decisional practice. The latter was defined as follows by Judge Antonin Scalia in his redaction of the Trinko judgment (adopted unanimously): « The mere possession of monopoly power, and the concomitant charging of monopoly prices, is not only not unlawful; it is an important element of the free market system. The opportunity to charge monopoly prices—at least for a short period—is what attracts “business acumen” in the first place; it induces risk taking that produces innovation and economic growth. To safeguard the incentive to innovate, the possession of monopoly power will not be found unlawful unless it is accompanied by an element of anticompetitive conduct^[x]”. 

A market practice may be sanctioned on the basis of Section 2 of the Sherman Act if a monopoly position is acquired, maintained or extended on a basis other than the merits. However, following the example of the Judiciary Committee’s report of October 2020, elements that fall within the European competition lexicon are emerging: damage to the competition process, market access, innovation and freedom of consumer choice. This is no longer the register adopted by the Supreme Court since 1979, which in its Reiter v Sonotone Corp. ruling[11]had taken over the definition of Section 2 as defined by Robert Bork in his Antitrust Paradox: a consumer welfare prescription[12].This shift attests to the influence of the arguments of the neo-Brandeis approach, which now goes beyond the Democratic Party alone[13]. However, antitrust proceedings initiated by the DoJ are brought before courts of law whose decision-making practice can only evolve through a reversal of the Supreme Court’s case law, the likelihood of which can be considered with some circumspection in view of its composition[14].

In any case, there are four central points to note before analysing the complaint more thoroughly.

Firstly, the theory of damage that seems to outline the DoJ is very close to the three Google decisions of the EU Commission of 2017, 2018 and 2019[15]… which had been criticised by part of the doctrine in that they would not have demonstrated the existence of damage to the consumer. 

Secondly, the similarity of the incriminated market behaviour on both sides of the Atlantic may give the impression that the practices covered by the procedure opened at European level in 2010, which gave rise to a first notification of grievances in 2015 and a second one in 2016, have not come to an end despite the three decisions cited above. However, the scope of the practices in question is wider due to the very development of the firm’s activities. The diversification of the firm towards voice assistants, on-board systems in the automotive sector and of course the Internet of Things both in the domestic space and in the industrial world gives rise to fears of the extension of the domain of dominance and the consolidation of the lock-in of the various members of the ecosystem.

Thirdly, the complaint filed by the DoJ presents some specificities with regard to the procedure conducted by the European Commission in that it does not limit its analysis of the functioning of the Android ecosystem to “vertical” links only, which consisted in reducing barriers to entry into the ecosystem for the complementors (through the supplying of APIs, access to the applications’ shop…) and increasing barriers to exit through loyalty reward mechanisms (e.g. financial incentives for the use of the data collected), which can constitute the equivalent of a financial penalty in the event of breach of agreements; the loss of a loyalty bonus acts as a penalty for disloyalty[16]. The American complaint concerns payments for the pre-installation of the search engine on Internet browsers competing with Chrome outside the Google ecosystem. These browsers may be horizontal competitors for Chrome, such as Mozilla’s Firefox, but may also belong to an ecosystem that is itself a competitor, such as Apple’s Safari. Certain issues could have been considered on the basis of Section 1 of the Sherman Act insofar as the financial incentives paid may echo both horizontal and vertical links (Apple, LG, Motorola, AT&T, T-Mobile, Verizon, Opera, Mozilla) that can be understood as exclusivity payments. 

Fourthly, if the threat, if not of dismantling, at least of structural remedies, is present in the complaint, it appears to be marginal and, to say the least, somewhat ritual. This was also the case in the report of the Judiciary Committee. If dismantling has indeed been carried out in the United States and is the distinguishing feature of Antitrust, with the dismantling of Standard Oil in 1911[17], the fact remains that the last operation of this type dates back to 1982 (AT&T[18]) and could have been part of a logic of sectoral deregulation rather than antitrust policy. Difficulties in implementing the latter may moreover play a repulsive role.  Thomas Sullivan notes that the support for the dismantling of AT&T monopolised Judge Green, who oversaw it for seven years during which he handed down 160 decisions[19]. The Microsoft case, as conducted in the United States by Judge Jackson[20]before it was settled by consent, could certainly have given rise to such a structural separation operation, but was it already technically possible more than twenty years ago? 

Examining the complaint reveals some of the now increasingly well-known themes that can be linked to the damage to competition resulting from practices implemented by digital ecosystem keystone operators as long as they are in a position to exercise control over access to the ecosystem (gatekeeper position) and private regulatory power over it. This structuring power makes the keystone capable of influencing the investments, prices and technological choices of its complementors and thus of deciding which operator and which service or content can access the market and the consumer. 

The result can be an ability to neutralise competitive threats directly at the heart of market power, whether it be search engines by pre-installing Google Search on the main alternative access pathways (Apple Safari, Mozilla, Firefox) or mobile operating systems. This second strategy could be achieved by putting obstacles to the development of forks. The case of Amazon Fire (§130), whose expansion could, according to the complaint, have been hindered by anti-fragmentation provisions, echoes the arguments of the European Commission in its Android decision of 2018.

Consolidation of the market position can also be achieved through horizontal and vertical arrangements used as mechanisms to limit competitive pressure within and between digital ecosystems (which as such would limit competition between the different silos formed by the respective GAFA ecosystems) and to consolidate the dominant position in online search services. 

The vertical foreclosure, as set out in the complaint filed by the DoJ, largely echoes the arguments of the European Commission in the Android decision cited above. Contractual provisions relating to access to the Android ecosystem could have a restrictive effect on competition in that they would limit the ability of complementors to act autonomously. Moreover, revenue sharing mechanisms linked to the exploitation of data create incentives for exclusivity and align the incentives of the complementors with those of the ecosystem keystone player, especially since their profitability depends on these payments. The loyalty of ecosystem members must therefore be assessed both in terms of the financial importance of RSAs – Revenue Sharing Agreements (§85) and direct technical aids such as the provision of APIs, which have the effect of lowering costs and improving the performance and quality of the services provided by the complementors. 

In this, the case of Mozilla is emblematic (§185). Mozilla Foundation’s 2018 financial report states that “the majority of Mozilla Corporation revenue is generated from global browser search partnerships, including the deal negotiated with Google in 2017 following Mozilla’s termination of its search agreement with Yahoo/Oath”.In each geographical area, Firefox uses a default search engine (Baidu in China, Yandex in some states of the former USSR and in Turkey…). However, users have the possibility to modify this default setting. On 20 October the Mozilla Foundation published a press release highlighting the possible destabilisation of its economic model by the procedure initiated by the DoJ:« The ultimate outcomes of an antitrust lawsuit should not cause collateral damage to the very organizations – like Mozilla – best positioned to drive competition and protect the interests of consumers on the web ». 

Compared to the European case, the American complaint highlights the importance of the horizontal agreements between Google and Apple, i.e. between two of the most important digital ecosystems and the two quasi-duopolistic players in the mobile operating system market so far. The DoJ highlights the existence of agreements with Apple to make Google Search the default search engine regardless of the technical device (§45). Indeed, the agreement initially concerned Apple computers (Google Search becoming the default search engine for Safari in 2005, iPhone mobile terminals in 2007 and finally connected speakers in 2016 with the installation on Siri). Finally, the only silo that actually competes with Google on the online search market is that of Microsoft with Bing. Coopetition does not only exist within ecosystems[21]but also between ecosystems. According to the 25 October 2020 edition of the New York Times, Google’s payments to Apple are estimated to be between 8 and 12 billion a year and represent between 14 and 21% of Apple’s annual profits[22].

At this stage, it is appropriate to look at the theories of harm that are arising from the DoJ’s complaint. They are even more interesting to consider because, according to the Supreme Court, Antitrust is based on a consumer welfare prescription.  

A first competition damage may be defined as harm to subsequent innovation and to potential innovation (see §9 and §115). This damage could be all the higher, since the alleged anti-competitive envelopment practices relate to emerging ways of accessing the Internet and hence to the current developments of online research (§160). There is also a dimension of damage in terms of quality (protection of privacy, etc.) which is also not in line with the tradition of consumer welfare and which has therefore led to a significant change in the practice of American antitrust since the 1970s (§167). 

A second competitive damage may arise from strategies designed to increase rivals’ entry and operating costs. This may, for example, be the case for “free” online research. Rival companies that do not have access to data flows from the Google ecosystem or Mozilla or Apple’s Internet browsers could suffer a performance handicap due to the lack of queries. As the DoJ’s complaint points out, the only way to generate them would be to pay for them (Microsoft – § 25). This strategy would generate an additional damage: that of preventing competitors from reaching the critical size from which they could become as efficient as Google (§167).

This question highlights the critical nature of the assessment of barriers to entry and development in the application of competition laws to digital ecosystems.

The investments required to enter the search engine market are particularly high. Following the report of the Judiciary Committee, the DoJ’s complaint highlights the investment required for the indexation of pages (crawling – §20 and §22). Ensuring a quality equivalent to the one provided by Google would imply acquiring technical capacities that are difficult to achieve and, above all, benefiting from the same flow and history of queries. It can therefore be considered that this market is characterised by barriers to entry preventing a competitor from being as efficient as the dominant operator. However, this first mover advantage still needs to be qualified in competition-law terms. Unless the search engine, or rather indexing data, is considered as an essential facility and access to it is regulated, it would be necessary in the American context to demonstrate that financial investments at entry (which correspond to investments previously made by the dominant company) are barriers to entry and that Google’s advantage does not correspond to the cases referred to in the Supreme Court’s Grinnell ruling[23]:« The offense of monopoly under § 2 of the Sherman Act has two elements: (1) the possession of monopoly power in the relevant market and (2) the willful acquisition or maintenance of that power as distinguished from growth or development as a consequence of a superior product, business acumen, or historic accident ».It would be necessary to demonstrate that the advantage of Google’s search engine lies in privileged access to data related to anti-competitive practices. The notion of a level playing field is for now purely European. 

A second barrier is interesting to consider: barriers to expansion. The DoJ complaint addresses this issue through the notion of scalability. Indeed, hindering the ability of competitors to grow is preventing them from being as efficient. The complaint emphasises the importance of the 4Vs that characterise the data (volume, variety, velocity, veracity) for implementing machine learning and therefore the performance of the algorithms (§8). As the DoJ complaint points out, competitors need both fresh queries, location-based queries and long-tail queries (rare queries) to compete effectively with the dominant operator (§36). However, the admissibility of possible antitrust remedies aimed at levelling the playing field remains to be assessed.

A third potential harm to competition could stem from the consolidation of the silos constituted by the different ecosystems of the GAFAMs, in this case Google’s. This consolidation can be achieved, for example, through contractual provisions with the various complementors. The effect of these clauses may be due to the increased costs of exiting the ecosystem. The loss of revenues linked to the sharing arrangements acts as a disloyalty penalty for the members of the ecosystem. These incentives can be seen both as a sharing of the wealth created but also as an incentive mechanism for the complementors not to take decisions that could go against the interests of the keystone firm. According to the complaint, Google’s payments in this respect in 2019 amounted to USD 1 billion (§148). Even beyond the sums at stake, the incentive effect is reinforced by the procedures put in place.For instance, renouncing the pre-installation of the search engine for a single terminal means that the penalties apply to all previously sold terminals (§152). The foreclosing impact is then close to that highlighted by the European Commission in the 2009 Intel decision for loyalty discounts that could lead to exclusivity[24]. An important point specific to the functioning of digital ecosystems is that the understanding of the possible competitive harm cannot be made on a single market but supposes to consider all the markets on which the keystone player (e.g. the dominant player of each digital ecosystem) is present, on which its actual or potential competitors are present or on which it is likely to enter. A strategy conducted in a specific side of the market may have a differentiated competitive or anti-competitive impact depending on the market. A consumer-friendly strategy in each market may be detrimental to current or future competition in other markets. These may be strategies to pre-empt data flows which may participate, as noted above, in the erection of barriers to entry and/or expansion of competitors.

A fourth damage to competition may therefore result from anticompetitive foreclosure strategies. Contracts concluded with the complementors can, as we have seen, act as exclusivity contracts. Even if they only require pre-installation of Google tools and do not prevent users from changing default settings, the importance of the status quo bias well known in behavioural economics should be noted (§16). Furthermore, the DoJ states in its complaint that handset manufacturers may be obliged to pre-install all Google’s critical applications, but also to make some of them non-removable, and not to make applications competing with Google’s appear on the same screen (§55, 137).Once again, the analysis of the contractual provisions that bind the different members of the ecosystem is decisive. Regulatory market power is exercised through non-negotiable contractual clauses (Mobile Application Distribution Agreements (MADAs), AFAs (Anti-Fragmentation Agreements) and ACCs (Android Compatibility Agreements). These are “adhesion contracts” whose terms are imposed by the economically most powerful party.  The definition of contractual terms can be in this perspective the expression of a superior bargaining power. The situation is what American institutional economists at the beginning of the 20th century referred to as a scarcity transaction.One of the two parties can withdraw from the transaction while the other has no alternative but access to Android and its “essential” applications in order to enter the market. This results in a cross consolidation of the dominant position in key markets via Android and exclusivity clauses: in order to have access to APIs (which lower barriers to entry), to benefit from side payments (rent sharing) and to access Google’s “star” applications (§128, the 6 core apps are defined in §134), the complementors are obliged to accept these transaction terms. These “aids” are presented as poisons pills in the complaint (§129). Indeed, they reduce barriers to entry but increase barriers to exit and consolidate Google’s dominant position.

This theory of damage highlights the possibility of the existence of kill-zones[25]. In this case, the gatekeeper and structuring platform position enables the keystone to exclude competitors from the market. This potential capacity to undertake such an exclusionary abuse is reflected in the expression found in the DoJ’s complaint “Cutting the [competitor] air supply” (§11). The literature on killer acquisitions often makes us lose sight of the fact that the elimination of potential or emerging competitors can easily be achieved within ecosystems through barriers (absolute or relative) to market access. 

The fourth type of damage may moreover relate to these relative barriers to market access. Those relate to self-preferencing strategies. The DoJ’s arguments are very similar to those advanced by the Commission in Shopping 2017 (§30). It is Google’s gatekeeper position and its overhanging position in relation to its complementors that makes these practices possible (§170). In the logic of the DoJ, the operating rules of the Android ecosystem can also be seen as a tool of self-preferencing for Google applications. It would be possible to cite an example not included in the Android decision of 2018. It would only be possible to make phone calls using Google Assistant when the smartphone is controlled by voice (§140).

It should be noted that parts of the problem could be due to the market behaviour of the dominant company, while other aspects are directly linked to its very dominance. Firstly, when a technical standard position is reached on a market, the dominance naturally tends to be strengthened. An anecdote can illustrate historical continuities. In the 1960s lawsuit against Rank Xerox, one of the arguments was that “to xerox” had become English for photocopying. The same argument was spelled out in the 2020s complaint with the verb “to google” to refer to online searching. However, the reinforcement comes not only from the users’ habits but also from the cross-externalities specific to multisided platforms. Winner-takes-all models combined with high barriers to entry may give rise to fears of structural market failure. This is in this case the argument of the European Commission with its “new competition instrument” as proposed in June 2020.

Such an approach is not a priori conceivable in the American contexts unless the notion of monopolisation should be interpreted very broadly, and no-fault monopoly notions reintroduced (which is not the case with this procedure). The higher the barriers to entry, the less contestable the market will be and the less self-regulating the market will be. In the 1960s and 1970s no-fault literature on no-fault, the durability of the dominant position appeared as one of the reasons for structural remedies[26].

The DoJ’s requests are in part standard (behavioural injunctions), perhaps rhetorical (with an opening towards structural remedies) but potentially interesting. They open up the question of how to restore competition in markets affected by Google’s practices (§194). These reflections echo the debates on the effectiveness of European sanctions and on the proposals for radical remedies that have been put forward in the literature, apart from assets divestitures, as Michal Gal and Nicolas Petit have shown[27].

However, as in the case of the Google decisions issued by the European Commission, two questions are expected to arise in the course of the proceedings. The first question relates to the definition of relevant markets. The difficulty is even more marked in this case as we are dealing here with practices implemented simultaneously on several markets in the context of ecosystems. The theory of anti-competitive leverage from the Microsoft case of the late 1990s, where it was a question of “simply” extending a dominant position from the relevant – easily definable – market of operating systems to that – just as easily circumscribed – of Internet browsers. The second issue is, as noted in the 2017 Shopping decision and the 2018 Android decision, the characterisation of competitive harm.

If the US Antitrust sticks to the criteria it adopted in the late 1970s, i.e. the case-by-case assessment of the net effect of the practices considered on consumer welfare, the procedure may run into a pitfall. The emphasis in the Judiciary Committee’s report on damage theories that are very European in their conception shows that a decision based on Section 2 of the Sherman Act could impose a return to the criteria that were those of American Antitrust during the post-war Warren Era. Mozilla’s press release highlighted this line of defence in its last paragraph: “Unintended harm to smaller innovators from enforcement actions will be detrimental to the system as a whole, without any meaningful benefit to consumers — and is not how anyone will fix Big Tech. Instead, remedies must look at the ecosystem in its entirety, and allow the flourishing of competition and choice to benefit consumers”. 

It should also be noted that arrangements within ecosystems can be defended on the basis of efficiency (balancing the improvement in the quality and performance of services rendered to users against possible effects restricting competition) and that agreements between ecosystems are not the subject of proceedings here, which, it should be recalled, are only initiated on the basis of Section 2 of the Sherman Act.

Frédéric MARTY

CNRS – GREDEG – Université Côte d’Azur

[1]Lande R.H., (2013), “A Traditional and Textualist Analysis of the Goals of Antitrust: Efficiency, Preventing Theft from Consumers, and Consumer Choice”, Fordham Law Review, 81, pp.2349-2403.

[2]Barnett T.H. and Wellford H.B., (2008), “The DOJ’s Single-Firm Conduct Report: Promoting Consumer Welfare Through Clearer Standards for Section 2 of the Sherman Act”, Competition Policy International, https://www.justice.gov/atr/dojs-single-firm-conduct-report-promoting-consumer-welfare-through-clearer-standards-section-2

[3]Ginsburg D.H., (2010), “Originalism and Economic Analysis: Two Case Studies of Consistency and Coherence in Supreme Court Decision Making”, Harvard Journal on Law and Public Policy, 33(1), pp.217-238.

[4]A procedure could be opened in late autumn by the Federal Trade Commission against Facebook, as Reuters reported on 22 October 2020.

[5]United States v. Microsoft Corp., 56 F.3d 1448 (D.C. Cir. 1995) -“Microsoft I” et United States v. Microsoft Corp., 147 F.3d 935 (D.C. Cir. 1998) – “Microsoft II” 

[6]Ohio Republican Senator, John Sherman, originally introduced the Sherman Act of 1890 in a very specific political context. It was both a matter of a context linked to the Republican primaries and a political situation characterised by the Democrats’ attempt to link the combined economic effects of the external trade tariff and the concentration of economic power through trusts. It might have seemed more appropriate in a context of industrialisation of the US economy to tackle trusts, which appeared to be the main problem in public opinion (see Kolasky W., (2009), “Senator John Sherman And the Origin of Antitrust”, Antitrust, 24(1), Fall). The analysis of the motivations of the Senate as described by Matthew Josephson is even more direct: “something must be flung out to appease the restive masses” (Josephson M., (1934), The Robber Barons, reed. 1962, HMH Books). 

[7]See for instance the mergers ban proposal (Pandemic Anti-Monopoly Act) submitted by E. Warren and A. Ocasio-Cortez during the 2020 spring. 

[8]Lina Khan herself had participated to this committee.

[9]Marty F., (2020), « Vers une européanisation de l’Antitrust américain ? », Medium, https://medium.com/@fred_marty/vers-une-europ%C3%A9anisation-de-lantitrust-am%C3%A9ricain-3755f93a88fc 

[10]Verizon Communications Inc. v Law Offices of Curtis V. Trinko, LLP, 540 US 398 (2004) 

[11]Reiter v Sonotone Corp, 442 US 330 (1979)

[12]Bork R.H., (1978), The Antitrust Paradox: A Policy at War with Itself, New York: Free Press (2ème ed., 1993) 

[13]Steinbaum M. and Stucke M. E. (2020) “The Effective Competition Standard: A New Standard for Antitrust,” University of Chicago Law Review: 87(2).

[14]However, Chief Justice Warren, who left his name in the most progressive era of Supreme Court decision-making practice, was a Republican (former governor of California) and was appointed by President Dwight Eisenhower. Chief Justice Stone (1941-1946) had also been appointed by a Republican President (Calvin Coolidge) and was generally supportive of Franklin Roosevelt’s initiatives.

[15]Comm. Euro. Google Search (Shopping), Case AT.39740, 27 June 2017; Google Android, Case AT.40099, 18 July 2018 and Google Search (AdSense), Case 40411, 20 March 2019 (decision not yet published).

[16]Marty F. and Pillot J., (2019), “Cooperation, Dependence and Eviction – How Platform-To-Business Relationships in Mobile Telephony Ecosystems Should Be Addressed in A Competition Law Perspective?,” CIRANO Working Papers 2019s-01, January.

[17]Standard Oil Co. of New Jersey v. United States, 221 U.S. 1 (1911) 

[18]United States v. American Tel. and Tel. Co., 552 F. Supp. 131 (D.D.C. 1983)

[19]Sullivan, E. T. (2002). “The Jurisprudence of Antitrust Divestiture: The Path Less Travelled”, The Minnesota Law Review, 86, 565-623. 

[20]United States District Court for the District of Columbia, US v Microsoft Corp., Civil Action n°98-1232 (TPJ), 5 Nov. 1999.https://www.justice.gov/sites/default/files/atr/legacy/2006/04/11/msjudge.pdf

[21]Marty F. et Warin T., (2020), « Concurrence et innovation dans les écosystèmes numériques à l’ère de l’intelligence artificielle », Concurrences, 1-2020 

[22]Bloombergindiquait déjà en janvier 2016 qu’un accord entre les deux firmes pour maintenir le moteur de recherche sur les iPhones se traduisait par un paiement annuel d’un milliard de dollars. 

[23]United States v. Grinnell Corp., 384 U.S. 563 (1966)

[24]Commission européenne, décision Intel, COMP/C-3/37.990, 13 mai 2009 

[25]Kamepalli S.K., Rajan R. and Zingales, (2020), Kill Zone, NBER Working Paper, w27146, DOI 10.3386/w27146, May. 

[26]See for instance Cartapanis M. et Marty F., (2020), « La Commission européenne lance deux études d’impact préalables relatives à l’encadrement concurrentiel et régulatoire des plateformes d’intermédiation électronique » Concurrences, 3-2020, pp.80-85.

[27]Gal M. and Petit N., (2021), “Radical Restorative Remedies for Digital markets”, Berkeley Technology Law Journal, Vol. 37, No. 1, forthcoming.