I. Introduction
1. At a February 2022 webinar, the economist Cristina Caffarra noted that the recent infringement decision by the Italian Competition Authority (ICA) fining Amazon for having leveraged the dominant position it holds in the Italian market for marketplaces to favour the adoption of its own logistics service s [1] “has been described in some quarters as perhaps the first New Brandeisian decision in Europe.” Elaborating on this, Caffarra explained that some have the impression, looking at this decision, that “the efficiency (…) of logistics operators versus Amazon was not an area of major consideration here (…) The concern was about Amazon and its size [as well as] its general conduct, less so about whether others are less efficient. ” [2] Antonio Buttà, the chief economist of the Italian Competition Authority, replied that the Amazon decision was not a neo-Brandeisian case and that in a few years the ICA’s decision would have “actually [been] remembered as a rather standard self-preferencing case.” He went on to say that the Italian Amazon case was undeniably “a remarkable case that attracted a lot of international attention.”
2. Piercing through the rhetorical veil enveloping this recent exchange between these two renowned competition economists, a number of stimulating questions emerge. The first one relates to the overall perception of the neo-Brandeisian movement on this side of the Atlantic. Admittedly, pigeonholing New Brandeisians as merely anti-Big and anti-efficiency paladins (“populists,” “hipsters,” etc.) is a rather unhelpful caricature. [3] Interestingly, however, it might be symptomatic of an unjustified lack of adequate attention and in-depth discussions in European competition policy circles (and elsewhere) of the distinctive features and ideas of this still-developing [4] albeit increasingly influential US antitrust movement, especially since Tim Wu and Lina Khan, two prominent New Brandeisians, have taken leading roles in “Biden Antitrust,” as special assistant to the president for technology and competition policy, National Economic Council, and as chair of the US Federal Trade Commission respectively. The dearth of intellectual engagement with this new Atlantic antimonopoly movement might in turn be due, at least in part, to what has been an accelerated emancipation in the last decades of EU competition policy from its US century-old venerable ancestor, which on the European continent was increasingly perceived as unable to face the new and significant challenges posed by the digital transformation in particular. A related question is to what extent European competition policy’s own evolution would be hampered [5] if academia, competition enforcers and other stakeholders do not fully engage in this rejuvenated transatlantic dialogue, besides more obvious preoccupations of international competition enforcement convergence. [6] This requires, however, spelling out clearly what is “distinctive” about the New Brandeis School, also relative to other voices (“Modernists,” etc.) critical of the current antitrust enforcement, both in the US and the EU . [7] Finally, it is indeed quite interesting to ask whether the Italian Amazon case might have features that could qualify as neo-Brandeisian, moving beyond catchy characterizations that, however, do little to encourage debate and dialogue.
II. At first glance : The Inaugural neo-Brandeisian “manifesto”
3. The origin of the neo-Brandeis School cannot exclusively be traced back to conceptual and theoretical work conducted within academic walls (e.g. “Harvard,” “Chicago”). As openly acknowledged by Tim Wu, acting in his new role in the Biden administration, the “prophet” of this more recent antimonopoly movement is Barry Lynn, a US journalis t. [8] Almost a decade after Lynn first published an influential book [9] painting “a portrait of corporate power running roughshod over traditional American values,” [10] a group of participants at a 2019 conference organized by the economic department of the University of Utah in Salt Lake City [11] drafted what could be read as the inaugural neo-Brandeis Sc hool [12] “manifesto,” the so-called Utah Statement . [13] Some of the drafters of this document subsequently went on to become the influential architects of “Biden Antitrust,” such as the already mentioned Tim Wu and Lina Khan, who has been at the head of the Federal Trade Commission since June 2021. This section provides an overview of some of the core tenets of this new School as specifically enshrined into the Utah Statement.
1. Acknowledging the essential link between antitrust law and democracy
4. The Utah Statement’s preamble comprises key principles articulating the relationship between antitrust law and democracy. It states that “[s]ubjecting concentrated private power to democratic checks is a matter of constitutional importance.” The idea that private power should be subject to “democratic checks” has a long and honourable tradition in Western societies. As famously recalled by former US President Franklin D. Roosevelt, “the liberty of a democracy is not safe if the people tolerate the growth of private power to a point where it becomes stronger than their democratic state itself.” [14] The well-known risk highlighted here is that, when left unchecked, concentrated private power can undermine democracy itself. [15] Following Roosevelt, Timothy Wu sees antitrust law “as one antidote to danger of private economic power that might rival public power .” [16] With regard specifically to the analogy between antitrust law and constitutional law, Wu argues that, as the Constitution is comprised of a system of checks and balances, antitrust law represents a new kind of limit in the form of a “check on private power, by preventing the growth of monopoly corporations into something that might transcend the power of elected government to control.” [17] Moreover, by passing the antitrust laws, according to Wu, lawmakers “made a critical, indeed Constitutional choice in industrial and national policy,” as “the nation rejected a monopolized economy and chose repeatedly over the decades to preserve its tradition of an open and competitive market.” [18] The preamble to the Utah Statement further articulates the relationship between antitrust law and democracy by stating that “[t]he protection of fair competition is a means to a (…) democratic society.” Conversely, “[e]xcessive concentration of private economic power breeds antidemocratic political pressures and undermines liberties.” [19] As further elaborated by Lina Khan, the concentration of economic power also consolidates political power “through enabling a small minority to amass outsized wealth, which they could then use to influence government,” as well as “permitting ‘private discretion by a few in the economic sphere’ to ‘control[] the welfare of all,’ undermining individual [20] and business freedom. ” [21]
2. Stating that the focus on consumer welfare as the sole goal of antitrust is not consistent with the legislative intent of the US antitrust laws
5. The Utah Statement frontally challenges the correctness of the argument that “Congress designed the Sherman Act as a ‘consumer welfare prescription.’” [22] To some extent, this neo-Brandeisian argument follows from the above-discussed point about the link between antitrust laws and democracy as historically enshrined into the antitrust laws. The “menace to republican institutions themselves” [23] was clear to lawmakers, who recognized that concentrated market power left unchecked could “subvert democratic decision-making and make decisions of public import.” [24] Moreover, the framers of the antitrust laws wanted to prevent large unjust wealth transfers from producers and consumers to large firms, [25] preserve open markets and enhance freedom of opportunity for new and small firms. [26] As Lina Khan wrote in 2018, “[t]he antitrust laws were passed against the backdrop of growing economic concentration and the rise of industrial trusts. Railroad barons had captured control over critical transportation networks, while monopolistic corporations dominated sectors ranging from oil and sugar to tobacco and steel. A small group of investment bankers, meanwhile, used their control of the credit system to further roll up industries, directing the fate of the econom y.” [27]
6. Famously, it was Robert Bork’s radically different reading of the legislative history of the Sherman Act that had a very influential impact on the Supreme Court itself. [28] Citing Robert Bork’s doctrinal work, the Supreme Court in 1979 declared that the “Congress designed the Sherman Act as a ‘consumer welfare prescription,’” [29] opening the gates for a wholesale reshaping of antitrust enforcement based on this specific, and, according to the neo-Brandeisians, fatally wrong understanding of the lawmakers’ intent in framing those laws. This has led “both enforcers and courts to focus mainly on promoting ‘efficiency’ on the theory that this will result in low prices for consumers.” [30]
3. Recognizing that the protection of fair competition is also an instrument for the distribution of power
7. The previous subsection explained how, according to the New Brandeisians, the genuine goal of antitrust has been wrongfully supplanted. Instead, as the Utah Statement spells out, “[t]he protection of fair competition is (…) an instrument for both the creation of opportunity and the distribution of wealth and power.” In this respect, it is also important to note that the neo-Brandeisians are putting forth a specific descriptive and normative account of power. Importantly, economic power is recognized as inextricably political. As Lina Khan puts it, “[p]ower in industry is the power to steer outcomes. It grants outsized control to a few, subjecting the public to unaccountable private power—and thereby threatening democratic order.” [31] Concentrated market power is a problem of domination, hampering our ability to self-govern. [32]
8. To the contrary, the consumer welfare’s focus on efficiency meant that even monopoly power ceased to be seen as threatening per se but instead viewed as beneficial, to the extent that it could be “both a product of metaphysical ‘free market’ forces and a key stimulus for them” and even “revered as a vital guardian of innovation.” [33] This is particularly significant “[g]iven the broad and general terms in which Congress wrote the antitrust statutes,” which means that “their application turns critically on notions of economic power.” [34] This observation alone puts under a rather different light the legal standards and the “more technical rules” with an alleged scientific, strongly economics-based pedigree that have emerged from the antitrust laws’ efficiency reorientation. Far from being politically neutral, the shift, according to Lina Khan, ushered in “a new ideology.” [35]
4. Requiring the reversal of the consumer welfare framework
9. While New Brandeisians, as seen above, oppose the view that the “Congress designed the Sherman Act as a ‘consumer welfare prescription,’” the Utah Statement is not explicit concerning the preferred, alternative antitrust standards. It is clear, however, that antitrust enforcement should move away from trying to measure specific outcomes per se, such as for instance lower prices. [36] Thus, according to Khan, being “for diversity and access to markets [and] against high concentration and abuses of power,” [37] “antitrust law was structured to preserve a set of structural conditions (competition) as a way of promoting a set of outcomes and principles.” [38] The core argument here is that competition refers to a process. Instead, to the extent that the focus on consumer welfare places competition in the service of efficiency, this “refers to an economic outcome, and is silent on the means by which it is achieved.” [39] Therefore, the “grotesque distortion of the antitrust laws that Congress passed” almost ironically stamped antitrust laws “with a value that is, in many ways, deeply antithetical to the goal of competition.” [40] The rejection of the outcome imperative and of its consequences for antitrust enforcement is hereto radical. As Khan writes, “[c]ontrary to how critics portray the New Brandeisians, this new school of thought does not promote using antitrust law to achieve a different set of social goals—like more jobs or less inequality. Doing so would replicate a key mistake of the Chicago School : overriding a structural inquiry about process and power with one that focuses on a narrow set of outcomes. Refocusing antitrust on structures and a broader set of measures to assess market power can return the law to focusing on the competitive process.” [41]
10. Notably, the focus on the specific economic outcomes distilled by the consumer welfare framework went hand-in-hand with the successful occupation of the field of competition economics by neoclassical orthodox thinking, while influences by alternative economic research agendas remained marginal at best. Supported by complex modelling as well as by the occasional judicial misinterpretations thereof, “[v]ertical coercion, vertical restraints, and vertical mergers” in particular have enjoyed a “presumption of benefit to the public” that the New Brandeisians now want to reverse. [42] The broader critique in this respect, however, is that relying on welfare-based tools focusing on narrowly defined outcomes would leave much harm caused by undue market power unaddressed, such as on “workers, suppliers, [43] innovators, and independent entrepreneurs—all harms that Congress intended for the antitrust laws to prevent.” [44] Moreover, those tools do not adequately reflect that “[t]he more pressing antitrust issues regarding the big tech platforms concern how their control over and access to data, as well as their integration across multiple lines of business, enables them to entrench their dominance in ways that undermine competition.” [45]
11. Given the present state of antitrust policy in the US, according to the New Brandeisians, well advised albeit modest changes within the same consumer welfare framework would not suffice to bring about “a strong revival of antitrust, and a return to its anti-monopoly roots.” [46] What the movement deems therefore necessary is to strike “at the heart of some of the theories that sustain the current framework” [47] in order to effectively “reckon with the ideological frame that cripples our enforcement tools.” [48]
12. While abandoning the consumer welfare standard is considered a must, the New Brandeisians are also perfectly aware that effectively dismantling this standard would require a complex, encircling strategy. Thus, for instance, the Utah Statement states that “[n]etting harms and benefits across markets, parties, or classes should not be a method for assessing anticompetitive effects.” The argument here is that the logic of the existing framework, which maps onto the “consumer welfare” approach, focuses on net effects while ignoring structure. The netting is rejected entirely—that is, not only in cases in which at least some economists would likely agree, such as in the Second Circuit’s analysis in American Express. [49] Moreover, the primacy of highly theoretical approaches, as proposed in particular by “modern” Industrial Organisation’s economists, is challenged. [50] Thus, the Utah Statement advocates that “[h]arms demonstrated by clear and convincing evidence or empirical study should never be ignored or discounted based on theories that might predict a lack of harm” and “[c]lear and convincing evidence of anti-competitive intent should be taken as a presumptive evidence of harm.”
13. Finally, according to the New Brandeisians, the preference for false negatives (not finding violations when the conduct harmed competition) over false positives (finding violations when the conduct did not harm competition) directly translates into a preference for under-enforcement. The neo-Brandeisians propose two main reasons why this should be corrected. First, this preference is based on a theoretically wrong understanding of market dynamics, such as that “[i]f the court errs by permitting a deleterious practice (…) the welfare loss decreases over time” because “[m]onopoly is self-destructive,” given that monopoly prices attract entry that will compete away market power. [51] Instead, “If the court errs by condemning a beneficial practice, the benefits may be lost for good.” [52] The overwhelming evidence of durable market power and the successful erection of entry barriers to exclude new rivals are in direct contradiction with this assumption. Second, while “[t]he preference for false negatives is offered as a way to guide enforcers through uncertainty,” [53] “concerns that enforcers and courts will decide improperly should be addressed through clarifying legal standards and minimizing the role of judicial discretion, not through requiring that enforcers and courts simply err on the side of not acting.” [54] This is not inevitable, as it merely reflects “the pegging of liability largely to welfare effects while neglecting the competitive process.” [55] What New Brandeisians instead suggest is that “present indicia can be used to structure analysis so as to encompass gradations of error analysis, rather than a binary choice between false positives and false negatives.” [56]
5. Creating space for the emergence of a new antitrust standard
14. The active and successful dismantling of the consumer welfare standard should create the necessary space for the emergence of a different antitrust paradigm, [57] which the New Brandeisians would want more in line with the US lawmakers’ legislative intent. In this respect, the Utah Statement relies on the creation of antitrust rules “through case development, agency rule-making, and legislation.” Moreover, it is stated that “[t]he States, the laboratories of economic experimentation, are a critical vanguard of enforcement efforts.” [58] The New Brandeisians thereby make clear that the coveted paradigm shift should be the result of a complex, experimental and democratic endeavour. [59]
15. While, accordingly, the Utah Statement is not directly suggesting an off-the-shelf, alternative antitrust standard, it is acknowledged by leading proponents like Lina Khan that “[s]ince Chicago introduced a new normative conception of antitrust, challenging its dominance will require offering an alternative normative vision of what the law stands for and how it can be operationalized.” [60] Thus, Tim Wu proposes replacing the consumer welfare standard with the “competitive process standard.” The latter focuses on whether “the conduct at issue [is] actually part of the competitive process, or [it is] enough of a deviation as to be unlawful.” The use of this standard would “produce a fact-intensive inquiry and an analysis of the pro-competitive justifications offered by the defendant [that would] yield a body of rules and standards which enforcers and judges can use to protect competition.” The main advantage would be that potential threats to the competitive process become far more obvious, as “[t]he allegation is that a powerful or unethical firm is seeking to disable the process of competition on the merits” without the need to assess whether consumer welfare has been served or not. [61]
6. Revisiting legal precedents under Section 2 of the Sherman Act
16. New Brandeisians are also perfectly aware that steering the course of US antitrust policy away from the alleged grip of the consumer welfare standard on the judiciary would require directly tackling court precedents. Thus, the Utah Statement makes concrete proposals [62] to change the antitrust standards, such as with regard to predatory pricing (“[t]he Brooke Group test for predatory pricing (…) should be overruled”), monopoly leveraging (“[t]he Berkley Photo standard for establishing monopoly leveraging should be restored”) and refusal to deal (“[t]he essential facilities doctrine should be reinvigorated for dominant firms that deny access to critical infrastructural services”) and price squeeze (“[t]he LinkLine doctrine holding that price squeeze allegations fail as standalone Section 2 claims should be overruled”). As already recalled above, these and other antitrust rules “should be created through case development, agency rule-making, and legislation.” Moreover, “[c]lear and convincing evidence of anti-competitive intent should be taken as a presumptive evidence of harm,” “[h]arms demonstrated by clear and convincing evidence or empirical study should never be ignored or discounted based on theories that might predict a lack of harm” and “[t]he determination by the antitrust agencies of relevant market definitions should receive judicial deference.”
III. The Italian Amazon infringement decision through a New Brandeisian lens
17. Lina Khan argued in her groundbreaking 2016 article that “Amazon’s business strategies and current market dominance pose anticompetitive concerns that the consumer welfare framework in antitrust fails to recognize.” [63] Part of the international interest that the Italian Amazon infringement decision attracted, as reported by the ICA’s chief economist, [64] might indeed relate to the antitrust standard that it concretely applied. The Utah Statement, as seen in the previous section, promotes the revision of a whole array of legal precedents in the area of monopolization. [65] Writing together with Sandeep Vaheesan, Lina Khan already made clear that “[t]he antitrust agencies and courts should look to European Union abuse of dominance law for a model to emulate.” [66] Based on the previous analysis of the Utah Statement, it could be argued that some elements of the recent Italian Amazon decision would already qualify as New Brandeisian. What can be seen as new and even “revolutionary” in the US today might indeed be “normal” in the EU. When Lina Khan and other New Brandeisians call to revitalize US antimonopoly traditions for the digital age, [67] their gazes are often directed towards cases and investigations against Big Tech on the other side of the Atlantic.
18. The Italian Amazon decision [68] is based on specific evidence that Amazon has, first, defined a set of functionalities enabling third-party sellers to obtain greater visibility of their offer and an improvement in their sales performance on Amazon’s marketplace. These constituted a set of exclusive and irreplicable advantages [69] relevant to third-party sellers’ profitability on Amazon.it. Second, the infringement decision relies on evidence that Amazon has made access to those functionalities conditional on subscription by third-party sellers to “Fulfillment by Amazon” (FBA), the integrated logistics service offered by Amazon itself. This encouraged the use of FBA by third-party sellers regardless of the characteristics of the service and its ability to meet their specific needs. According to the ICA, this is an abuse of a dominant position in the national market for intermediation services on e-commerce platforms (marketplaces) implemented by Amazon in breach of Article 102 of the Treaty on the Functioning of the European Union (TFEU). With regard to the Italian market for marketplace intermediation services, the Italian competition enforcer considers that Amazon holds a position of “super dominance.” [70] By engaging in the above-mentioned practices, Amazon was unduly promoting its own logistics service to the detriment of competing services. The effect was increasing Amazon’s share in the related national market for e-commerce logistics services (leverage effect), as well as strengthening its dominance in the core market where it supplies intermediation services on e-commerce platforms. On the one hand, competing logistics operators were deprived of a significant part of the demand of retailers active on Amazon’s marketplace, making it harder for them to compete and therefore preventing the development of effective competition in that market. On the other hand, Amazon made the simultaneous sale on several platforms (so-called multihoming) more costly for retailers active on Amazon.it, causing them to decrease their offerings on competing marketplaces. According to ICA’s findings, this strategy was not justified by any reason of efficiency in the management of the platform, which was objective, demonstrated and otherwise unattainable. Amazon was therefore fined €1,128 billion for a breach of Article 102 TFEU. In addition, the Authority imposed significant behavioural obligations on Amazon, to be monitored by a trustee.
19. With regard specifically to self-preferencing, Lina Khan, in her writings, pointed to Amazon’s aggressive expansion into multiple business lines, with the result that, in many instances, the Big Tech’s rivals were also its customers. She proposed the example of “retailers that compete with [Amazon] to sell goods [who] may also use its delivery services.” “At a basic level,” she noted, “this arrangement creates conflicts of interest, given that Amazon is positioned to favor its own products over those of its competitors.” [71] In a 2018 contribution, Khan pointed to the real-world example of Amazon prioritizing its own brands of apparel “on and restricts rivals’ access to certain prominent promotional areas on its website.” As a source of inspiration for how to tackle this type of self-preferencing practice, Khan directly referenced [72] the EU Google Shopping decision by the European Commission, [73] which has been recently confirmed by the General Court. [74]
20. To some, labelling the Italian Amazon case as New Brandeisian would be perceived as disparaging, in the sense of being annoyingly political, sanctimoniously anti-Big per se, economics-illiterate, oblivious of flagrant efficiencies, etc. As a matter of fact, the New Brandeisians fully acknowledge that the antitrust laws, as they deal with the power allocation and use in society, have a strong political dimension. As seen in subsections II.1 and II.2 above, according to the Utah Statement, the US antitrust laws have a distinctive value-based content (“democracy,” “dispersion of power,” etc.) that the consumer welfare standard has wrongfully replaced by a different political agenda. No less political, for instance, is following the lodestar of “legitimacy and accountability,” [75] if it translates into the choice of non-action disguised by a “scientific” preference for false negatives.
21. Obviously, New Brandeisians are not anti-Big as such, as was instead glibly insinuated for instance by the economist Cristina Caffarra during the webinar that inspired this contribution. [76] Thus, this School recognizes that certain network industries and others tend naturally towards monopoly. In such cases, they do not necessarily suggest breaking these firms up, designing “a system of public regulation that prevents the executives who manage this monopoly from exploiting their power [as well as] to ensure that executives face the right incentives to provide the best service possible to everyone who relies on the monopoly to sell or to buy a particular product or service.” [77] Similarly, this School is also far from promoting the rejection of an evidence-based approach, as some “antitrust reactionaries” [78] would seem to imply. On the contrary, as the Utah Statement affirms, the focus should indeed be on “clear and convincing evidence or empirical study.” [79]
22. While being political in the New Brandeisian sense does not imply “demonizing the elites” nor indulging in other overt moralizing inclinations, this School considers that “forms of economic power are not inevitable but can be restructured” to promote power decentralization and dispersion. [80] Following Lina Khan, “[e]ven under existing statute and case law, a different theory of power would illuminate under-used legal authorities and could encourage antitrust authorities to bring cases that reach forms of harm neglected by the current regime.” [81] In this respect, the Italian Amazon decision could be New Brandeisian in a more profound sense than generally perceived. First, as seen above, the ICA refers to Amazon as holding a position of “super dominance” in the Italian market for marketplace intermediation services, as confirmed on the basis of different indicators. [82] Surely, in the context of online platforms, we are confronted with extreme forms of economic power. In the EU, this realization is also triggering unprecedented regulatory responses, such as in particular the Digital Markets Act, [83] currently finalized by the co-legislators. Second, the Italian decision might also be seen as laser-focusing the competition policy analysis on specific ways in which extreme manifestation of economic can be exercised in the digital era, giving rise to the ability and incentives to cause harm to consumers, complementors and competition as a whole, both of an exclusionary and exploitative type, [84] for which well-tread theories of harm offer only limited guidance. [85] New Brandeisians would definitely support putting under the spotlight any type of “[h]arms demonstrated by clear and convincing evidence.” [86]
IV. Conclusion
23. The conclusion is that the recent Italian Amazon infringement decision contains some New Brandeisian elements. Seen from an EU competition policy perspective, however, those elements are not entirely new nor alien. Hence, this is not the first New Brandeisian antitrust decision in Europe.
