Ever since Brunswick was decided almost 50 years ago, competition law in the United States has been thought to protect the interests of the competitive process over the interests of competitors. That paradigm is being challenged today, however, with attacks on the consumer welfare approach, often articulating “self-preferencing” by dominant firms as a basis for legal challenge. But promoting one’s own products over those of rivals is the essence of competition, and putting that procompetitive activity at legal risk can only harm competition and the economy as a whole—at least without careful legal rules that distinguish the harmful from the benign. The necessary analysis for distinguishing the two has often been lacking. What is clear, however, is that requiring firms to pull their competitive punches just to benefit rivals makes no sense. The purpose of this paper is to suggest an approach that allows condemnation of truly anticompetitive “self-preferencing” while recognizing that simply promoting one’s own wares over those of rivals should be encouraged, even for dominant firms.
Self-preferencing, loosely defined, has been seen by some antitrust authorities as inherently anticompetitive, particularly in digital markets. But because much self-preferencing is supported by common-sense procompetitive justifications, including competition “on the merits,” self-preferencing alone cannot sensibly be viewed as a standalone monopolization offense. Doing so would sacrifice these benefits for no sound competitive purpose. Self-preferencing has been targeted by regulators most often when the conduct is seen to be associated with activities such as refusals to deal, tying, bundling, or consumer deception. These aspects of unilateral conduct, however, are already recognized as potential offenses under existing antitrust or consumer protection laws and should therefore be conceptually distinguished from simple self-preferencing—elevating one’s own products over those of rivals. The recent attacks on simple self-preferencing are largely inconsistent with the longstanding judicial treatment of the same category of conduct—albeit prior to the emergence of the large Internet platforms—without any policy basis for ignoring what has been considered settled law. Regulators are seeking to protect rivals by developing new rules for platform business models without regard for the likely impact on consumers.
In many countries and jurisdictions, the self-preferencing buzzword is at the center of legislative and regulatory scrutiny of the market behavior of dominant firms. The term generates far less enthusiasm in U.S. court rulings. The U.S. courts did not even mention the term “self-preferencing” until 2021: a Lexis search for U.S. cases for the term “self-preferenc! and antitrust” resulted in only six cases and among them only three cases discussed self-preferencing under the antitrust laws—Dreamstime, Rumble, and Epic Games. But even in these cases, the discussion of the lawfulness of self-preferencing was largely lacking.
This relative absence of allusions to “self-preferencing” in U.S. antitrust case law appears, however, to be more a matter of nomenclature than an actual blind spot. Viewing self-preferencing as simply preferential treatment granted by a platform to its own products and services, it becomes easier to find echoes in prior U.S. antitrust jurisprudence—a small set of cases that by and large treats what is now often labeled as self-preferencing behavior as competition on the merits.
In Bayou Bottling, Inc. v. Dr Pepper Co., a local Pepsi bottler complained that the local Coke bottler (an alleged 75%-80% share monopolist) would not allow Pepsi into vending machines or coolers that the Coke bottler supplied or serviced. In choosing to use these machines only to facilitate the sale of Coca-Cola products, the defendant’s behavior could clearly be described as self-preferencing in modern parlance. In its ruling, the circuit court opined: “Without anything more, these practices are not barred by the antitrust laws. They are competitive acts. It ought to be apparent that ‘a monopolist’s right to compete is not limited to actions undertaken with an altruistic purpose. Even monopolists must be allowed to do as well as they can with their business.’”
Similarly, in Christy Sports, LLC v. Deer Valley Resort Co., the plaintiff had long operated the ski shop at Deer Valley Resort, but the defendant resort (DVRC) elected to evict the plaintiff and operate the ski shop itself. The court assumed monopoly power but ultimately rejected the plaintiff’s claim, saying: “The Sherman Act does not force DVRC to assist a competitor in eating away its own customer base, especially when that competitor is offering DVRC nothing in return.”
Perhaps the most significant U.S. government position regarding “self-preferencing” was the Federal Trade Commission’s 2013 Statement explaining its unanimous decision to close its Google investigation. There, in a 5-0 ruling, the Commission found no violation from Google’s placement of Google Shopping results over third-party comparison-shopping engines on google.com, as well as other claims of what is now called self-preferencing. The Commission acknowledged that the conduct at issue had lowered the rankings of competitor websites, but nevertheless concluded:
Product design is an important dimension of competition and condemning legitimate product improvements risks harming consumers. Reasonable minds may differ as to the best way to design a search results page and the best way to allocate space among organic links, paid advertisements, and other features. And reasonable search algorithms may differ as to how best to rank any given website. Challenging Google’s product design decisions in this case would require the Commission—or a court—to second-guess a firm’s product design decisions where plausible procompetitive justifications have been offered, and where those justifications are supported by ample evidence. Based on this evidence, we do not find Google’s business practices with respect to the claimed search bias to be, on balance, demonstrably anticompetitive, and do not at this time have reason to believe that these practices violate Section 5.
More recently, although the “self-preferencing” label was not used, the court in United States v. Google LLC granted summary judgment in favor of Google on the claim that Google was unlawfully favoring its own specialized “vertical” websites (such as shopping or hotels) over those of rivals. The court concluded that the state attorneys general, led by Colorado, had failed to prove that any such “favoring” had anticompetitive effects in the relevant markets alleged, and that the states’ speculation was insufficient to carry their burden of proof.
The landmark antitrust case of the digital age, United States v. Microsoft Corp., also involved elements of what some may now term self-preferencing; but the opinion in fact condemned only those actions designed to exclude rivals—where the exclusion made no economic sense for Microsoft but for the exclusion of rivals. In the lawsuit, some of the alleged anticompetitive conduct was based on Microsoft’s contracts with original equipment manufacturers, which made Internet Explorer the exclusive browser to be pre-installed (thus prohibiting the pre-installation of Netscape), prevented OEMs from removing any “desktop icons, folders, and Start menu entries,” and had the effect of thwarting the distribution of rival browsers that many users preferred. The court also found unlawful Microsoft’s agreements with Internet access providers or online services (such as AOL) that largely prevented their use of any browser other than Internet Explorer and agreements Internet service vendors requiring them to use only IE in any software development. All this took place in an era where the preclusion of pre-installation was close to exclusive dealing; downloading Netscape on a 14.4 kbs modem could take an hour.
The agreements between Microsoft and computer manufacturers (and similar agreements with Internet service vendors and online services) did little to improve Microsoft’s own product and were designed principally to exclude rival browsers, mainly Netscape. It was not a manner of simply “preferring” IE; it was a (successful) strategy to prevent access to Netscape and thus inhibit competition in ways that did not involve any enhancement to Windows or the quality of IE. No part of Microsoft really departed from Bayou Bottling from decades earlier or was inconsistent with Christy Sports, which came afterwards. The case said nothing about simply promoting one’s own wares over rivals.
The UK Streetmap decision is largely consistent. There, the British court was asked to restrain Google’s ability to preferentially treat its own maps product, Google Maps. A competitor to Google Maps, Streetmap, complained that
by the visual display at or near the very top of its SERP [search engine results page] of a clickable image from Google Maps, and no other map, in response to certain geographic queries, and the consequent position in the market for online search and online search advertising, Google was abusing its dominant position in the market for online search and online search advertising.
The British court found, however, that Google’s preferential treatment of its online map product was unlikely to give rise to an anticompetitive foreclosure. Among other reasons, the court noted that “although Google Maps is the only online map to benefit from a visible thumbnail, the Google SERP . . . include[d] clickable links to other relevant online maps; and there is no particular difficulty for a user to click on those blue links.” The court’s finding in Streetmap that Google users would have experienced little inconvenience in switching from Google Maps to competing products distinguishes Streetmap from Microsoft and makes it more consistent with Bayou Bottling and Christy Sports. But in rejecting the claim based on the availability of alternatives, the court did not reject the idea that simply favoring one’s own product could be a violation.
By contrast, Google’s promotion of Google Shopping in its general search engine faced a stiffer challenge and opposite result in continental Europe. The European Commission argued that Google abused its market dominance as a search engine by giving an advantage to its own shopping results and demoting competitors’ comparison shopping services in its search results. The European Commission found the conduct anticompetitive on the grounds that it had the potential to foreclose competing comparison shopping services and was likely to reduce the ability of consumers to access the most relevant comparison shopping services. On appeal to the General Court, the Commission’s decision was upheld in part. The mere “special display and positioning” of the platform’s own products and services was not in and by itself deemed abusive. What was ruled as illegal was Google’s demotion of results from competing comparison services by means of adjusted algorithms.
As noted, however, on the same facts, the U.S. Federal Trade Commission found that Google’s promotion of its own shopping site and its concurrent demotion of comparison-shopping sites benefited users. It concluded that “Google likely benefited consumers by prominently displaying its vertical content on its search results page,” that “Google would typically test, monitor, and carefully consider the effect of introducing its vertical content on the quality of its general search results, and would demote its own content to a less prominent location when a higher ranking adversely affected the user experience,” and that “data showing how consumers reacted to the proprietary content displayed by Google also suggest that users benefited from these changes to Google’s search results.” The EC’s contrary ruling unambiguously favored the interests of Google’s comparison-shopping competitors over those of consumers.
If any common ground exists between the General Court’s Google Shopping decision, Streetmap, and the U.S. cases, it is found in the courts’ tolerance of a firm’s right to promote and preference its own products regardless of the firm’s monopolistic or dominant status. The disagreement is regarding the impact on competitors. The uplifting of one’s own products and services in many cases, however, necessarily means the demotion of rivals. To treat these impacts differently makes little practical sense without a good test to distinguish the anticompetitive from the benign.
The recent EC proceedings against Amazon demonstrate this difficulty. On September 20, 2022, the EC preliminarily found that, as a dominant online marketplace for third-party sellers, Amazon abused its dominant position in breach of Article 102 of the Treaty by: (1) relying on non-public sales data of sellers active in its marketplace to adjust its own retail offerings; and (2)
artificially fav[o]ring its own retail offers and offers of marketplace sellers that use Amazon’s logistics and delivery services (the so-called ‘Fulfilment by Amazon’ or ‘FBA’ services), to the detriment of other marketplace sellers and consumers, when (i) selecting the single prominently displayed offer on Amazon’s product detail page (the winner of the ‘Buy Box’); and (ii) enabling sellers to offer products to users of Amazon’s loyalty program[] (the ‘Prime program[]’) under the Prime label.
Here again, the EC seems to have equated promoting and “preferencing” one’s own product with harming marketplace sellers, while just presuming harm to the consumers. Although the use of rival’s data arguably makes this case different from pure self-preferencing, it is hard to see why using a seller’s data from transactions on one’s own platform would be problematic, at least when the collection and use of the merchant’s data is stipulated in agreements between the merchants and the platform or online marketplace.
The EC resolved the two proceedings after extracting commitments from Amazon to not use data it collects from sellers on its platform to compete against the sellers and to not discriminate against sellers that do not use Amazon’s logistics and delivery services. In a sense, the two proceedings ended in a cliffhanger because Amazon’s commitments are set to expire in a few years and because the EC may challenge similar practices by other firms that end up in European courts. However, the recent case filed by the FTC against Amazon, which advances similar claims, seeks similar relief with no expiration date.
Competition authorities in some member states have already moved aggressively in targeting self-preferential conduct. Following competition concerns by the UK authorities, for example, Amazon offered commitments not to use the data generated through transactions on its websites to give an edge to its own retail business that compete against third-party sellers that use Amazon. The Polish Competition Authority, as another example, recently imposed a huge fine on Allegro, Poland’s dominant online shopping platform, for using its own algorithm and consumer data to boost the sale of its own wares and to position them more prominently on its website compared to the merchandise of third-party sellers who use Allegro’s platform.
The current U.S. DOJ and FTC appear now to be taking the EC approach. In the July 2023 draft update of their Merger Guidelines, the U.S. agencies say:
The Agencies protect competition on a platform in any markets that interact with the platform. When a merger involves a platform operator and platform participants, the Agencies carefully examine whether the merger would create conflicts of interest that would harm competition. A platform operator that is also a platform participant has a conflict of interest from the incentive to give its own products and services an advantage against other competitors participating on the platform, harming competition in the product market for that product or service. This problem is exacerbated when discrimination in favor of a product or service would reduce access to distribution for rivals in the participants’ market and deprive rivals of network effects in the platform market, both extending and entrenching a dominant position.
By labeling a platform operator’s participation in its own platform a “conflict of interest,” the U.S. competition agencies would effectively prevent (or at least inhibit) platform operators from competing with platform participants. As discussed more in detail below, this trend of preferring competitors to the process of competition has nothing to commend it.
The history of monopolization jurisprudence’s conduct requirement is a history of finding the delicate balance between the anticompetitive and procompetitive effects of the monopolist’s conduct. As the recent regulatory and legal scrutiny of self-preferencing has largely focused on digital markets, legal analysis often pivots on how to properly measure the degree of anticompetitiveness of the monopolist’s conduct in these markets and how to credit the potentially offsetting procompetitive effects of the same conduct.
Digital markets pose a challenge to the legal analysis. In two-sided or multi-sided markets, users and consumers on one side often pay little to nothing (apart from the cost associated with user attention) for their use of digital products. On the other side, platforms may be compensated directly by advertisers and third-party merchants; the question for them is whether the platform provides a positive return on investment. Because of these factors, the degree of competition in a given digital market cannot be easily gauged by standard metrics such as price. The alternative measures include output effects, continuing innovation, and sustained level of investment. In multi-sided markets, output tends to be the best measure. But output must be measured properly, which can be difficult.
The argument for condemning digital self-preferencing stems from the idea that some Internet firms are so central that their services may be deemed essential facilities, basically public utilities. In other words, the drive to outlaw self-preferencing among these firms really stems from the drive to turn digital platforms into common carriers that are subject to utilities-style regulation. In turn, some commentators have cautioned that this threatens to reduce digital platforms’ incentive to invest in consumer welfare-improving innovations.
U.S. antitrust law went through a phase of utilities-style regulation from the 1880s through the early 1970s, some of which of course remains in effect today. Prior to the deregulation movement in the late 1970s, several important industries, mainly utilities, were considered prone to market failures. In these industries, administrative agencies were created to “oversee economic functioning, particularly prices, costs, and entry.” This type of regulation was later widely criticized for “distort[ing] firms’ incentives and reward[ing] inefficiency rather than reduced costs and innovation.” “Significant criticisms of the costs and market distortions that accompanied regulation prompted serious review of regulatory regimes . . . and persuade[d] policymakers to move toward deregulation in almost all regulated markets.” Many industries, such as transportation and communications, were significantly deregulated as a result.
Later, in 2004, the Trinko decision essentially rejected the utilities-style approach to antitrust enforcement. The Court there made clear that “[f]irms may acquire monopoly power by establishing an infrastructure that renders them uniquely suited to serve their customers. Compelling such firms to share the source of their advantage is in some tension with the underlying purpose of antitrust law, since it may lessen the incentive for the monopolist, the rival, or both to invest in those economically beneficial facilities.”
Condemning a leading firm’s promotion of its products in an adjacent market has aspects of the same utilities-style regulation Trinko rejected. To the extent that the firm needs to and continues to make investment in innovation in the second market, the imperative to protect the investment and to encourage future investment is both an economic rationale for policymakers and regulators not to punish self-preferential conduct in adjacent markets, and a basis for a legal defense against any suggestion that the firm’s conduct is not conducive to competition in these markets. In the Streetmap case, for example, the UK court took note that Google’s “presentation of a thumbnail map on the SERP in response to a geographic query was a technical ‘efficiency”’ and noted that Google can legitimately improve its search product.
The FTC’s 2013 approach and the Microsoft decision provide a useful current guide to the analysis going forward. The Commission focused on whether the self-preferential conduct improved the quality of a product and user experience and balanced that against the potential for anticompetitive foreclosure. In reviewing some vertical websites’ allegations that Google “prominently displayed Google vertical search results in response to certain types of queries, including shopping and local,” the FTC decided that the key issue “was to determine whether Google changed its search results primarily to exclude actual or potential competitors and inhibit the competitive process, or on the other hand, to improve the quality of its search product and the overall user experience.” The FTC found that, “in the main, Google adopted the design changes that the Commission investigated to improve the quality of its search results, and that any negative impact on actual or potential competitors was incidental to that purpose” and that “these types of adverse effects on particular competitors from vigorous rivalry are a common byproduct of ‘competition on the merits’ and the competitive process that the law encourages.” Microsoft is entirely consistent. The decision condemned exclusionary agreements, not self-preferencing, and made clear that conduct on one’s own platform should not be condemned absent proof that the conduct made no economic sense but for the exclusion or marginalization of rivals.
One might reasonably ask whether even the 2013 FTC and Microsoft balancing approaches go too far. What if a company legitimately tries to create a new and better product or feature that also has the effect of making it harder for rivals to compete effectively but fails because the new product is not better at all? Regulatory second-guessing of the company’s behavior might provide a short-run benefit in the specific matter at hand, but would also be a signal to the rest of the world that, if the new product is not in fact “better,” it could be condemned as an antitrust violation. That could have a significant chilling effect on new investment incentives. Courts should continue to be wary of assuming the job of speculating about the degree efficiency gains and benefits of new products. If balancing is to be done at all, it might be best to do it with overweight on the side of innovation.
This notion that it is not unlawful to promote one’s product in an adjacent market in order to improve the actual or perceived quality of a dominant product in the primary market is important. Absent evidence that the conduct is designed solely or primarily to disadvantage rivals, product improvements typically evidence continued investment and innovation, a strong procompetitive effect of the “self-preferential” conduct. This is especially true of product design decisions. Courts properly have been reluctant to interfere with such decisions absent clear evidence of anticompetitive effects.
The European Commission, unfortunately in our view, habitually rejects any efficiency or consumer benefit justifications for this same variety of self-preferencing. For example, the EC rejected in the Google Android case the efficiency justifications for prohibiting phone OEMs from “forking” Android, i.e., from selling a phone as an Android phone without complying with Google’s technical requirements. Google argued that these anti-forking measures prevent “software fragmentation and the potential diffusion of incompatible versions of the software,” factors that could ruin the product in the eyes of consumers. The European General Court responded that “[i]t is not necessary to settle the dispute between the parties as to the harmfulness or benefits which fragmentation might have represented for Google and for the entire sector.” Instead, it concluded that “the extremely rapid growth of the ‘Android ecosystem’ from the early 2010s onwards makes Google’s claims regarding the hypothetical risk that the threat which it describes to the very survival of that ‘ecosystem’ could have continued throughout the infringement period implausible.” The Commission’s antagonism towards Google’s anti-forking/efficiency argument could well be not so much about restraining self-preferencing, as about allowing rivals to free-ride on Google’s Android investments. This again sounds quite a bit like preferring competitors to the process of competition.
What sets Microsoft apart from Bayou Bottling and Christy Sports, as we noted earlier, is that Microsoft involved other types of activities that are otherwise actionable. The U.S. cases, as well as some aspects of the European General Court’s reasoning in Google Shopping, illustrate the need to distinguish self-preferential conduct from other actions that can rightly be prosecuted. Refusals to deal, tying, and consumer fraud or data privacy violations are commonly implicated in antitrust enforcement actions that involve activity that might be characterized as self-preferencing. But these types of conduct are already treated under existing laws. Adding in a “self-preferencing” count in these contexts adds little or nothing.
There generally is no duty to deal with competitors. Refusals to deal violate Section 2 when the refusal makes no economic sense apart from the exclusionary impact on rivals. There are, therefore, some limited instances where a monopolist’s refusal to deal with a competitor can violate Section 2 of the Sherman Act. Some of those might involve what could be called self-preferencing.
What we have called “pure” self-preferencing should stay clear of refusal to deal liability except where the refusal violates the no economic sense test. In Streetmap, for example, there was no allegation that Google would have profited more from promoting the Streetmap product over Google Maps either in the long run or in the near term. In Google Shopping, the European General Court narrowed the European Commission’s finding and held that if Google’s conduct consists “solely in the special display and positioning” of the platform’s own products and services, it is not necessarily abusive.
Although Trinko poured a large bucket of cold water on the doctrine, some U.S. authorities have recognized an “essential facilities” exception to the general rule that a monopolist has no duty to deal with competitors. The doctrine is frequently evoked in European antitrust litigation. In the EU,
a refusal to deal many trigger an antitrust violation when: (i) access to the product or service is indispensable to a firm’s ability to do business in a market; (ii) the refusal is unjustified; (iii) the refusal excludes competition on a secondary market; and (iv), if intellectual property rights are involved, it prevents the emergence of a new product for which there is potential consumer demand.
As Trinko recognized, this essential facilities argument can be applied far too broadly. The lessons learned from the utilities-style enforcement of antitrust statutes in 1970s should discourage any regulatory attempt to declare a digital product an essential facility just because of its popularity. Europe has moved in the opposite direction. The new laws on self-preferencing in Europe deviate drastically from U.S. jurisprudence in numerous respects, prominently including the Europeans’ readiness to subject nearly all the big-name (U.S.) tech firms to the essential facilities doctrine. The European Union Digital Markets Act (DMA), for instance, terms many large online platforms as “gatekeepers,” and in turn the new law provides, among many other things, that these “gatekeeper” firms would be enjoined from treating their own products more favorably than rivals. The DMA’s requirement of treating rivals the same as the “gatekeeper” firm appears not to consider or care about the negative effect on large firms’ incentives to develop new and better products. Why undertake such costly investments if there is no benefit to be gained?
As the Trinko court warned, these kinds of broad regulatory attempts cut against the very purpose of antitrust laws of encouraging the invention of socially and economically beneficial tools and facilities. Many of the “gatekeepers” were non-existent two decades or even ten years ago, and their success can be viewed as entrepreneurship and the genius of innovation. The “gatekeeper” laws against self-preferencing impose a heavy price on their success and may ironically snuff out the next generation of “gatekeepers” still in the cradle.
“Pure” self-preferencing can also be distinguished from tying because preferential promotion of one’s own product or service does not necessarily entail coercing customers to use it. Moreover, to the extent tying is used to achieve a self-preferential outcome, the U.S. antitrust policy towards tying has transformed over a long period and pivoted from the hostile approach of the early per se rule. In Jefferson Parish, the earlier hostility was turned to a modified per se rule that permitted the consideration of possible tying efficiency gains (with four judges in favor of a rule of reason). And the Supreme Court’s 2006 decision in Illinois Tool Works acknowledged that tying arrangements often have procompetitive effects, a proposition fundamentally inconsistent with any per se rule. Tying arrangements can have severe anticompetitive consequences, but adding “self-preferencing” to the analysis adds nothing.
The Italian Competition Authority’s separate investigation into Amazon, and the subsequent imposition of billion-plus euro fine, focused not on Amazon’s use of the data collected on Amazon’s sites in competition against the third-party sellers on these sites, but rather on Amazon’s requirement that these sellers must use Amazon’s delivery services to be eligible for the Prime program. The Italian authorities deemed this to be an act of tying and improperly leveraging the dominance of Prime to force Amazon’s logistic service on the third-party sellers. Similarly, the Dutch Competition Authority imposed a penalty on Apple for requiring dating-app providers that appear in Apple’s App Store to use Apple’s payment services. In essence, Apple was deemed to have used its dominance of the App Store to restrict the app developers’ freedom of choice in picking their own payment processors. Although one might well question these results, what is clear is that the true transgression in the Dutch case, as in the Italian case, was tying.
Exclusive dealing arrangements have long been examined under the rule of reason, and “foreclosure” has for decades been the critical issue in evaluating any exclusive dealing claim. While the foreclosure concept was developed as a useful proxy for analyzing harm to competition, as the sophistication of the antitrust analysis has increased, foreclosure as a proxy for analyzing harm to competition has been found inadequate even in cases where foreclosures has properly been defined. The relevant question is instead “whether there has been an adverse effect on price, output, quality, choice, or innovation in the market as a whole.” Self-preferencing generally falls outside this arena, but where the effect of the conduct is to make a large portion of the relevant market unavailable to rivals, there will be a violation absent very substantial countervailing efficiencies. The number of instances where pure or simple self-preferencing (i.e., just favoring your own product) might truly foreclose rivals in this manner would seem few and far between. But the Microsoft case again provides a useful basis for comparison. There, Microsoft did not simply promote its own browser over Netscape; it entered into agreements with computer OEMs, Internet service vendors, and others that were effectively exclusive arrangements properly analyzed under an exclusive dealing framework. Engaging in a separate “self-preferencing” analysis would have added nothing.
Self-preferencing using deception, similarly, can be and is addressed under consumer protection statutes. The issue arises most often in the context of digital platforms’ collection and use of consumers’ personal data for their commercial benefits. For example, a business might mislead consumers about the collection and use of personal location data. To the extent that this deceit eventually affords the business a significant and unfair advantage over its competitors in designing and developing related products, the deceitful conduct can be prosecuted under existing antitrust and consumer protection statutes.
There are frequent news reports about businesses that mislead consumers about their commitment to the privacy of users’ personal data in order, for example, to promote a mobile application where the expanded collection of personal information is then often combined with consumers’ Internet activity to give businesses greater insight to users’ habits and behavior; this can give businesses a competitive edge. In the digital economy, profits and commercial advantage are to be sought in the businesses’ aggressive race to collect more data and more granular data about their users. But if some businesses gain the edge through fraud and deceit, then they can be prosecuted for that fraud and deceit. Further, if they use that edge either to foreclose competition in the native market or to harm competition in an adjacent market, then there may be a viable antitrust case to be made. Legal analysis for antitrust violations by businesses that fraudulently obtain consumer data would hardly dwell on self-preferencing. Rather, the pivotal issue here is whether the deceit gave the defendant an unfair advantage in competition and whether the unfair advantage eventually stifled competition.
Self-preferencing can often be a significant feature of leading tech firms’ commercial and research strategy when these firms straddle multiple digital markets. The adjacent areas are typically where they focus major parts of their research and investment; and this is where many exciting new products are born. It is time regulators and courts set clear and consistent legal standards for treating the subject. In our view, any recommended treatment of self-preferencing should stay closely aligned with the existing U.S. case law, which in the past several decades has created a stable and predictable legal framework conducive to huge investment, huge reward, and relentless innovation and progress in the tech sector. There is no principled basis for preferring competitors over consumers in competition cases. To use Steve Salop’s example, doing so would mean that a merger that raised prices to consumers would be just fine if rival profits increased by an equivalent amount. Logically extended, a large portion of existing antitrust law would have to be revisited. Has the economy suffered since Brunswick called for the opposite result? Hardly. We have seen the greatest technological progress, with associated societal benefits, in world history. That should be celebrated, not reversed.
Pure self-preferencing should be presumptively legal. The U.S. case law from Bayou Bottling to Christy Sports is unambiguous in this aspect. New nomenclature does not alter the fundamental legal perspective on promoting one’s own product over those of rivals. Where self-preferencing is effectuated by actual exclusionary conduct, the exclusionary conduct itself provides a sound basis for condemnation. The case has not been made for a new species of violation.
