Did Facebook Cheat?: A Test Case of Antitrust Ethics

Abstract

Citing corporate concentration and lax enforcement since the Reagan era, the Biden administration has declared a new era of aggressive antitrust prosecution, bringing antimonopoly actions against tech giants such as Meta, Google, and Amazon. But what’s so bad about monopoly or corporate concentration? The standard answer appeals to economic consequences, such as higher prices or deadweight losses. This paper offers a different framework. It argues monopolizing can be a form of cheating, which is a wrong that attaches to means, not just ends; an athlete who cheats but loses still does wrong. In particular, this paper argues that certain market-controlling strategies constitute a form of cheating I call ‘structural cheating,’ best illustrated by the metaphor of creating an unlevel playing field: rather than compete fairly on merits such as product quality and price, a firm that acquires rivals biases the market in its favor, thereby entrenching a dominant position that effectively forces would-be competitors to compete uphill. By framing (alleged) antitrust violations as cheating, while using the FTC’s lawsuit against Facebook (now Meta) as a test case, this paper provides a needed corrective to those citing market success as evidence of merit or skill. A further upshot is the structural cheating account better explains the distinctively problematic features of social media market concentration than Heath’s Market Failures Approach. More generally, this paper provides a normative lens for analyzing fair market competition and shows why it’s not only winning or losing that counts in capitalism, but how one plays the game.

Introduction

Antitrust enforcement is having a moment. Citing increasingly concentrated markets and forty-plus years of lax enforcement, the Biden administration’s Federal Trade Commission has declared a new era of aggressive antitrust prosecution,¹ bringing antimonopoly lawsuits against major firms such as Amazon, Google, and Facebook (now Meta).

But what’s so bad about monopoly or corporate concentration? Standard answers appeal to consequences. These include higher prices or deadweight losses in the economic arena, or inequalities of power or influence in the political arena. Undoubtedly, these consequences are significant. Yet this paper offers a different framework. It argues that monopolizing can be a form of cheating, which is a wrong that attaches to processes or means, not just outcomes or ends; an athlete who cheats but loses still does wrong. In particular, I argue that certain (alleged) antitrust violations—especially acquisitions of major competitors and the liberal use of non-compete clauses—can constitute a particular kind of cheating I call ‘structural cheating,’ which is best illustrated by the metaphor of creating an unlevel playing field. Rather than win customers via superior merits such as product quality and price, for example, a firm that buys out or contractually excludes rivals effectively tilts the field in its favor, biasing the market and forcing would-be opponents to play uphill.

There are several important reasons for employing the moral framework of cheating in addition to the legal framework of “anticompetitive conduct.”² One is that distinguishing cheating from competing fairly is crucial for determining merit or desert.³ Winners are admired for their successes, assuming their victories are fair. Cheaters (attempt to) win by unfair means, and, if successful, do not deserve the esteem or adulation rightly given to legitimate victors. In the twenty-first century, tech founders such as Zuckerberg, Bezos, Musk, and Jobs are widely lauded if not idolized for their entrepreneurial successes, and their victories over weaker competition or obsolescent technologies are celebrated as exemplars of vision and strategy in business schools around the world. Capitalist economies more generally are often perceived in terms of winners and losers, with winners presumed to have won via merit the losers lack.⁴ But what if some gains are ill-gotten, due more to underhanded or unfair competition than meritorious conduct? Highlighting the potential role of cheating can thereby serve as a powerful corrective to the claim that market success is evidence of skill, merit, or desert.

Adopting the framework of cheating can also make clear the normative stakes and standards of putative antitrust violations. An action does not count as cheating only it has significant negative effects, nor is cheating rendered permissible if it has positive effects. Yet firms are often convicted of antitrust violations—or accused of them in the first place—only if certain negative consequences materialize (e.g., high prices), while monopolizing actions are often condoned if certain positive consequences do materialize (such as lower prices, at least in the short-term).⁵ Courts of law may shy away from what seem to be non-demonstrable philosophical assertions of immorality, yet the moral framework of cheating adds value to how one understands the stakes of the competitive process. It suggests cheating not be overlooked simply because it resulted in (economic) benefits; after all, a fan’s enjoying a team’s tainted victory does not justify the means.

The strategy undertaken in this paper is to use the FTC’s antitrust lawsuit against Facebook (now Meta) as a jumping off point. The FTC sued Facebook in August of 2021, accusing it of monopolization via “anticompetitive conduct and unfair methods of competition.” These charges include Facebook’s acquisitions of social media and messaging giants Instagram and WhatsApp, as well as its requirement that employees and third-party software developers sign non-compete agreements as a condition of accessing the Facebook platform. Yet acquiring companies and signing contracts are not prima facie impermissible. Explaining how such actions could nonetheless constitute wrongful cheating is an important upshot of the present account.

Another contribution of this paper is that the structural cheating account provides a superior explanation of antitrust ethics than the recently popular Market Failures Approach, as championed by Joseph Heath (2014). One reason is the MFA cannot easily explain the distinctively problematic features of concentration in social media markets, which differ from traditional concentration problems as explained by the neoclassical theory on which the MFA is based. Second, I show the MFA either misclassifies the moral status of thwarting innovation, or cannot explain its wrongmaking features. The reason is the lack of innovation is not technically a market failure, i.e., a failure to realize the idealized conditions of a perfectly competitive Pareto-efficient market. Because the structural cheating account naturally explains thwarted innovations in terms of cheating future competition in dynamic markets, it is, on balance, stronger.

The structure of this paper is as follows. I first briefly set out the FTC’s case against Facebook. I then sketch an account of what I call ‘structural cheating,’ which occurs when a powerful player tilts the field to their own advantage, thereby putting its competitors (if any) at a structural disadvantage. The next section fleshes out the structural cheating account by responding to objections aimed at the sports-theoretic framing of antitrust (e.g., in terms of merit, cheating, or level fields of play). Next, I consider and rebut challenges to the links between cheating, monopolization, and moral impermissibility. The final section shows how the structural cheating account provides a better account of antitrust ethics than the MFA, while also providing a supplement to sociopolitical accounts of monopolistic wrongdoing. More generally, this paper provides a normative lens for viewing fair market competition, and shows why it’s not only winning or losing that counts in capitalism, but how one plays the game.⁶

The FTC’s Complaint Against Facebook: an Overview

I begin with a condensed presentation of the FTC’s 2021 antitrust suit against Facebook (now Meta), which runs as follows.

After its inception in 2004, Facebook quickly became the world’s preeminent social media site upon displacing MySpace c. 2006. This dominant position was held steadily over the next several years, and into the following decade. Yet the world of social media was about to change. As smartphones became ubiquitous, Facebook’s edge, born in the desktop era, looked precarious. Facebook’s app seemed to work less well on the new mobile devices than that of its upstart rival Instagram, whose glossy photo-centric interface seemed tailor-made for the smaller mobile phone screen. Meanwhile, the WhatsApp instant messaging service was growing increasingly popular as well, and seemingly poised to displace not only Facebook’s “Messenger” instant messaging service, but even standard (SMS) text-messaging as the preferred means of mobile communication.

In the FTC’s account, Facebook initially tried to compete head-on with Instagram and WhatsApp from roughly 2009 to 2011. It cites various internal documents detailing Facebook’s attempts to transition to a mobile platform, incorporate more photo-sharing and photo-interactive elements to its newsfeed, maintain its display ad click rates despite the smaller and differently behaving mobile screen, and to develop and enhance its mobile messaging services. These attempts were largely unsuccessful, and Instagram and WhatsApp continued to grow. Rather than continue to compete, Facebook decided to buy out the competition, purchasing Instagram for approximately $1 billion in 2012, and WhatsApp for a staggering $19 billion in 2014. To this day, Instagram and WhatsApp remain popular services. Yet they are no longer competitors of Facebook, but components or complements under the larger ‘Meta’ umbrella.

The FTC asserts these mergers and acquisitions constituted anticompetitive practices. It alleges that Facebook simply acquired its competitors to prevent them from competing, rather than compete on the merits. Moreover, it alleges that Facebook acquired these companies because it lacked the talent or ability to compete on the merits. The FTC holds that this is not effectively different than Facebook “bribing” Instagram and WhatsApp to not compete (FTC vs. Facebook, 2005, p. 5).

These are not the only acquisitions subject to the FTC’s complaint but also the highest profile. As part of the same charge the FTC alleges, Facebook continually monitors the market for acquisition targets and, as a strategic policy, has purchased or has attempted to purchase any firm deemed likely to pose a competitive threat. This included failed attempts to purchase Twitter (now “X”) as far back as 2008 and Snapchat on several subsequent occasions.

Facebook is also said to have complemented this acquisition strategy by actions falling under another count of anticompetitive conduct levied in the complaint: that it required employees and third-party software developers to sign non-compete agreements as a condition of accessing the Facebook platform. This “conditional dealing policy” allowed Facebook to entrench its monopoly position, the FTC argues, by leveraging the innovations of potential rivals to improve the user experience of Facebook, while simultaneously preventing those potential rivals from becoming actual rivals, thus stifling competition. Conjoined, the (alleged) net effect is that users and advertisers are deprived of “the benefits competition,” including a greater variety of products with superior quality. So the FTC is seeking redress: it aims to divest or separate Facebook from Instagram and WhatsApp, and ban Facebook from using similar acquisition and contractual strategies in future.

In this paper I will assume the FTC’s case is materially correct, i.e., that Facebook did what the FTC says it did, such as purchase Instagram. I do this not to beg questions; these facts are compatible with prosecutors failing to meet the current (yet historically variable) relevant legal standards.⁷ Instead, the point is to use the details as a springboard, or test case, for the more general moral questions they prompt. This includes whether there are good reasons for construing the legal concept of unfair competition in terms of the moral concept of cheating. In the next section, I argue there are.

Unfair Competition and “Structural Cheating”

What is cheating? The best attempts at analysis come from philosophers of sport. Despite some marginal disagreements, cheating is generally understood as taking unfair advantage, with ‘unfair advantage’ itself cashed out as a violation of the conditions of equality that should characterize a competition (Aspin, 1975, p. 55; Dixon, 1999, p. 138; Feezell, 1988, pp. 58, 61–2; Luschen, 1977, p. 67).⁸ The reasoning is clear. Competitions are supposed to be a test, a mechanism for determining or revealing merit, skill, or desert. As MacRae puts it, “competitions are tools for detecting excellence” (MacRae, 2019, p. 343). Yet competitions reveal excellence or merit only if competitors have a fair or equal opportunity of winning on the basis of that merit. Because the cheater avails herself of tactics or strategy not available to the other competitors—because they’re outlawed—the cheater undermines the merit or excellence-detecting function of the contest. Equivalently, the cheater subverts a meritorious outcome by privileging herself at the expense of fair competition.

Applying this account to Facebook’s alleged behavior is straightforward. The FTC holds that Facebook avoided competing on the merits, and that it bought out its competition for the very purpose of avoiding competing on the merits. To the casual outside observer, Facebook may appear as the deserved victor in the competition for social media preeminence. Yet the FTC’s charges imply Facebook achieved this position by means other than acumen or skill, i.e., via anticompetitive or unfair tactics. Despite the word’s absence from the FTC’s complaint (and from antitrust law more generally), such behavior is reasonably construed as cheating.

Yet one might be skeptical. An initial objection points to an apparent dissimilarity to paradigm cases of cheating in sport. Athletes do not generally announce their use of performance-enhancing drugs, nor that they are spying on an opponent’s game-planning or strategy sessions. More generally, one might presume cheating requires deception, or the attempt to conceal one’s self-interested actions from authorities. Yet Facebook’s actions were open, or at least not covert. The acquisitions of Instagram and WhatsApp were public market transactions, for instance. And many of Facebook’s internal communications the FTC cites—including a seemingly damning email written by CEO Mark Zuckerberg in 2008, stating “it is better to buy than compete”—seem to have been written without concern for their discovery.⁹ Nor is Facebook accused of using non-disclosure agreements or other means of concealing their activities that might imply a guilty conscience. From this, one may infer that Facebook didn’t cheat.

The objection fails, however, as cheating does not require deception. Bernard Gert gives the example of a boss blatantly undercounting strokes at a company golf outing, knowing his employees dare not call him out (2005, p. 18). More generally, Gert argues that those with the power to avoid sanction can cheat without deception (ibid.). Unfair advantage is still taken, in this case by a powerful party who can play by his own rules.

It is telling that Gert’s example does not involve professional sports. Professional sports are not corrupt, by and large; the wealth or prestige of participants does not in general unduly influence the adjudication of contests, for instance. That this is so even in games contested by millionaires shows the powerful commitment to norms of fairness and merit. Of course, there are dubious cases on the margins, e.g., an umpire being reluctant to eject a star player during a championship game, but one rarely sees in professional sports a flagrant abuse of power akin to the boss in Gert’s example. Such cases being rare, however, do not mean that those with power openly taking unfair advantage are not a form of cheating.¹⁰

Cheating can also occur by corrupting officials; a competitor who hires biased referees is undoubtedly cheating (Feezell, 1988, p. 59). A similar concern arises in business competition. Consider “regulatory capture,” whereby industry or special interests unduly influence government regulators for private gain, at the expense of the public interest. A firm that captures an agency or gets laws passed to favor its products over its competitors, regardless of merit, takes unfair advantage, and so satisfies the criteria for cheating. This is so even if the behavior is done openly or without deception. The conditions of equality characteristic of a fair competition are still undermined, preventing the competition from detecting or revealing excellence, skill, or merit; that the favored product acquires a dominant market share in this biased scenario would be no indication of the product’s superior quality.

This reasoning allows for the introduction of a crucial distinction. Familiar examples of cheating in sports typically involve deception because participants do not have the power to openly cheat or unduly influence officials. Due to these background condition of equality, when cheating does occur, it generally involves the use of deceptive tactics or strategies. It therefore concerns the illicit manner by which one competes on an otherwise level playing field. Call such cases ‘tactical cheating.’ Yet sufficiently corrupting the referees might make illicit or deceptive tactics unnecessary. In such cases, one could enjoy an unfair advantage even if one otherwise played cleanly, the dirty work being outsourced to another party. Or consider the commonly used metaphor of competing on a level (or unlevel) playing field. Part of professional sports being fair entails that no team must literally play uphill against an opponent with the luxury of playing downhill. Yet suppose one side did have the power or authority to create an actual inclined or unlevel field and required its opponent to play uphill. This would introduce a structural inequality into the game itself, providing an unfair advantage that would undermine the favorite’s claim to merit or desert if victorious, even irrespective of tactical cheating. Let ‘structural cheating’ be the creation or exploitation of unequal structural conditions that provide an unfair advantage that undermines a meritorious outcome. Thus, whereas ‘tactical cheating’ involves the unfair manner by which one competes on an otherwise level playing field, structural cheating tilts the field to favor one side, which can (but needn’t) obviate the need for tactical cheating. Put another way, tactical cheating concerns the manner of competition. Structural cheating concerns the conditions of competition.

I suggested above that the FTC’s accusations against Facebook are tantamount to an accusation of cheating. The further claim defended here is that these actions plausibly constitute a particular kind of cheating: structural cheating. Facebook is not accused of deception, or using unfair tactics against background conditions of equality. Rather, the accusation is that by acquiring major competitors, and contractually restricting the behavior of would-be competitors, Facebook effectively tilts the field in its favor to avoid competing on the merits. This suggests the ‘structural cheating’ label is apt.

Nonetheless, one might object. Facebook is not accused of corrupting regulators, or unduly or illicitly influencing government actors to create rules favorable to itself at the expense of its competitors, which is how the concept was just introduced. Second, what Facebook is accused of—acquisitions and contractual stipulations that allegedly constitute unfair or anticompetitive behavior—was done with the consent of the relevant parties; e.g., Instagram and WhatsApp agreed to be purchased, and third-party app developers accepted the non-compete clauses Facebook offered. Yet cheaters do not typically if ever receive the consent of their opponents. This might suggest Facebook’s actions were not cheating, structural or otherwise, especially if one agrees with Green (2004, p. 140) that cheating requires someone (or something) to be cheated.

While plausible, the objections fail. One reason is that even if the focal behavior involves consensual arrangements between some parties, there are at least four other possible objects or victims of Facebook’s (alleged) cheating.¹¹ The first are competitors not party to Facebook’s agreements, such as Google or Apple, as well as countless smaller firms seeking to acquire market share in social media markets.¹² Second are users of social media, who are allegedly “deprived of the benefits of competition” they would have otherwise enjoyed. A third possible victim is society at large. Suppose a cheater uses the spoils of victory to pursue unjust causes. Or suppose a fair competition might have checked the power of a dominant player to act unilaterally in ways largely viewed as negative. Those who believe Facebook’s influence on politics or culture to be deleterious can claim to have been cheated out of the better or more just outcome that would have resulted from a competitor who checked or balanced Facebook’s influence.

The fourth possible victim of cheating is competition itself, including future competition. It may initially sound implausible to consider a process such as competition a possible object or victim of cheating. Yet the FTC has traditionally cited the responsibility to protect competition—not individual competitors—in its mission statement.¹³ Further, this role is plausibly comparable to a sports commissioner whose mission includes protecting the integrity of the game, over and above the interests of any individual or team, and regardless of consensual arrangements between players. Yet if cheating or corrupting a process still sounds inapt, one needs only consider future competitors the relevant possible victim of cheating despite some parties’ consent. On either construal, it is adducing the status of the future that matters here. Doing so also helps explains why Facebook’s actions can constitute structural cheating even without unduly influencing regulators or the law, and so answers the previous objection as well.

Consider the difference between static and dynamic competition: the former occurs at a given time between a fixed number of firms, whereas the latter occurs over time between a varying number of firms, as firms exit or enter the market. Dynamic competition is naturally linked to innovation, as new firms or products enter the market, and firms unable to keep up with innovative competitors exit. To shield themselves from this eventuality, however, extant firms may seek to erect barriers to market entry that would prevent future competitors from competing. One method is having workers or contractors sign non-compete clauses. Such clauses prevent future competitors from even stepping onto the field by effectively tilting it vertically, to continue with the metaphor.¹⁴ (Non-competes are especially common in tech industries, where incumbent-threatening innovations may just be an app away.) Although non-competes tend not to apply in perpetuity, they can buy an incumbent time, during which further advantages may be entrenched. Social media companies in particular may exploit network effects for this purpose. Network effects obtain when the value to one user increases as other users join the network; in the case of social media, the value to a user is proportional to the network’s size. Network effects therefore serve as a barrier to entry, as even a novel social network with technically superior features will be seen as less valuable than an incumbent site already populated by a critical threshold of users.

Network effects may be endemic to social media, so a company need not be blameworthy for benefiting from them. One can nonetheless distinguish deserving to benefit from building a superior network as opposed to simply purchasing and absorbing a competitor’s network. Moreover, it is the conjunction of acquired network effects and non-competes that is especially worrying, as these are complementary or even synergistic strategies for avoiding competition; non-competes buy time while acquiring competitors buys (market) power, making it increasingly difficult for an upstart firm to dislodge a monopolizing incumbent, despite the merits. This explains why both appear as counts in the FTCs complaint.

This also explains why Facebook’s actions are plausibly considered structural cheating, despite not being an example of regulatory capture, nor involving the corruption of government officials or agencies. The reason is that non-compete restrictions and monopolization by merger, especially when conjoined, constitute non-merit strategies to acquire market power and prevent future competitors from competing under anything resembling conditions of equality. Put differently, the exploitation of purchased network effects and contractual restrictions on would-be dynamic competitors gives undue weight or power to incumbency in lieu of (internal) skill or (external) product quality. This constitutes “field-tilting” against future competition, which must effectively compete uphill (if at all) against the incumbent’s built-in but non-meritorious advantage. Such strategies satisfy the criteria for structural cheating.¹⁵

If one is skeptical, consider the matter another way. Professional sports teams do not purchase each other instead of competing, and they certainly don’t declare themselves champion after bribing a team to not compete in a championship match. So the idea of acquiring the competition as cheating does not have a familiar analog in professional sports. Were such a purchase to happen, however, there is no doubt the purchasing team would be accused of cheating to win.¹⁶ Were this purchased championship then used to create a built-in advantage in future seasons or tournaments—the analog of tilting the field against dynamic competition—the unfairness would only be compounded.¹⁷

Business and Sporting Competition on Level Fields of Play

This section aims to flesh out the structural cheating account, especially with respect to those still skeptical of framing market competition as akin to sporting competition in a manner that makes the framework of cheating apt.

Why would one be skeptical? Sports are governed by well-defined rules relative to which some actions clearly constitute cheating. But what are the rules of business competition, and who makes them? And what exactly constitutes a level playing field in a market? Perhaps one might worry that without clear rules to break, or a literal playing field, the distinction between fair and unfair business competition might be murky or without content. And this might in turn threaten the distinction between meritorious and non-meritorious actions. For instance, why shouldn’t one think Facebook’s “strategic acquisitions” or conditional dealing polices aren’t meritorious practices, indicative of business acumen?¹⁸ Or perhaps even more strongly: why think business competition aims to reveal or determine merit at all? Perhaps markets are merely mechanisms for allocating resources or promoting welfare, rather than spectacles or contests where society tests who has merit or skill. If so, perhaps the distinction between cheating and competing on the merits might be rendered otiose or inapt.

The objections can all be met. Consider first the matter of the rules of business competition, which are largely given by antitrust law, or ‘competition law’ as it’s called in Europe. It consists of legislation, such as the Sherman Antitrust Act of 1890, and the Clayton and FTC antitrust acts of 1914, as well as judicial precedent or case law.¹⁹ This legal corpus explicitly delineates permissible from impermissible business practices, such as bans on predatory pricing, price gouging, or product-tying, while granting government agencies such as the FTC the power to police competition and market concentration. The result is a broad similarity with governance in sports: antitrust law plausibly constitutes (part of) the rulebook governing fair competition, akin to any sports organization’s rulebook, while executive agencies and the judiciary play governmental roles analogous to sports organizations such as Major League Baseball or the National Football League- namely, that of interpreter, adjudicator, administrator, and enforcer.

Of course, the objector to the sports-theoretic framing of business competition would not deny the existence of antitrust law, or claim that business competition lacks rules altogether. What may be more worrisome is the apparent indeterminacy or unclarity of antitrust law, especially regarding the line between permissible and impermissible mergers and acquisitions. Mergers and acquisitions are routine occurrences, and only a small percentage are blocked. Some mergers even seem desirable, from an economic point of view, and it may not be antecedently obvious whether a given acquisition is above board. So the general worry is that the framework of cheating may be inapt in light of a potential disanalogy with sports, i.e., that sports yield a more determinate or black-or-white distinction between rule-following and rule-breaking than does antitrust law, and more antecedent clarity on whether prospective behaviors are proscribed. The particular worry is whether the structural cheating account can distinguish permissible from impermissible mergers, or if it’s committed, perhaps implausibly, to considering all mergers to be cheating.²⁰

But both worries can be met: the structural cheating account can distinguish permissible from impermissible mergers, while the putative differences between sports and business with respect to indeterminacy are overstated. On the latter point, first consider a non-sports example. It is currently debated whether students using AI chatbots such as ChatGPT counts as cheating on academic coursework. But it being debatable if not indeterminate whether using ChatGPT is cheating does not entail there is no such thing as academic cheating, or that the concept of cheating is inapt. Much the same applies to sports. Using steroids was not banned in Major League Baseball until 1991. Mandatory testing did not begin until 2003. Was a player who used steroids in 1988 cheating? Was a player who used non-steroidal performance-enhancers (such as Human Growth Hormone) in 2002 cheating? Many players with otherwise Hall of Fame worthy resumes who have been tainted by suspicion of PED use during this era have been denied entry into the Hall of Fame for being cheaters, even if they may not have been by the letter of the law.²¹ So did they cheat? The answer is debatable if not indeterminate. Yet this does not render the concept of cheating inapt. More bluntly, there is such a thing as cheating in baseball even if it is sometimes debatable what constitutes cheating in baseball.

Precisely because it might seem unjust to be punished or judged as a cheater retroactively,²² however, governing bodies generally attempt to reduce indeterminacy in rule-application whenever possible. An important example in the antitrust context is the 1976 passage of the Hart–Scott–Rodino Antitrust Improvements Act. The act requires pre-merger notification for large mergers, and a waiting period during which the proposed merger can be assessed for antitrust compliance prior to consummation. The FTC and DOJ have also provided merger guidelines since 1968 detailing the conditions under which antitrust prosecution is likely. These guidelines—which are frequently revised and updated—are expressly intended to “enhance transparency and promote awareness” of how antitrust rules are understood and applied, especially regarding changes over time.,²³²⁴ Yet again sports leagues are no different. For example, Major League Baseball often disseminates memos or directives telling umpires to enforce certain rules more strictly or literally than they have in past, or that umpires should now consider certain behaviors as running afoul of those rules even if those behaviors were previously tolerated.²⁵ So in both the sports and business cases, neither historical variability nor indeterminacy in rule-application undermines the applicability of a rule-following-vs.-cheating framework.

Still, lines must be drawn. What distinguishes permissible from impermissible mergers? And what does it mean to compete fairly or meritoriously on a level playing field in the business or antitrust context? The account runs as follows.

Meritorious fair market competition has a positive or ‘thou shalt’ aspect, and a negative or ‘thou shalt not’ aspect. The former involves competing on the basis of price, quality, and variety of goods or services,²⁶ as these are the public or consumer-facing qualities that provide consumers with information relevant to making a utility-enhancing purchase decision. The latter involves not interfering with or sabotaging a rival’s ability to compete on the basis of price, quality, or variety of goods or services. This includes not putting up significant obstacles or barriers that make it difficult for consumers to find or access goods that would otherwise be competitive on the merits. When firms honor these imperatives, the field is effectively level.

Another sports example anchors the account. It may be recalled that figure skater Tonya Harding arranged for her rival Nancy Kerrigan to be attacked just prior to the 1994 Winter Olympics, for the purpose of injuring Kerrigan sufficiently to put her out of the competition. Such actions obviously undermine a meritorious outcome; were Harding to win only because her opponent was injured, her nominal victory would be no indication of sporting merit or superior ability. The reason is Harding opted for a “behind-the-scenes” attack to compromise the quality of her opponent’s offerings, rather than allow the public-facing qualities of meritorious skating be the determining factor.

The market analog of Harding’s behavior is industrial sabotage, e.g., blowing up a rival’s factory or assaulting their employees. Note that this is a tactic often employed by gangsters who are not exactly known for fair competition.²⁷ Granted, it might seem absurd to compare M&A to acts of industrial sabotage. Yet the strategies do often share a crucial feature: they remove competitors from the market rather than improve the merit of focal products. This is especially clear in cases of what Cunningham et al. (2021) call “killer acquisitions,” and what the FTC and others have referred to as a “buy and bury strategy,” which consists of acquiring firms with competing products only to shut them down.²⁸ Acquiring firms thereby avoid competing on the consumer-facing merits of price, quality, or product variety, instead using business-facing or behind-the-scenes strategies that adversely affect what consumers are offered.

A more subtle example comes from Amazon’s (alleged) “self-preferencing” behavior (Kahn, 2017). The accusation stems from Amazon’s “dual role” as marketplace that links independent buyers and sellers, as well as a retailer on that very same marketplace. The worry is that Amazon alters its algorithms so that its own brands appear more prominently in keyword searches, while also requiring fees from third-party vendors to not be ‘buried’ in search results, which artificially raises prices.²⁹ This practice also makes a neutral or fair comparison of products more difficult—i.e., it creates significant barriers to consumers choosing on the basis of merits such as (fair market) price or quality. So Amazon’s (alleged) tactics here plausibly rise to the level of structural cheating: instead of allowing or providing equal access to competing goods qua marketplace, Amazon qua retailer biases or tilts the market to benefit itself at the expense its competitors (allegedly).³⁰

Return now to the question of what distinguishes permissible from impermissible acquisitions. Why does the FTC allow many but not all mergers? Its reasoning is straightforward: mergers between small firms in competitive markets are not likely to yield the deleterious effects of monopoly or market concentration. So the FTC is traditionally only concerned with large or ‘mega-mergers’ that might drastically change an industry’s competitive landscape.³¹ Yet one might worry this line of response is not really open to the structural cheating account, given its emphasis on means or processes, rather than ends or consequences. But this is not the case, and a similar (or roughly coextensive) distinction can be drawn; large mergers may constitute cheating, while smaller mergers may not.

An initial skepticism that there can’t be a moral difference between actions that differ only in degree should be put aside; differences of degree or magnitude often distinguish legitimate actions from cheating in sports. A baserunner peering in from second base to decode a catcher’s signs to the pitcher is permitted in baseball, whereas a spy using binoculars from the centerfield bleachers isn’t. Jostling under the rim for a rebound is permitted in basketball, whereas targeted elbow strikes or full-on tackles are not.³² More generally, differences of degree are real differences. A small-magnitude token of an action type is often permissible even if an action of a similar type but with greater magnitude is not.

But what principle (if any) justifies drawing the line? A rough characterization is that the distinction corresponds to whether the practice reveals skilled or ingenuitive play, as opposed to obviating the need for skilled or ingenuitive play. Baserunners decoding signs from second base requires baseball skill; a runner must be quick enough to act in the few moments they’re in the right position, and subtle enough to relay those signs discreetly while standing on an open field in front of millions of people. This contrasts with using binoculars from centerfield, or putting spy cameras in the opponent’s dugout. Such tactics obviate the need for skill and do not require real-time acuity. Similarly, maneuvering oneself under the basket in a good position to get a rebound is a basketball skill, whereas assaulting someone and then taking their now-open position would obviate the need for those basketball skills.

The analogous scenario holds in the business case. In a competitive market with many small firms, one small firm acquiring another won’t typically provide a structural advantage that obviates the need to compete on the merits, nor undercut the ability of rivals to compete on the merits. Such a merger may therefore be reasonably (but defeasibly) construed as a legitimate means of improving the public-facing merits of one’s products, for example, by creating price-reducing operational efficiencies or expanding a product line. A large firm in a concentrated market buying out rivals, however, is typically different. The extent to which a non-negligible percentage of the competition is removed is the extent to which one no longer needs to compete on the merits, with the limit case being a monopolized market with no competition at all. In such cases, “behind-the-scenes” strategies obviate the need for open or fair market competition, and are therefore reasonably construed as (structural) cheating.³³

Drawing the lines in this way also allows one to handle a challenge that may be lurking in the background. Sports franchises differ as to their wealth and resources, which may seem to endow a rich team with an “unfair advantage” over a poorer team, especially regarding which players the teams are able to acquire. This may not constitute cheating, but it might seem to challenge the presumption that even sporting competition occurs on a level playing field, or constitutes a fair competition determined by merit.³⁴

Yet this challenge can also be met. The key move is to not only distinguish on-field competition from behind-the-scenes strategies, as above, but to distinguish behind-the-scenes strategies that contribute to one’s own on-field merit (the ‘thou shalt’ imperative) from those that undermine the merit of opponents (the ‘thou shalt not’ imperative). It is certainly true that greater resources may allow a richer team to acquire more talented players, and this may raise the prior probability of victory, or the credence assigned to its occurrence. Suspecting prior to the contest that one side has more merit does not mean the contest doesn’t reveal merit or isn’t on the level, however. To say otherwise conflates the epistemic situation of the observer with the causal situation of the participant; the team with greater resources must still compete on the merits to demonstrate they are in fact the better team. Much the same applies to firms. Consider a large firm achieving economies of scale through operational efficiencies that allow it to sell cheaper goods than a local “mom and pop” store. Even if the larger firm’s efficiencies make it more likely to offer lower-priced goods, thereby creating the appearance of an inevitable victory, the competition is fair as long as its outcome remains determined on the open market by merits such as price and quality. The contrast in both cases is with behind-the-scenes strategies that obviate the need for fair competition by tilting the field or sabotaging an opponent’s ability to compete at full strength.³⁵ Only the latter is cheating, and so unfair in the relevant sense.³⁶

One last (and more global) objection is worth addressing in this section: that the entire framework of meritorious market competition is simply inapt. Perhaps markets only aim at the efficient allocation of resources, or welfare-enhancing distributions of prosperity, the objector suggest, rather than serving as a forum for victory and defeat.

But this is a false dichotomy. Certainly, markets are a mechanism for welfare-enhancing allocations or distributions of resources. Yet capitalist markets are also widely believed to be meritocratic, or at least as aiming at meritocracy. Put another way, capitalism’s ideology is undoubtedly meritocratic.³⁷ Unlike traditional caste or patrimonial social hierarchies, in which property, wealth, and status are inherited rather than earned, capitalism is believed to be a forum of social mobility, where anyone with the skill, talent or determination can rise to the top. Consider America’s traditional reputation as the land of opportunity, or its long-standing admiration for “Yankee ingenuity” and “the self-made man,” all of which point toward a widespread belief that with talent, gumption, or an enterprising spirit, one can earn a higher station in life via success in the marketplace.

Of course, many pillory these beliefs as myths, or consider them ideology in the pejorative sense. Yet many who complain capitalism is not meritocratic because the game is “rigged”—or who point to the many unearned advantages of those born into socioeconomic privilege—still use the meritocratic paradigm of a level (or unlevel) field, even if it’s conspicuous only by its absence (Hayes, 2012; Markovits, 2019; Sandel, 2020).

The belief in (the ideal of) market meritocracy is prevalent in political and legal circles as well. Consider conservative polemics against progressive taxation that suggest a “flat tax” would not punish winners and be more fair. Or consider free market economists such as Mankiw (2013) who argue that what appear as exorbitant CEO salaries are proportionate to their productivity and therefore warranted. Even antitrust enforcement has traditionally carved out an exemption for success believed to be merited. Famed “trustbuster” Theodore Roosevelt and progressive jurist Louis Brandeis—the latter of whom was instrumental in the creation of the FTC and Clayton Antitrust Acts of 1914— distinguished between legitimate and illegitimate means of growing a business (Urofsky, 2009; Wu, 2018). Judge Learned Hand followed this tradition when he argued in 1945 that “the successful competitor, having been urged to compete, must not be turned upon when he wins.”³⁸ And in 2021, a judge dismissed an antitrust suit against Apple’s allegedly anticompetitive use of its app store on the grounds that “success is not illegal.”³⁹ Even today’s hawkish FTC accepts the Supreme Court precedent that achieving a monopoly via superior products or “business acumen” need not be punishable,⁴⁰ a claim which relies on drawing a line between what is and isn’t merited.

It is in this cultural context that many enjoy the spoils of being perceived as victors in market competition. This is especially the case in tech industries; simply witness the laudatory if not apotheosizing attitude many have toward tech entrepreneurs, as mentioned earlier. But what if gains are ill-gotten, or market success is due to competing unfairly? Because of the link between cheating and a lack of merit, an analysis in terms of (structural) cheating is an important corrective to social and legal claims of desert, while at the same time accommodating the widespread belief that marketplace success can be merited. The account sets out the difference between success on the merits, by providing consumers with products they choose over a wide range of viable alternatives, as opposed to success due to restricting what consumers are able to choose, from mergers that remove products from the marketplace to tactics or strategies that bias the market toward the firm with the power to tilt the field in its favor.

Put a different way, the goal here is to not be revisionary by abandoning the widespread framework of market meritocracy, while at the same time not arguing that actual capitalist competition is meritocratic or occurs on a level playing field; certainly, the opportunities of countless individuals are hampered by any number of unjust obstacles. Instead, what I am arguing is that construing some but not all means of market competition as (structural) cheating or field-tilting makes clear that not all business success results from fairly winning a competition, even apart from the question of whether an initial endowment of resources was deserved. Sometimes winners don’t just have an unearned advantage based on inherited privilege; sometimes they cheat.

Might not Structural Cheating—or Monopoly—be Permissible?

An underlying presumption of the structural cheating account has been that cheating is wrong. Yet some have argued that cheating can be excused if not justified in certain circumstances. Still others might think that so-called antitrust violations might nonetheless be permissible if not desirable. The goal of this section is to be rebut these objections, while solidifying the links between cheating, monopolization, and immorality.

That cheating could be permissible or even justifiable might seem implausible. Many find it intuitive that the word ‘cheating’ has a built-in negative moral valence or connotation of intrinsic wrongness, roughly akin to the word ‘murder,’ which means not just any killing but a wrongful one (Feezell, 1988, p. 67; Vorstenbosch, 2010, p. 168; MacRae, 2019, p. 341). Yet some reject the conceptual link. A recent example comes from Eabrasu (2020), who argues cheating may be excused or even justified when a context is sufficiently “unfair” as to make cheating a legitimate “compensatory tool” against “arbitrary and/or undeserved privileges” (p. 526), or when one is subject to immoral rules imposed by illiberal political regimes, or the exceedingly wealthy or powerful.

But cheating being permissible (or even justified) in these situations will not help exonerate Facebook in particular, or license structural cheating more generally. Eabrasu defends cheating as a means by which less powerful agents may attempt to extricate themselves from an unjust, unfair, or abject position. It is a means of redress whereby one may attempt to level a playing field that has been tipped toward one’s disfavor. Structural cheating, however, is a phenomenon whereby a powerful player uses its power to tilt the field to its own advantage. It is a strategy used by a possessor or perpetrator of power, not its victim. So even if cheating is permitted as a tool to fight back against unjust power or unfairness, this will not excuse structural cheating.

Turn now to another attempt to excuse what might otherwise be termed antitrust violations. Some philosophers distinguish between harming and merely failing to benefit.⁴¹ Yet recall the FTC claims Facebook harms users and advertiser by depriving them of the benefits of competition- a claim which appears to be mirrored or endorsed by the structural cheating account. But if there’s a difference between failing to benefit and harming—and especially if the former isn’t sufficient for the latter—then one might infer that it’s “no harm, no foul” when it comes to thwarting competition.

This objection can be met in two ways. One is by defending the link between harming and failing to benefit in the relevant cases. Even those who argue that not all failures to benefit are harms nonetheless admit that some are (see esp. Shiffrin, 2012, pp. 372–373 and Purves, 2019, p. 2654). The reason is that some failures to benefit not only make one comparatively less well off, but result in states of disutility (e.g., pain or suffering). For example, failing to benefit someone by refusing to give them a lifesaving drug plausibly harms that person (cf. Boonin, 2014, p. 53, n2; quoted in Feit, 2019, p. 817). If so, the reasoning extends to the pharmaceutical company that withholds a drug on which it has a monopoly, or the pollution-causing fossil fuel company that thwarts the development of healthier alternative energy sources. To the extent that competitors in pharmaceutical or energy markets would have offered benefit-enhancing or harm-reducing alternatives, one can tie violations of fair competition to wrongful harming.⁴²⁴³

Yet even if harming and failing to benefit are distinct in such cases, the present account is not undercut, as one can be blameworthy or culpable for failing to deliver benefits even if it’s not technically a harm. More generally, it’s plausible that promoting progress is a good thing, ceteris paribus, and that when progress is attainable without being excessively demanding or supererogatory, the failure to realize it is lamentable, if not blameworthy when actively thwarted or opted against. Antitrust emphasizes how competitive markets can foster price reductions, quality improvements, and innovations, and the ways in which non-competitive markets can thwart the same. Highlighting the welfare-enhancing—and pain reducing—benefits of competition is one way to highlight the moral and not just economic import of fair competition.

A more serious objection to the structural cheating account appeals not to harms (or failures to benefit) due to monopoly or market concentration, but to its potential or actual benefits. Some claim that monopolies are efficient because they avoid “ruinous competition,” or obviate the need to allocate resources to the competition itself—consider marketing budgets, for instance—instead of improving (or cheapening) products.⁴⁴ Others point to the existence of “natural monopoly,” which is a market or industry most efficiently or cost-effectively served by a single firm. Crucial here is that some argue that tech industries in particular trend toward natural monopoly, including social media. One reason is the characteristic presence of network effects, as discussed earlier. Another is the high fixed cost/low marginal cost structure of IP or other information-theoretic assets.⁴⁵ So some hold it would be counterproductive or inefficient for antitrust authorities to “break up” monopolistic social media networks, or otherwise police tech industry M&A (e.g., Haskel & Westlake, 2022).⁴⁶

There are several available responses. Defenses of monopoly are typically consequentialist, in that they cite the (putative) benefits of consolidated control. Recall that the structural cheating account focuses not (just) on consequences, however, but on the process of competition itself. This can serve as a helpful reminder that the ends may not justify the means. Suppose the New York Yankees are the most popular baseball team, and that their success would maximize advertising revenue for the league. That would not justify their cheating to win. The same goes for a political commitment to democracy; the end of having one’s preferred candidate in office does not justify the means of rigging the election. In institutions where the process of competition matters, there is a distinct moral value to letting the process play out fairly, or on the level, after which one should let the chips fall where they may. The market is plausibly one such institution; while certain ends regarding efficiency might be desirable, the means also matter. Sometimes the integrity of the process is more valuable than a hoped-for end- especially when bypassing the process undermines the legitimacy of that end.

This is not to say the structural cheating account cannot countenance natural monopolies, or that it must wrongly misclassify all monopolies as impermissible. Suppose there is a genuine or on the level competitive bidding process for the contract that results in monopolistic provision of electrical power in a given region. There is no reason to say that simply because a monopoly results that there must have been cheating along the way. More generally, whether a firm competes fairly or not is orthogonal to the question of whether a monopoly exists, or should be tolerated. For the latter concerns the consequences of a monopoly, regardless of how it was obtained. The former concerns the process itself, and whether a firm earned or deserved its monopoly position.

That said, the structural cheating account does not abandon an analysis of consequences altogether, nor the need to police (natural) monopolies even if initially merited. The reason is that an industry’s being most efficiently served by a single firm may be contingent on the technology of the day. So it is imperative that future competition not be impeded, especially when the process of innovation or “creative destruction” may obviate the technology that justified the natural monopoly in the first place.⁴⁷ Granted, protecting future competition does not necessarily require government intervention; if “natural” barriers to entry do not arise from the very nature of the industry’s cost structure, new firms may spontaneously enter the market if profitable opportunities arise. If industry structure tends to generate natural monopoly and barriers to entry, however, then intervention may be necessary. This goes doubly when artificial barriers to entry—e.g., legal or contractual barriers such as non-compete clauses—are used to protect an incumbent firm against dynamic competition. So even if the initial natural monopoly were merited, an active role for antitrust authorities might still be warranted to protect dynamic (future) competition and progress (cf. FTC vs. Facebook 2021, p. 73).⁴⁸ More generally, in this way the structural cheating account weds a concern with process to a concern for outcome: if the consequences of monopoly (or concentrated industry) is impoverished competition, then policing the process of competition in the present to ensure healthy dynamism and fair competition in future can be defended on both procedural and consequentialist grounds.

Competing Accounts of Antitrust Wrongmaking

In the previous section, I defended the structural cheating account against objections that cheating or monopolization might be permissible if not desirable. In this section, I assume the moral wrongness of monopolization, and turn to alternate theories or accounts of that wrongness. I argue the structural cheating account is better placed to explain the wrongs of antitrust violations than the recently popular Market Failures Approach (MFA), championed by Heath (2014) and earlier by McMahon (1981), while also being compatible with accounts that locate the wrongs of monopoly in their larger-scale social or political effects.

The MFA’s Shortcomings as an Account of Antitrust Ethics

The MFA’s primary focus is not antitrust, but it can provide an account of at least some antitrust wrongdoing. This shouldn’t be surprising; the MFA is based on standard or general neoclassical theory, which identifies the ways in which concentrated or monopolized markets depart from the ideal of perfect competition. That theory holds that a perfectly competitive market is Pareto-optimal, meaning resources are put to their maximally (Pareto-) efficient use. This occurs when supply and demand equilibrate at a market-clearing price, i.e., when resources are produced and purchased at quantity/price mixes that satisfy both producers and consumers. The standard theory corollary is that when a market is monopolized, the monopolist restricts the supply of goods to keep prices at a profit-maximizing level, resulting in underproduction and overpricing relative to competitive levels. Such “deadweight loss” is a form of economic or allocative inefficiency, as potential efficiency- or welfare-improving exchanges are not made. Cartels and price-fixing arrangements are understood to have similar effects; the more closely these market structures resemble monopolies by rendering a market non-competitive, the greater is the inefficiency due to higher prices or reduced output.

The above is essentially a textbook account, and not distinctive to the MFA. What is distinctive is the MFA’s normative take on neoclassical theory. It holds that Pareto-efficiency is the goal of the market mechanism, and that actions are to be normatively evaluated by whether they advance or impede allocative or Pareto-efficiency. By creating deadweight loss (overpricing and underproduction), monopolization can therefore be construed as unethical. Price-fixing agreements or cartel-like behavior, along with other antitrust violations, can be explained as unethical departures from the Paretian ideal as well. This naturally extends to M&A that concentrate markets above perfectly competitive levels.⁴⁹

There are interesting criticisms of the Market Failures Approach as a complete business ethic that I will not entertain here.⁵⁰ For my purposes, the MFA is at least a plausible way of explaining some traditional problems of market concentration. Yet the MFA and its neoclassical framework are not especially well suited to capture the distinctive harms of tech industry M&A and market concentration in the digital era. There are two crucial reasons for this.

One stems from the neoclassical framework’s being fundamentally price-theoretic. The supply/demand models of neoclassical theory see the world in terms of output and price (the x and y axes on its standard graphs). It therefore sees market failures as failures to achieve the optimal price/output mix. In the case of monopoly or market concentration, this means overpricing and underproduction. But this is not the problem in social media markets. Facebook’s services are free to users, and social media sites do not underproduce social media services to raise market price above cost. If anything, by “selling” at zero social media usage is in effect priced below cost (a habit that is often supported by capital markets or other outside funding sources not derived from operating revenue).⁵¹ More strongly, it is plausible that social media services are overproduced, at least from the point of view of social welfare—that is, many take social media usage to be highly damaging or harmful—which implies that it’s underpriced from the same perspective.⁵² That the MFA sees concentration problems as problems of overpricing and underproduction, when the problem is more plausibly underpricing and overproduction, suggests a different theoretical lens is needed.⁵³

The second but related concern is that neoclassical theory is essentially a static end-state theory, as equilibrium theories tend to be. Equilibrium is by definition a state in which nothing (net) happens; it is a stasis due to balanced opposing forces that require an exogenous “shock” or external force to change. The neoclassically inspired MFA is therefore not well suited to explain the good-making features of innovation and dynamism, which are change or disequilibria-inducing processes, and so not states at all. More strongly, the MFA risks misclassifying innovation and dynamism as ethically problematic insofar as they disturb the static equilibrium the MFA takes as the end goal of market activity. This in turn renders the MFA ill-equipped to explain the wrongmaking features of Facebook’s anticompetitive actions, which may include preventing new firms from creating new products and even new markets, rather than its interfering with the efficient allocation of preexisting goods provided by a fixed number of firms in a static market.⁵⁴

To conjoin and summarize these two problems, the MFA locates the bad-making features of market concentration as a state of overpricing and underproduction, which prevents the market system from reaching a static equilibrium. Yet the problem with concentration in social media is the opposite on each front: the bad-making features are underpricing and overproduction, which prevent a dynamic system from triggering disequilibrium- or, as Schumpeter famously put it, “creative destruction.”

The claim might sound surprising, but it shouldn’t. The reason is that a lack of innovation isn’t technically a market failure. As indicated above, market failures are defined in terms of the maximally efficient supply/demand equilibrium that neoclassical theory predicts a perfectly competitive market will achieve if its idealized conditions are met (cf. Heath, 2014, p. 4). Market failures are therefore failures of actual conditions to meet ideal conditions. For example, because perfect or symmetrical information and zero transaction costs are idealized features of a perfectly competitive Pareto-efficient market, information asymmetries and significant transaction costs are market failures, the rectification of which are ethical imperatives on the MFA, and the exploitation of which are correlatively verboten (Heath, 2014, pp. 36–38).

Yet innovation is simply not a characteristic or condition that a perfectly competitive market instantiates. So failing to innovate is not a market failure. If anything, one could construe the presence of innovation as market failure, as another idealized condition of a perfectly competitive Pareto-efficient market is that products are identical or homogenous. Because innovations are differentiators, an innovative product or market “fails” in this respect.⁵⁵

Note that this point dovetails with and reinforces two points made above: that the MFA risks misclassifying innovation as ethically problematic, while also being unable to account for its presence due its being an exogenous or extrinsic shock to an equilibrated system. A fortiori, this implies the MFA lacks the resources to explain the good-making features of innovation, as well as the correlative bad-making features of thwarting innovation and dynamic competition. The framework of structural cheating not only has no such problems, but it offers a straightforward explanation: cheating future competition is tantamount to undercutting the ability of future competitors to create the innovative products or services that would differentiate future firms from their extant rivals. So not only does the structural cheating account ground the ethics of antitrust in different normative principles than the MFA—in procedures rather than outcomes—but it can also address a lacuna in the MFA’s classification of problematic outcomes via its intimate connection with the process of cheating in dynamic markets.

Supplementing Social and Political Accounts

Social and political consequences—not just economic consequences—have also traditionally motivated antitrust enforcement. So it is important to note, in closing, that the structural cheating account is not incompatible with these sociopolitical concerns. It can even provide a valuable supplement.

What are those motivations? Some take the morally problematic character of monopoly to lie in its threats to democracy. A notable example is Senator John Sherman, for whom the Sherman Antitrust Act of 1890 is named. Sherman argued that the great risk of monopoly is its endowing a monopolist with a “kingly prerogative, inconsistent with our form of government.”⁵⁶ His worry was that a monopoly would in effect be above the law, not subject to the checks and balances constitutional democracies institute to disperse power. Adam Smith expressed a similar concern when he wrote that a monopoly would resemble an “overgrown standing army,” dangerously capable of usurping the proper role of the state, and thus constituting a rival government in itself.⁵⁷ A contemporary version focuses on its bureaucratic manifestation, as monopolists or representatives of concentrated industries are poised to “capture” regulatory agencies, or otherwise exert an outsized effect on legislators or lawmakers, especially toward the ends of rent-seeking or other private gains at the expense of the public good.

A related concern is that monopoly power or excessive corporate concentration threatens the classical liberal values of freedom and autonomy. Anderson’s (2017) worries about dictatorial or coercive “private government” in workplaces are exacerbated by monopolistic firms, for example. Those who analyze freedom as a state of non-domination (e.g., Pettit, 2014) may see those subject to corporate power as rendered unfree.⁵⁸ Competition allowing self-determination and coercion protection through choice is also a prominent historical theme motivating strong antitrust enforcement (e.g., Doctorow, 2022; Wu, 2018).

A third traditional worry concerns economic opportunity. The “American dream” traditionally includes being one’s own boss or running one’s own business. Yet monopolist or monopsonist control of crucial sectors of the labor market plausibly leads to declines in entrepreneurship and small business ownership,⁵⁹ preventing many from attaining economic independence or self-sufficiency.⁶⁰

A fourth concern is wealth and income inequality. Economists and legal scholars are increasingly linking the lack of antitrust enforcement since the Reagan years with the dramatically increasing wealth inequality over that same period, with all the polarization and destabilization of society that comes with it.⁶¹ And given the potential link between corporate concentration and inflation, which hits harder for those with fewer means, there are reasons to think social mobility is diminished by concentrated markets as well, exacerbating inequality even further.⁶²

That these big-picture sociopolitical values have traditionally motivated antitrust enforcement may lead one to object that it is the threat to these values that explain the wrongmaking features of antitrust violations, as opposed to the account in terms of cheating given here.⁶³ But there is no reason there can’t be several wrongmaking features of antitrust violations, which may be additive or supplementary. While cheating is a prima facie and pro tanto wrong in and of itself, for example, it can also lead to harmful further consequences, among them threats to democracy, autonomy, freedom, equality, and opportunity.⁶⁴ And it is certainly not implausible to think the structural cheater’s willingness to dominate a market could spread to other domains, in addition to the adverse economic effects of concentration. At the same time, an advantage of the structural cheating account is that it can be applied in cases where larger social or political charges cannot reasonably be levied, or when those problems have yet to emerge.⁶⁵ Structural cheating focuses on the process of competition, or the means by which market power is obtained, not only the end-result. In such cases, the structural cheating account provides a useful supplement to accounts that focus on the political consequences of consolidated market power.

Conclusion

Regardless of its outcome, the FTC’s case against Facebook presents an opportunity to consider the meaning of fair competition under capitalism. In this paper, I have argued that “anticompetitive conduct and unfair methods of competition” can be construed in terms of cheating, in particular a form of cheating I call ‘structural cheating.’ I have argued this construal has many advantages. It explains the wrongs of unfair competition irrespective of its consequences, provides a corrective to false claims of merit or desert, and provides a superior explanation of the wrongmaking features of market-concentrating antitrust violations compared to the Market Failures Approach, especially regarding dynamic competition and innovation in tech markets. More generally, this paper provides a normative lens with which to view antitrust violations, supplementing social or political accounts by focusing on the means of accumulating power, not just its effects, as important as they are. In so doing, this paper shows why it’s not only winning or losing that counts in capitalism, but how one plays the game.