Abstract
A novel theory of antitrust law may be tested in the case of Federal Trade Commission (FTC) v. Facebook. It focuses on how pricing might be monopolistic even when the goods delivered to end users are zero-priced. While there is considerable political momentum behind a regulatory push to toughen antitrust sanctions on digital platforms in general and Facebook in particular, the economic theory behind the Government’s antitrust case is shown to be uncompelling. That does not mean it will necessarily be rejected by a given court, but the chances of the case succeeding and then surviving the full gamut of appeals is low. However, that predicted outcome may well calibrate the considerable space between the existing legal equilibrium and an emerging electoral policy equilibrium. If so, the expected outcome may well fuel the populist movement pushing legislation to fundamentally alter the antitrust statutes.
I. Introduction
Public choice research has shown that antitrust prosecutions tend to be heavily influenced by intersectional political factions and rent-seeking rivalries across industry segments. 1 In the present climate, in which antitrust enforcement issues have risen to national prominence, 2 a popular cause may elicit critical, perhaps marginally determinative, political support. While antitrust issues are typically complicated matters of law and have difficulty achieving salience in conversations influencing electoral politics, that burden may be met under current circumstances.
In the high-profile debate—perhaps rising to a furor—over market power wielded by large digital platforms, one perceived barrier to bringing antimonopoly cases is that many of the platforms charge zero prices for their services. This relates, in differing degrees, to the GAFAM enterprises: Google, Amazon, Facebook, Apple, and Microsoft. Facebook offers social media access, globally providing over two billion users with essentially free web hosting and additional digital connection algorithms that allow them to easily find and communicate with friends, family, experts, celebrities, businesses, and affinity groups all over the world. Google, likewise, supplies an array of products that are generally accessible simply by signing up; creating a name and password allows one unlimited Internet search, viewing of user-generated YouTube content, travel directions, restaurant reviews, and a long list of other digital services. In either case, the audience created for the services supplied allows the host platform to monetize a second side of the market, often by selling advertising. The resulting commercial communications tends to be made more valuable by the use of data about the end users’ online behavior, allowing messages to hit highly targeted audiences, producing much better responses than sales pitches randomly distributed across networks.
Some have noted that, despite achieving very high market shares in particular Internet services, regulators have been reluctant to use antitrust regulatory tools due to the nature of Internet pricing. Cabral writes that the zero-pricing issue is thought to have shielded monopoly power,
3
and quotes a New York Times essay that complains that antitrust regulation has been overwhelmingly focused on the welfare of the consumer. No cost to the consumer, no problem. That opened the door for Google, Facebook, Apple and Amazon—which offered digital services that were cheap or free—to become immensely profitable and powerful.
4
The assertion (noted but not championed by Cabral) is curious. First, low prices enable consumer gains, and these constitute the benefits that disappear when anticompetitive conduct replaces efficient market rivalry. It is not surprising that “cheap” prices would be unchallenged under pro-consumer antitrust laws. Second, Apple and Amazon, two of the four examples offered, primarily distribute paid goods. Apple is well known for the premiums charged for its products, a high-priced policy that nearly drove the firm to extinction in the 1990s. Moreover, the firm explicitly guarantees buyers that it will not sell online behavioral data, a rejection of the two-sided models pursued on other platforms. While Amazon borrowed Walmart’s slogan, “Everyday Low Prices,” it has generated enormous eCommerce volumes by selling retail goods. This is also true of their Amazon Prime delivery of digital video content. That multiple business models have, indeed, enabled rival firms to become “immensely profitable and powerful” goes to the conclusion tendered by Cabral who endorses Cremer’s view that markets, operating under existing law and economics institutions, can best supply emerging markets: “There is no need to rethink the fundamental goals of competition law in the light of the ‘digital revolution.’” 5 With high prices, low prices, or zero prices, the same basic features of economic rivalry exist and can be evaluated to determine how market power is being created and utilized.
A pathbreaking antitrust complaint filed by the Federal Trade Commission (FTC) against Facebook, if successful in ongoing litigation, potentially relies on that implication of economic theory.
6
The lawsuit, filed on Dec. 9, 2020, by FTC accuses Facebook
7
of monopolistic actions violating the FTC Act.
8
A parallel antitrust complaint was filed on the same day by forty-eight states, but that legal action has been dismissed by a federal court.
9
The federal prosecution continues, and is due to be tried in about mid-2024.
10
As summarized by the Financial Times, the principle allegation is that Facebook has defended its social media monopoly for years with a “buy or bury” approach to its rivals, such as Instagram and WhatsApp. If the FTC . . . can convince the courts that Facebook’s allegedly anti-competitive behaviour has damaged the market, the default solution is nothing short of the dismantling of the company.
11
The key analytical innovation provided in the case is that free goods offered by Facebook are said to be monopolistically priced. The seeming contradiction has to do with quality adjustments that have been made by Facebook as it has grown larger and allegedly achieved a greater degree of control over the market. When it was smaller and had more, and more important, rivals, the firm allegedly provided higher quality service. The margin which the FTC focuses on is privacy. As Facebook has achieved market power, old protections against the use of personal information have been loosened, claims the FTC. This relaxation is tantamount to a quality reduction. Even as the nominal price paid by end users to access Facebook services remains constant, at zero, the quality-adjusted price is claimed to have risen.
This argument is logically possible. It attempts to work the antitrust allegation into a consumer welfare model. But the theory will likely prove uncompelling in a court of law, because the story of consumer harm from better targeted advertising does not fit the facts of quality reduction. Of course, legal verdicts are rendered with variance; the prediction of a Government loss underlying this article may prove incorrect.
Section II will review the FTC’s argument about how Facebook rose to achieve market power. Section III considers the monopoly pricing claims made by the Government, and the FTC’s reliance on Facebook CEO Mark Zuckerberg’s smoking gun email saying “it is better to buy than to compete.” Section IV ends with a surmise about the likely political reaction to Facebook’s victory in the suit brought by the FTC.
II. Facebook’s Rise to Market Dominance
The Government’s case against Facebook argues that the company started in a competitive mode, then vanquished rivals, bought out emerging threats, and achieved monopoly power, securing its position by making its platform closed to third-party applications. At that point, it effectively raised prices, lowering consumer welfare and violating the FTC Act. It is key that the alleged market power sprang, as per the FTC, not from efficiency but from its 2012 merger with Instagram and its 2014 acquisition of WhatsApp. This accumulation and exploitation of market power led to the antitrust transgression.
The basic business model pursued by many digital platforms, including Facebook, is to offer “free” (zero-priced) services to Internet users in exchange for access to the data generated by their resulting usage. A customer creates a log-in, without charge, and then reads content and makes posts, exchanges “Likes” or emojis with other users, searches for old friends or lost family members, and hosts what is tantamount to a Facebook-supplied personal website for commenting on public events, publicizing commercial enterprises, organizing an online community of enthusiasts, or leading a cause. In these activities, information about individual preferences and their likely online behavior is gleaned by Facebook. These data help target advertising messages to audiences likely to be interested in certain activities, organizations, goods or services, and demand specific products. The improved ad-targeting creates a premium paid for advertising message slots and is the value captured by the Facebook platform. Gains from trade in this two-sided business model are distributed to three sets of parties: end users (zero-priced access to valuable social media), advertisers (paying customers), and Facebook (the platform owner).
As in all antitrust cases, market definition is key. The Government generally wants to narrow the relevant product market (where few rivals exist), the defendant tries to expand it (claiming substitutes everywhere). So here. The FTC complaint cites Facebook Chief Operations Officer, Sheryl Sandberg, speaking to investors in 2012: “We’re not TV, we’re not search. We are social advertising.”
12
Thus, the FTC characterizes the firm’s competitive superiority: Facebook in particular has a preeminent ability to target users with advertising due to its scale, its high level of user engagement, and its ability to track users both on and off Facebook properties to measure outcomes . . . Facebook’s social advertising business, serving ads to users of its personal social networks reflecting its vast access to data, is extraordinarily profitable.
13
This success may appear to be an efficient outcome. Yet, the asserted anticompetitive twist is that “Facebook has maintained its monopoly position by buying up companies that present competitive threats and by imposing restrictive policies that unjustifiably hinder actual or potential rivals that Facebook does not or cannot acquire.” 14 This aggregation strategy has underpinned the firm’s achievement: “Facebook holds monopoly power in the market for personal social networking.” 15
The market power associated with Facebook is not, by itself, anticompetitive. It needs to be demonstrated as harming consumers by prices above competitive levels that would obtain but for the illegal conduct of the defendant. Here that is claimed to involve two acquisitions and then a lock-down of the platform, imposing rules of access (by users and third-party vendors, perhaps emerging platforms themselves) that do not allow the data amassed by Facebook to be profitably used by rivals. 16 The “anticompetitive conduct,” writes the FTC “has had three main elements: acquiring Instagram, acquiring WhatsApp, and the anticompetitive conditioning of access to its platform to suppress competition.” 17 The acquisition of Insta in 2012 for $1 billion was problematic, says the FTC, because the photo-sharing application was arguably already competing with Facebook and would have likely become a highly successful direct substitute in social media had it been left an independent contender. Thus, despite the fact that Instagram was nascent (it had no revenues and only thirteen employees), the merger with Facebook “neutralized” a potential rival. 18 The 2014 merger with WhatsApp, for $19 billion, involved a far more substantial target, one with a robust business. Its messaging service was highly complementary to the primary Facebook social media platform and directly competitive to the Facebook Messenger service already available.
The Complaint contains a distinctive temporal aspect: the FTC attacks transactions that the Government had previously approved, yet leaving the Instagram and WhatsApp acquisitions unchallenged. The practice in antitrust enforcement, explicit since the 1976 Hart-Scott-Rodino Act, has been for federal agencies to have advance notice of all proposed mergers of a substantial size and then to decide, ex ante, to oppose the deal or to let it proceed uncontested. If the former, a trial will be scheduled should the merging parties elect to continue, and a judge will decide the matter under the antitrust laws. The typical cause of action in situations where the FTC or Department of Justice (DOJ) opposes a merger comes under Sec. 7 of the 1914 Clayton Act that makes illegal combinations that “may . . . substantially lessen competition . . . or tend to create a monopoly.” 19 An ex post attack on mergers seeking to divest integrated firms of their joint assets was found to be unduly disruptive. 20
In the 2020 FTC suit, the agency has declared that its earlier verdicts were incorrect; that Facebook had achieved monopoly power by 2011; that it is now appropriate to disentangle an existing platform.
21
It has elected to bring this case when the defendant’s arguments have already been written—by regulators, and not just in the United States. In approving the 2012 Facebook–Instagram combination, the U.K. antitrust office wrote, Facebook launched its mobile photo app in May 2012, weeks after it had announced that it would acquire Instagram. Facebook’s app has similar functionality to Instagram’s. It allows users to apply filters, tag photos, comment on photos, and post the photos to Facebook.
22
But there was much more to the market than Instagram and Facebook: “ . . . virtually all smart phones have a photo app installed by the . . . manufacturer.” 23 While such programs were technically similar to the social media apps, allowing for posting and sharing pictures, their more limited features did not make them “close substitutes to Instagram.” 24 But an array of stand-alone photo-sharing apps were “identified as being the closest substitutes for Instagram [including] Camera Awesome, Camera+, Flickr, Hipstamatic, Path and Pixable.” 25
While photo-sharing apps were just beginning to develop (hence, the entry by Facebook in 2012 and the emergence of Instagram as a promising upstart), the approaches taken by the merging firms exhibited disparate business models and complementary features. UK’s Office of Fair Trading (OFT) found that “Instagram is not currently a competitor to Facebook for advertising revenue and it has limited social networking functions.” The merger would allow the competitive advantages of Facebook to integrate with the photo-sharing popularity of Instagram, creating synergies, particularly in bringing desktop social media usage (where Facebook was strong) with emerging smart phone social media (where Insta appeared to be creating a successful platform). Noting these structural factors and considerable entry into the budding photo-sharing space, OFT found no competitive threat in the merger.
In 2014, the interest of Facebook executives, managing their now world-leading social media platform, was turning from emerging mobile uses to the rising popularity of messaging. It saw the service as potentially enhancing its communications network, or perhaps giving life to a new ’n improved competitor, such as WhatsApp, a texting service exhibiting rapid user growth. The firm had yet to enter Facebook’s market, but might soon do so, as memos inside Facebook noted. Many other texting apps were already popular, including those ubiquitously embedded as mobile phone features: Apple’s Facetime, WeChat (dominant in the Chinese market), Microsoft’s Skype and LinkedIn, Snapchat, and multiple Google programs.
In the intervening years, the analysis of the regulatory agencies has played out. The social media ecosystem burgeoned and, measured in usage or advertising revenues, has grown impressively. The marketplace exhibits substantial benefits from “network effects,” 26 creating value in the increasing ability of individual users to both find more information of interest to them and to broadcast their viewpoints to larger audiences. Positive feedback loops enable such platforms to grow rapidly, scale efficiently, and occasionally to become extremely valuable. If I “Like” you and you “Like” me, our connections might become something more than either sentiment would be by itself. The scope offered by growing nodes of interaction yields value that increases more than linearly—Metcalfe’s Law. 27
Facebook overwhelmed initial incumbents. The market tilted decisively away from Friendster and MySpace—even after its 2005 acquisition by News Corp, dubbed one of the five “Media Monopoly” firms said to control U.S. news and entertainment 28 —as Facebook surged. The newcomer’s competitive superiority has been associated with Facebook’s more cooperative business model in which innovative third-party programs were easily integrated by users under the platform’s “Open API” policy. 29 While the FTC cites the 2012 and 2014 mergers as decisive, it is curious that the Facebook’s movement to industry leadership was clearly initiated prior to either combination—the FTC Complaint accuses Facebook of asserting monopoly control of the industry by 2011. 30 It is indisputable that considerable economic value was created. While the FTC Complaint focuses only on company profitability, 31 the evidence suggests that consumer surplus has been a generous multiple of firm earnings. In 2017, experiments (in which subjects are offered payments to disengage from the service) revealed the median price required by customers to drop Facebook was $40 per week, far higher than the value of the firm to shareholders. 32
III. Monopoly Pricing at Zero
A. Exchange as Suspect Behavior
In addition to quality and service, the exchange between the social network and the user must account for any barter of goods in kind. In the case of Facebook, users agree to trade their data to the platform in exchange for its services.
33
Despite the generation of value for end users, there is purportedly a fly in the ointment. With the rising importance of the Facebook social network, competitive choices have been reduced for platform users. “Since toppling early rival MySpace and achieving monopoly power,” alleges the FTC, Facebook shored up its market power by key mergers and then reduced access to its network by potential rivals via rules making interconnection terms more onerous. 34 And Facebook has demonstrated its monopoly power by raising prices to end users via reduced privacy protections.
The Facebook service is made available to users on a barter agreement, with personal behavioral data traded for subscription access. 35 In 2007 and for some years thereafter, Facebook was an innovative firm, competing against established incumbents in social media. The FTC Amended Complaint 36 details how Facebook surpassed Friendster and MySpace by building “Facebook Platform,” easing the pathway for third-party software application developers, with tools to improve functionality, that could help build-out the burgeoning ecosystem. “Facebook Platform offered developers a unique distribution channel for their products and services” such that “developers were induced to rely on Facebook’s open access policies and invested in developing compatible products.” 37
One of the key features of this ecosystem, important to popular adoption by users, Dina Srinivasan writes, was that while “Facebook entered a competitive social media market [it] disintermediated competition by offering superior consumer privacy protections.” 38 But once it had attained dominance, it shifted gears. At that point, “after other social networks like MySpace and Orkut exited the market . . . Facebook was able to reverse course, and initiate consumer surveillance.” With greater market power, the demand curve facing the firm became less elastic, prodding the firm to raise price to a (new) profit-maximizing level. It did this by reducing privacy protections, extracting more (and more valuable) data about personal online behavior, and so making higher revenues from ad sales. 39
The idea, then, is that the user’s price to access Facebook social media rose, and output was reduced, revealing an anticompetitive outcome. A quality-adjusted admission fee might, if in evidence, be associated with a decline in consumer welfare. 40 Of course, there is another side to the market, the ad sales side, which might offset this asserted decline. 41 But the FTC does not focus on the ad market in its Complaint. It is reasonable to conclude that, with regard to competition for advertising dollars, social media platforms like Facebook compete against multiple other advertising sellers and the market would be defined quite broadly under antitrust law.
The approach by the FTC opens the way to a price theory argument of consumer harm, despite zero-pricing, in FTC v. Facebook. The Commission claims that, with monopoly power, FTC was able to restrict output by reducing quality with respect to privacy protections: Consumers have been harmed by a lack of sufficient competitive constraints on Facebook, which has enabled Facebook to exercise its monopoly power. Without meaningful competition, Facebook has been able to provide lower levels of service quality on privacy and data protection than it would have to provide in a competitive market.
42
Elsewhere, the FTC references this as “‘direct’ evidence that Facebook is a monopolist with the power to profitably reduce quality (which is equivalent to a price increase) and exclude competition” (FTC 2021B, 2). And it frontally attacks the notion that Facebook’s zero-priced access policy suggests competitive conditions, mocking Facebook’s spurious arguments that it cannot have market power because the price of its product is nominally “zero” . . . [T]hese assertions are bad law and bad economics, because the exercise of market power can manifest in ways other than increases in nominal prices charged to consumers.
43
But the FTC nowhere establishes that the privacy policies of Facebook are not preferred, given the bargain sought by customers (end users), and indeed offers evidence that the asserted quality-adjusted price increases imposed by Facebook were not price increases at all: “‘the ability to withstand significant user dissatisfaction while experience[ing] a minimal loss of user engagement’ is an indicator of market power.” 44 (FTC 2021B, 12, citing FTC). Actually, no. A monopolist should observe a decline in quantity demanded when increasing prices; indeed, the reaction will be demand elastic in equilibrium. Only by asserting a Cellophane Fallacy (see just below) does the FTC make the leap between the cited user agreement changes and Facebook pricing.
The exchange between social media users and Facebook no doubt involves quality issues, of which privacy protections are key. But the terms of contracts in allowing data to be used for ad-targeting is distinct from the data breach issue. The latter is an unambiguous loss to the consumer who has been assured that personal information will not be made public. The former, however, involves data sharing that does not unambiguously harm the consumer. Not only does that barter pay for access to the social media services sought, but it improves the selection of ads shown the user from the standpoint of the user’s interest and intention. The reason that Orvis pays a premium to place an ad on a page being viewed online by a person likely to be a fly-fishing enthusiast is the mirror image of the reason that a fly-fishing enthusiast would more likely prefer to have a message from Orvis pop up than an ad for, say, knitting supplies. The selection of audience via online usage patterns improves matching. Not only is it not a categorical reduction in quality to increase ad value, it is likely a general improvement. The user gets zero-priced access to Facebook, the Facebook network grows over time to become more valuable, and ads displayed for users better appeal to them. The quality depreciation cited is not compelling. The FTC’s claim characterizes gains from trade as anticonsumer and anticompetitive. 45
Security breaches or policy failures by platforms such as Facebook may harm consumers. The Cambridge Analytica controversy was a problem for users who found that their data had been enlisted, not via their consent, but by other Facebook users volunteering access to their personal networks—producing sensitive information about their online friends without the consent of said friends. Facebook, which conceded error in the matter, tightened its policies; nonetheless, it lost substantial market value, adjusted for overall stock price trends, in the year following the release of news about the breach (Hazlett 2019). It is an open question whether additional rules are needed, which are not without cost. A reduction in competition via small websites has been observed in Europe following the 2018 adoption of the GDPR. 46 The sharing of personal data for ad-targeting is a contract issue. Does the user accept the terms being offered? It is argued that consumers do not. 47 They are deceived into accepting lower privacy protections by tricky terms and false statements. If so, then contract law, and perhaps criminal fraud statutes, is available for addressing the illegal conveyance. The FTC itself regulates such alleged activity. Indeed, in 2019, Facebook was fined $5 billion for violations of user privacy. But that did not address a market power issue; the FTC filed a separate FTC Act Complaint the following year.
B. A “Cellophane Fallacy” 48
In the Supreme Court’s 1956 opinion in the DuPont antitrust case,
49
the defendant cited evidence as to the many competing products available to consumers, including aluminum foil, waxed paper, and polyethylene. Considering “all flexible wrapping materials” seemed to demonstrate that Du Pont faced an elastic demand for its cellophane products, and the company claimed it could exercise no substantial ability to profitably raise prices. The Supreme Court found this approach compelling: If a slight decrease in the price of cellophane causes a considerable number of customers of other flexible wrappings to switch to cellophane, it would be an indication that a high cross-elasticity of demand exists between them; that the products compete in the same market.
50
Alas, this became the Cellophane Fallacy. 51 In fact, price theory suggests that all firms, including monopolies, face firm demand curves that are necessarily elastic at the price–quantity combinations obtaining in equilibrium. A profit-maximizing firm operating in the inelastic portion of the firm’s demand curve could raise revenues by increasing price; while output declines with a price increase, it does so by a lower percentage (per the definition of inelasticity). Because Total Costs are constant (if marginal costs = 0) or declining (if marginal cost > 0) as outputs decline, profits necessarily increase as prices rise. This is true whenever firm demand is inelastic. Hence, to maximize profits, the firm (whether competitive or monopolistic) always charges prices that target the elastic portion of its demand curve.
The trick that Du Pont’s lawyers performed was an audience misdirection. While Du Pont’s cellophane faced many effective substitutes at the prices charged, had more competitive prices been in place, the cellophane product might have faced only inferior substitutes. 52 The FTC case against Facebook, interestingly, demonstrates an Inverse Cellophane Fallacy.
The FTC alleges that, upon acquiring monopoly power in social media, Facebook exploited the situation to effectively raise price by lowering privacy protections: “Without meaningful competition, Facebook has been able to provide lower levels of service quality on privacy and data protection than it would have to provide in a competitive market.” 53 While Facebook was forced, as a start-up, to offer superior terms to users, this allegedly gave way to “lower levels of service” in the monopolization phase. This allowed greater data flow to advertisers, profiting Facebook, according to the FTC.
The claim is distinct from losses resulting from privacy breeches or from (other) contractual failures by Facebook, which are subject to legal, regulatory, and economic sanctions. This is discussed below. What is relevant here is that the Commission’s case against Facebook states that more aggressive data harvesting constitutes a price increase for end users (even if the nominal price for platform access is $0.00, quality had declined) and that Facebook’s unique position is revealed: [H]istorical events indicate that even when Facebook’s conduct has caused significant user dissatisfaction, Facebook does not lose significant users or engagement to competitors. This is an indicator of market power . . . Facebook’s ability to withstand significant user dissatisfaction while experiencing a minimal loss of user engagement on Facebook Blue [the social media platform] indicates inelastic demand and market power.
54
This argument takes the general form of the Cellophane Fallacy, where marginal demand conditions are taken for a categorical characterization of the marketplace. Yet, instead of seeing abundant competition as in the Du Pont case, however, it sees no competition—no substitution away from Facebook services occurs in response to a price increase. Price theory predicts, however, that quantity demanded falls (given a downward sloping demand curve), and by more, in percentage terms, than price increases (demand is elastic at the margin). The argument given by the FTC is contradicted by theory, then, as the observed lack of consumer response implies that consumers did not value the asserted change as the Commission hypothesized—in other words, no quality-adjusted price increase was actually observed. 55
A similarly dubious argument appeared in the debate over the XM-Sirius merger in 2007–2008. 56 The National Association of Broadcasters, which opposed the combination as “merger to monopoly,” advanced the theory that XM radio had raised (pre-merger) prices for its satellite radio service from $9.95 to $12.95 and yet “subscriber growth continued at such a rapid pace [after] the price increase” that the evidence was said to “underscore the low elasticity of demand faced by” satellite radio providers.
IV. Merging to Monopoly
The FTC claims that Facebook has exerted monopoly power since 2011, and that the firm has protected its position since that time via the 2012 and 2014 acquisitions of Instagram and WhatsApp. Both are identified as platforms that offered loose substitutes for the social networking services of Facebook Blue, either of which might have become close substitutes had they been allowed to develop independently. The FTC seizes on a memo written by CEO Mark Zuckerberg in 2008 which noted, in part, “it is better to buy than compete,” 57 a phrase quoted four times in the Complaint. The allegation is made that Facebook could not “compete on the merits” 58 and was, instead, intent on acquiring upstart rivals by merger. This turns out to be a smoking gun lacking pop.
First, the FTC’s selected exhibit of the buy-not-compete strategy was taken from a time when Facebook’s internal growth was soaring and the firm was overtaking the market leader, NewsCorp/MySpace. 59 This market rise was facilitated by the advantages rendered users by Facebook, including the network effects spawned by the relative openness of the Facebook platform, and was not the result of mergers. 60 That Facebook managed to “topple” MySpace was a demonstration of competition on the merits, as per the FTC’s terms, and did not connote a strategy of merging to monopoly. 61 And if the buy-don’t-compete strategy was then in effect, it would have made Facebook a target, not an acquirer.
Second, as a matter of profit maximization, every firm would rather buy than compete conditional on the price being right. That does not imply that there is a categorical profit-maximizing rule that any merger will extinguish market competition or prove beneficial. The cost–benefit calculus is simply an extension of the trade-off analysis outlined in the classic article about how firms decide whether to use produce inputs themselves as opposed to purchasing them using “the price system.” 62
Third, as a practical matter, any strategy to inevitably buy rather than compete is doomed to failure. That is because there is a highly elastic supply of entrants willing to hold-up incumbents. It is a losing strategy to protect a monopoly by buying up all the real or imagined upstarts. That is why theories of predation embed credible threats to destroy entrants rather than reward them with capital gains via takeover premia.
Fourth, as an empirical matter, Facebook did not buy all of the platforms competing in social media. Not only did it not purchase, and then outcompete, Friendster, Orkut, MySpace, and Google+, but the FTC Complaint notes, Facebook has tried and failed to buy other companies that have drawn its competitive attention, including Twitter and Snapchat. For example, in lamenting that Twitter had “turned down [Facebook’s] offer” to be acquired in November 2008, Mr. Zuckerberg wrote: “I was looking forward to the extra time that would have given us to get our product in order without having to worry about a competitor growing.”
63
Yet “failing” to acquire a target is another way of saying that Facebook elected not to pay the asking price. That violates the “buy-don’t-compete” maxim, as Facebook has done many times. With the acquisitions of Instagram and WhatsApp, on the contrary, the firms have been well integrated in the platform. Instagram has been appraised as having a market value of $100 billion if a stand-alone company, 64 and WhatsApp, while not generating substantial revenues by itself, has attracted over one billion users, driving substantial demand for other elements of the Facebook platform (Jenkins 2020). Both of these acquisitions have been retained as brands, neither a “killer acquisition,” but extending the scope of the Facebook network.
V. The Backlash from FTC v. Facebook
The antitrust case against Facebook was filed by the FTC during a Republican administration and is being vigorously pursued during the Democratic administration that followed. There appears substantial bipartisan support to bring new regulatory restrictions, by antitrust or other means, to the large digital platforms in general, and Facebook, in particular. The FTC Act complaint against Facebook attempts to explain how Consumer Welfare may be reduced by monopolization where the price of the relevant service is free.
Yet the mergers in question were not only considered benign when reviewed a decade ago by U.S. and foreign regulators, they have proven successful in extending and improving the Facebook ecosystem. The claims made by the Government about Facebook inevitably pursuing mergers with potential rivals is theoretically dubious and is then contradicted by marketplace facts (including those presented in the FTC Complaint itself). The plausibility of quality-adjusted price increases is lacking. Contracts between Facebook users and the platform routinely create value by bartering service for data, and the increased use of data for advertisement targeting does not imply a categorical reduction in consumer surplus, but the reverse. There are many margins on which Facebook (and other digital platforms) may appropriate users (say, by contract breach) or generate social costs (say, by making detrimental activities less costly), and the possible existence and scope of such practices are being explored by legislatures and policy makers. Some identified practices of Facebook have already been sanctioned in law.
The irony is that the antitrust suit brought by the FTC may fail and have large impact in that failure. That a popular case is prosecuted and is yet rejected in court may spur further political support for statutory reforms that would generically extend the regulatory reach of antitrust enforcement agencies. Whatever the outcome of the antitrust case brought by the FTC against Facebook, the battle for enhanced antitrust enforcement against the practices of digital platforms may just be beginning.
Footnotes
Acknowledgements
The author thanks the organizers for the invitation and the participants for their generous comments.
1.
Roger Faith, Roger, David Leavens, & Robert Tollison, Antitrust Port Barrel, 25
2.
Antitrust policy generally is not a frontpage topic in political reporting. Yet it has become so. President Joe Biden devoted “valuable real estate” in his Mar. 1, 2022, State of the Union Speech to pledge action on antitrust enforcement, and both major parties have made antitrust reforms high-profile causes, focusing on new legislation governing large digital platforms. Leah Nylen, Dems Are Going after Big Tech. It’ll Affect Almost Everything You Do Online,
3.
Luis Cabral, Merger Policy in Digital Industries, 54
4.
D. Streitfeld, To Take down Big Tech, They First Need to Reinvent the Law,
5.
Cabral, supra note 3, at 1.
6.
This supposition is drawn from the FTC Complaint as well as from supporting documents arguing for such litigation but composed by scholars outside of the FTC. Dina Srinivasan, The Antitrust Case against Facebook: A Monopolist’s Pervasive Surveillance in Spite of Consumers’ Preference for Privacy, 16
7.
The firm has since renamed itself Meta. We use “Facebook,” as it is called in the 2020 lawsuit, or Meta. “Facebook Blue” is the internal name given to the company’s major social media platform.
8.
Federal Trade Commission v. Facebook, Inc., FTC Matter/File Number 191 0134 (Dec. 9, 2020) (“Complaint”).
9.
Salvador Rodriguez, Judge Dismisses FTC and State Antitrust Complaints against Facebook, CNBC (June 28, 2021). The suit by the states was dismissed with prejudice, ending that litigation. The court dismissed the FTC claims without prejudice and later reinstated the case when the Government refiled (and expanded) its analysis.
10.
The trial is “estimated to start 2 H 2024.” Robert Kaminski, Meta U.S. Antitrust Primer Tracker,
11.
Hannah Murphy, Kiran Stacey, & Kadhim Shubber, US Antitrust Charges Are ‘All-or-Nothing’ Attempt to Break up Facebook,
12.
Complaint at ¶ 14. This internal Facebook comment by a top executive appears selected by the FTC to buttress the FTC market definition. Counter arguments have been crafted, unsurprisingly, in Facebook briefs. The firm cites numerous rivals, including Google/YouTube, Apple iMessage, Snapchat, Twitter, Spotify, Netflix, Hulu, LinkedIn, and Strava. Memorandum in Support of Facebook, Inc.’s Motion to Dismiss FTC’s Complaint, U.S. District Court for the District of Columbia, Case No. 1:20-cv-03590-JEB (Mar. 10, 2021), 6 (“Facebook Memo 2021”).
13.
Complaint at ¶ 50.
14.
Complaint at ¶ 1.
15.
Complaint at ¶ 2.
16.
These exclusions, the FTC alleged, were interconnection barriers, “API rules blocking directing competing services, with APIs referring to ‘application programming interfaces’” (Complaint at ¶ 22). These strategic foreclosures via unfriendly APIs created an environment that “suppresses, deters, hinders, and eliminates personal social networking competition,” and “deprives users of . . . increased choice, quality, and innovation” (Complaint ¶ 27).
17.
Complaint at ¶ 3.
18.
“By acquiring Instagram, Facebook neutralized Instagram as an independent competitor to Facebook Blue. Since the acquisition, Facebook has taken actions to reduce the impact of Instagram on Facebook Blue, confirming that Instagram is a significant threat to Facebook’s personal social networking monopoly.” Complaint ¶ 102. “Facebook Blue” was the internal name for the Facebook social media platform. In fact, Instagram was but integrated into Facebook, run as a separate brand (still). It may have been “neutralized as an independent competitor,” of course, a conjecture that depends on what one assumes about the trajectory of what Insta would have been without the merger. This does distinguish the Instagram (and for identical reasons) and WhatsApp mergers from what have been dubbed “killer acquisitions.” On this issue, see Robert W. Crandall & Thomas W. Hazlett, Antitrust in the Digital Era, 65
19.
Clayton Act, Ch. 323, § 7, 38 Stat. 730, 731–32 (1914), amended at 15 U.S.C. § 18 (2012). See also: Peter C. Cartensen & Robert H. Lande, The Merger Incipiency Doctrine and the Importance of Redundant Competitors, 2018
20.
Kenneth G. Elzinga, The Antimerger Law: Pyrrhic Victories, 12
21.
“‘We currently expect that [the remedy] will include divestiture of Instagram and WhatsApp,’ the FTC said . . . William Kovacic, a former FTC chairman and a current law professor at George Washington University said: ‘The FTC is all-in on the pursuit of the break-up. I think it is fully committed’. Joel Mitnick, a partner specialising in antitrust at law firm Cadwalader, Wickersham & Taft, said: ‘They are saying to the court that anything short of a really dramatic restructure is not going to cure the anti-competitive effects here. It’s kind of an all-or-nothing roll of the dice.’” US Antitrust Charges Are ‘All-or-Nothing’ Attempt to Break up Facebook,
22.
United Kingdom, Office of Fair Trading, Anticipated acquisition by Facebook, Inc of Instagram Inc, ME/5525/12 (full text of decision published Aug. 22), par. 15 (“OFT 2012”).
23.
OFT 2012, par. 16.
24.
Id.
25.
Id.
26.
S. J. Liebowitz & Stephen E. Margolis, Network Externality: An Uncommon Tragedy, 8
27.
This is exactly the source of value that Facebook has asserted as its key to success, and it is identically the source of Facebook’s market power. The FTC takes the following passage, for instance, to buttress its claims about the unrivaled power of the network: “Facebook Chief Operating Officer (‘COO’) Sheryl Sandberg described Facebook Blue as the ‘world’s first global platform that lets marketers personalize their messages at unprecedented scale.’” Complaint at ¶ 13.
28.
Ben H. Bagdikian, The New Media Monopoly (7th ed. 2004).
29.
API refers to an “application programming interface.” Gwangjae Jung & Byungtae Lee, How Did Facebook Outspace MySpace with Open Innovation? An Analysis of Network Competition with Changes of Network Topology,
30.
“Facebook Blue has been the dominant personal social networking provider in the United States since at least 2011.” Complaint at ¶ 62.
31.
“Facebook’s unmatched position has provided it with staggering profits . . . Last year alone, Facebook generated revenues of more than $70 billion and profits of more than $18.5 billion.” Complaint at ¶ 4.
32.
Erik Brynjolfsson, Avinash Collis, & Felix Eggers, Using Massive Online Choice Experiments to Measure Changes in Well-Being, 116
33.
Morton & Dinielli supra note 6 at 8 (footnote omitted).
34.
Complaint at ¶ 5.
35.
Morton & Dinielli, supra note 6.
36.
Complaint, at ¶ 24–42.
37.
Complaint, at ¶ 39.
38.
Srinivasan, supra note 6 at 81.
39.
The FTC puts it this way: “Facebook has . . . degraded the user experience, including the misusing or mishandling of user data. For example, the FTC charged Facebook with engaging in a range of serious user privacy and related abuses in 2012 and 2019 . . . Facebook’s ability to harm users by decreasing product quality, without losing significant user engagement, indicates that Facebook has market power.” FTC v. Facebook, First Amended Complaint for Injunctive Relief, Case No.: 1:20-cv-03590-JEB (Aug. 19, 2021) (“Amended Complaint”) at ¶ 206.
40.
Herbert Hovenkamp, Progressive Antitrust, 2018 U. Ill. L. Rev. 71, 113 (2018).
41.
Offsets for welfare changes on either side of a two-sided market were set forth by the Supreme Court in Ohio v. American Express Co., 585 U.S. ___ (2018).
42.
Amended Complaint, at ¶ 221.
43.
Federal Trade Commission v. Facebook, Inc.: Memorandum of Law in Opposition to Defendant Facebook Inc.’s Motion to Dismiss, U.S. District Court for the District of Columbia, Case No. 1:20-cv-03590-JEB (Nov. 17, 2021) (“FTC Law Memo”) at 14.
44.
Id. at 12.
45.
The Commission pursues this reasoning in initiatives beyond FTC v. Facebook. As one financial analyst puts it, “[T]he FTC is targeting the business of targeted advertising and the data-driven Internet ecosystem at large.” Robert Kaminski, Internet: First FTC Steps on Privacy Regulations; Targeting Targeted Ads,
46.
GDPR is the General Data Protection Regulation. See Michal S. Gal & Aviv Oshrit, The Competitive Effects of the GDPR, 16
47.
Srinivasan, supra note 6, 92–97.
48.
This subsection is based on Thomas W. Hazlett, Some Dynamics of High-Tech Merger Analysis in General and with Respect to XM-Sirius, 4
49.
United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377 (1956).
50.
Id., at 400.
51.
Donald F. Turner, Antitrust Policy and the Cellophane Case, 70
52.
Lawrence J. White identifies the difficulty in producing such a hypothetical market alternative as the great challenge of market delineation in monopolization cases. White, The Dead Hand of Cellophane and the Federal Google and Facebook Antitrust Cases: Market Delineation Will Be Crucial, 67
53.
Amended Complaint at ¶ 221.
54.
Facebook Law Memo at 12 (quoting Amended Complaint at ¶ 206).
55.
It is also notable that the FTC asserts that Facebook “surpassed” both Friendster and MySpace as the leading social media platform in “early 2009,” and that at “least since 2011, Facebook has had monopoly power in the United States with respect to personal social networking” (Amended Complaint ¶ 236). The key changes in Facebook privacy policy took place in 2016, however, long after the asserted monopoly formed, casting further confusion on the agency’s theory that a change in market power drove a monopolistic price increase.
56.
The author served as an economic expert for the merging parties in that transaction.
57.
Complaint at ¶ 5 (emphasis added by the FTC).
58.
Complaint at ¶ 83.
59.
Gil Press, Why Facebook Triumphed over All Other Social Networks,
60.
Jung & Lee, supra note 29.
61.
Complaint at ¶ 5.
62.
R.H. Coase, The Nature of the Firm, 4
63.
Complaint at ¶ 73.
64.
Ellen Simon, How Instagram Makes Money,
Author’s Note
A draft of this paper was given at the Annual Meetings of the Public Choice Society (Mar. 9–12, 2022, in Nashville, Tennessee). In Spring 2022, as this paper was being written, the author was a Visiting Scholar at Hoover Institution, Stanford University.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.