Many of the problems with the FTC complaint against Facebook that was dismissed in court last month appear to be fixable. However, given that the case still has to confront other unfavorable realities, the dismissal very likely improves the chances for new antitrust legislation.
Last month, a federal District Court dismissed the antitrust complaints filed against Facebook by the Federal Trade Commission (FTC) and, separately, by a coalition of state Attorneys General. The dismissal order by Judge James E. Boasberg, an appointee of President Barak Obama and a centrist, raises two questions: first, what are the problems with the current FTC complaint, and can they be fixed? The court gave the FTC 30 days to make repairs. Second, does the dismissal affect the prospects for the numerous antitrust bills currently pending in Congress? The dismissal of the states’ complaint was based on timeliness and will be harder to fix. It can be appealed, however, over a dispute concerning whether the state AGs are “public” or “private” enforcers of the federal antitrust laws.
These cases were at the motion to dismiss stage, which means that Judge Boasberg was obliged to accept the pleadings as true and, except in narrow circumstances, not examine the evidence supporting them. In 2007, however, the Supreme Court raised the pleading standards, requiring more detailed facts, and here the FTC’s complaint fell short. Judge Boasberg was particularly concerned about several things: most importantly, the market was not well defined. As a result, “the FTC has failed to plead enough facts to plausibly establish … that Facebook has monopoly power….” Further, the FTC’s unsupported allegation that Facebook had a “60 percent-plus” share of this market was “speculative and conclusory.”
The court also noted, however, that this case involved “no ordinary or intuitive market.” Digital “two-sided” markets, where consumers are on one side and advertisers and other suppliers on the other, yield several idiosyncrasies, including prices of zero on the consumer side. The failure to delineate the market also undermined the challenge to Facebook’s acquisitions of Instagram and WhatsApp, because the FTC had sued under the monopolization statute rather than the merger statute. Finally, the court concluded that Facebook’s refusals to share data or permit interoperability with rivals “cannot form the basis for [antitrust] liability” no matter what the market.
What counts in a monopolization case is not absolute size, which Facebook clearly has, but rather market power, which is the power to profit by raising prices above cost. Measuring market power traditionally requires a plaintiff to identify a “relevant market” and show that the defendant has a dominant share of it. Relevant markets usually include products or services that are close substitutes for one another, like “automobiles.” The FTC’s alleged market for “personal social networking” (PSN) services combined activities such as posting videos, photographs, messaging, and the like that are not substitutes for one another. It would be somewhat like defining a market of “small appliances” that includes toasters, blenders, and coffee makers. The courts do recognize a concept of “cluster” markets for firms such as hospitals, which group noncompeting services together. This would require the FTC to show that customers find the particular aggregation of products desirable or that there are cost savings from providing them together; and second, that the aggregation would be difficult for a new competitor to duplicate. The FTC should be able to fix this problem. In addition, the complaint did not identify who Facebook’s competitors were—another important oversight, but one that can be fixed.
The court was also troubled by the fact that when services are offered free, coming up with a unit for measuring market shares is difficult. Revenue on the user side will not work as a measure, but there are alternatives, such as the number of users or usage measured by time. The judge appeared to believe that these were mainly problems with how the issues were articulated, not with substance. The FTC may also be able to fix these by repleading. Finally, the FTC claimed that it could measure market power directly, which is a particularly promising alternative in digital markets. Direct measurement can be more accurate but is technically complex. However, the FTC did not put enough details about this methodology into its complaint.
The court also concluded that it was not unlawful for Facebook to refuse interconnection, or “interoperability” with competing apps. This claim is governed by the US law of refusal to deal with competitors, which is particularly difficult for plaintiffs. Repleading this claim is less promising, because the extremely narrow refusal-to-deal rules apply even to confirmed monopolists. A strong case exists that the law of refusal-to-deal should be loosened up, particularly in networked markets where interoperability is important, but the courts have not yet gotten that message. This district judge is of course bound to follow precedent.
The court found that the FTC had made a better case in its challenge to Facebook’s acquisitions of Instagram and WhatsApp. The complaint itself recounts how Facebook regarded Instagram as a dangerous emergent competitor and purchased it for just that reason. It also agreed with the FTC that mergers can be challenged as acts of monopolization when they prevent the emergence of new competitors. The court ultimately dismissed the complaint on this issue because it had not alleged Facebook’s market power sufficiently under the monopolization standard. A promising way to address this problem would be to replead the merger challenge under the merger statute, §7 of the Clayton Act, which does not assess the same monopoly power requirement.
Many of the problems with the FTC complaint appear to be fixable, but the case still confronts other realities. The filing of these complaints occurred amid a wave of popular enthusiasm for doing something about the four major Big Tech platforms. Nevertheless, they must still be litigated in federal courts that exhibit a strong anti-enforcement bias. A district judge has only limited power to do much about that. The text of the antitrust laws is in fact quite broad. Section 2 of the Sherman Act, which the FTC enforces via the FTC Act, reaches every act of “monopolization.” The anti-merger statute reaches every merger where the effect “may be” to “substantially lessen competition.” But the law has been narrowed by many years of judicial interpretation, particularly since the 1980s.
These dismissals will very likely improve the chances for new legislation, which has recently been proposed in both the House and the Senate. Between those in Congress who are enthusiastic for radical changes and those who favor doing nothing are some who would prefer to wait and see if existing antitrust law can do the job. These dismissals provide some evidence that they cannot. The question then is whether to wait through a lengthy appeals process that could take years or do something legislatively. If the answer is the latter, it is important to act wisely. Breaking up Facebook is legally possible, but it would injure users’ experience. Reducing product quality or variety is not antitrust’s purpose. It causes unnecessary injury to consumers as well as suppliers of inputs, including labor. There are more promising alternatives, including the prevention of mergers, stronger and more explicit rules identifying and prohibiting anticompetitive agreements, and mandatory interoperability rules that would require data sharing and give a foothold to potential rivals.
Disclosure: Herbert Hovenkamp serves in an unpaid capacity on the international board of advisors for the Global Antitrust Institute (GAI) at the Antonin Scalia Law School at the George Mason University, where he debates against the Federalist Society on antitrust issues. GAI has received funding from tech companies, including Google, Amazon and Qualcomm.
Editor’s note: The author’s disclosure has been updated after publication.