On April 24, 2023, the U.S. Court of Appeals for the Ninth Circuit affirmed a district court decision holding that Apple does not violate federal antitrust law by restricting app distribution on iOS devices to its App Store or by requiring in-app purchases on iOS devices to use Apple's in-app payment processor.
The decision implicates competition issues that have garnered considerable legislative interest. This Legal Sidebar provides an overview of general antitrust principles, the litigation challenging Apple's app-distribution restrictions, and the Ninth Circuit's decision. It concludes with options for Congress.
Antitrust Basics
The Sherman Antitrust Act contains two core prohibitions. Section 1 of the statute prohibits concerted conduct that unreasonably restrains trade. Section 2 bars unilateral activities that "monopolize" commerce.
Under Section 1, some practices—like price fixing among competitors—are per se illegal, meaning courts need not inquire into their effects in individual cases. Most restraints of trade, however, are evaluated under a standard called the "rule of reason." Under the rule of reason, courts conduct fact-specific assessments of a defendant's market power and the details of a challenged agreement to determine a restraint's competitive effects.
Plaintiffs can establish that a defendant possesses market power either directly (via evidence of supracompetitive prices and restricted output) or indirectly (via evidence that the defendant possesses a large market share and that the market contains entry barriers).
The indirect route requires plaintiffs to define a relevant market—an exercise that turns on the range of items that are "reasonably interchangeable" with one another. In conducting this inquiry, many courts employ a methodology under which a proposed market is iteratively expanded until a hypothetical monopolist in that market would be able to profitably impose a "small but significant non-transitory increase in price." Stated differently, market definition involves an attempt to identify the smallest grouping of products for which consumer substitution to other items would not prevent a monopolist from profitably raising prices.
In some cases, one brand of a product can itself constitute a relevant antitrust market. For example, courts have recognized the existence of aftermarkets—markets in which demand for a product or service (e.g., razor blades) depends on an earlier purchase of a durable good (e.g., razors) in a foremarket. In acknowledging the possibility of relevant aftermarkets, courts have reasoned that switching costs and consumer unawareness of aftermarket restrictions (e.g., unawareness that a durable good is incompatible with complements produced by rivals) may prevent foremarket competition from disciplining a firm's aftermarket conduct.
Once a market has been defined, courts assess whether a defendant possesses the market power necessary for liability in a Section 1 rule-of-reason case. Section 2 plaintiffs are held to a higher burden and must establish that a defendant possesses monopoly power, which entails "something greater" than market power. While the Supreme Court has not identified a minimum market share for a monopolization claim, lower courts have tended to conclude that a share of more than 70% is necessary to prove monopoly power via indirect evidence.
In both Section 1 rule-of-reason cases and Section 2 litigation, plaintiffs also must establish that the defendant engaged in anticompetitive conduct. This inquiry usually involves some variation of a multistep burden-shifting framework that courts have described as being "essentially the same" under both provisions.
In that framework, the plaintiff has the initial burden to prove that challenged conduct has a substantial anticompetitive effect. If the plaintiff does so, the burden shifts to the defendant to show a procompetitive justification for its conduct. If the defendant makes this showing, the burden shifts back to the plaintiff to demonstrate that the relevant procompetitive benefits could be reasonably achieved through less anticompetitive means. Some courts have also added a fourth step in which they balance a restraint's anticompetitive and procompetitive effects.
Epic Games v. Apple
Apple's iOS and Google's Android are the leading operating systems for mobile devices. According to a 2020 congressional report, iOS runs on more than half of smartphones and tablets in the United States. iOS is often referred to as a "walled garden," as opposed to an open ecosystem in which software developers can transact with consumers without Apple serving as an intermediary.
More specifically, the district court found—and Apple did not deny—that Apple restricts app distribution on iOS devices to its App Store and reviews apps to ensure that they meet certain security, privacy, and reliability standards before making them available. The company also requires developers to use Apple's payment processor for any purchases that occur within the developers' apps. Apple generally collects a 30% commission for this service. Additionally, Apple prohibits developers from using certain communication methods—like in-app links—to inform users about out-of-app payment options. Apple imposes these restrictions through a licensing agreement that developers must execute to distribute apps on iOS devices.
The District Court Decision
In 2020, Epic Games—the developer of the videogame Fortnite—challenged these limitations under both the Sherman Act and California law. Epic argued that Apple possesses monopoly power in single-brand aftermarkets for iOS app distribution and payment processing for iOS apps and that the relevant restrictions harmed competition in those markets, in violation of Sections 1 and 2 of the Sherman Act. Epic also contended that Apple's requirement that in-app purchases use Apple's payment processor constitutes an unlawful tying arrangement. (Tying arrangements can be illegal under either a modified per se rule or the rule of reason.)
Apple denied Epic's allegations of monopoly power, arguing that the relevant market consists of video game distribution generally and thus includes sales involving game consoles, personal computers, and Android mobile devices. The firm also defended the challenged restrictions on the grounds that they enhance user privacy and security and allow Apple to monetize its intellectual property (IP).
In 2021, a federal district court held a bench trial in the case. The court ultimately issued a lengthy opinion in which it rejected Epic's federal antitrust claims.
On the issue of market definition, the court arrived at a conclusion that fell between the two parties' positions. Instead of a single-brand aftermarket for iOS app distribution or a market for video-game distribution generally, the court held that Apple competes in a market for digital mobile gaming transactions, which includes transactions on iOS and Android mobile devices. The district court rejected Epic's narrower proposed aftermarkets for several reasons, including Epic's failure to establish consumer unawareness of Apple's restrictions or to produce evidence of the magnitude of the relevant switching costs. The court ultimately concluded that Apple possesses market power—but not monopoly power—in the market for digital mobile gaming transactions.
After assessing Apple's market power, the district court proceeded to consider Epic's Section 1 claims. It ultimately concluded that those claims failed for multiple reasons.
First, the court held that the challenged restrictions did not represent concerted conduct within the meaning of Section 1 because Apple's licensing agreements with app developers were contracts of adhesion drafted by Apple and generally offered on a take-it-or-leave-it basis, rather than bargained-for agreements.
Second, the court determined in the alternative that the challenged restrictions were permissible under the rule of reason. In applying the rule of reason, the court found that
Epic established evidence of the restrictions' anticompetitive effects.
Apple proffered valid procompetitive justifications for the restrictions based on their promotion of security, privacy, and Apple's monetization of its IP. (While the court credited Apple's IP monetization justification generally, it rejected that justification "with respect to the [App Store's] 30% commission rate specifically.")
Epic failed to show that Apple could achieve the relevant procompetitive benefits through less restrictive means.
The district court also rejected Epic's tying claim on the ground that app distribution and in-app payment processing are not separate products. The court denied Epic's Section 2 claims based on the firm's failure to establish that Apple possesses monopoly power and prove that the procompetitive benefits discussed above could be achieved through less restrictive means.
Although the court sided with Apple on the federal antitrust claims, it concluded that the company's antisteering provisions prohibiting certain communications about out-of-app payment options violated California competition law.
The Ninth Circuit's Decision
Epic and Apple both appealed. Epic argued that the district court erred by rejecting Epic's proposed single-brand aftermarkets; holding that contracts of adhesion fall outside the scope of Section 1; misapplying the rule of reason; and concluding that app distribution and in-app payment processing are not separate products. Apple challenged the district court's determination that its antisteering rule violated California competition law.
While the Ninth Circuit panel determined that the district court committed several errors, it held that those errors were harmless and affirmed the court's decision with respect to both the federal antitrust claims and California law. (One judge on the panel concurred in part and dissented in part, arguing that some errors identified by the majority were not harmless and that the court should have thus remanded the case to the district court.)
In addressing market definition, the Ninth Circuit explained that the governing case law required Epic to prove that consumers are generally unaware of Apple's app-distribution restrictions in order to establish the existence of single-brand aftermarkets. The panel concluded that the district court did not clearly err in finding that Epic failed to establish this lack of consumer knowledge. This finding, the majority reasoned, rendered harmless a separate error in the district court's market-definition analysis.
Proceeding to the lower court's assessment of Apple's conduct, the Ninth Circuit held that the court erred when it concluded that contracts of adhesion like Apple's developer agreements fall outside the scope of Section 1. The majority determined that this error was harmless, however, because the district court properly applied the rule of reason in the alternative.
In affirming the district court's rule-of-reason analysis, the panel held that the court did not err when it rejected Epic's argument that Apple could secure the relevant procompetitive benefits by implementing a "notarization model" derived from Apple's Mac computer operating system. Under this model, Apple would permit developers to distribute iOS apps outside of its App Store but append a warning to those apps indicating that Apple had not reviewed them for malware. The Ninth Circuit affirmed the district court's finding that Epic failed to show that the notarization model would be as effective as Apple's restrictions in improving user security and privacy or how such a model would allow Apple to monetize its IP.
The panel also reviewed the district court's omission of an explicit fourth step in applying the rule of reason. The appellate court expressed skepticism about the usefulness of a totality-of-the-circumstances balancing step given a defendant's burden at step two to proffer non-pretextual justifications for its conduct. Nevertheless, despite acknowledging some inconsistencies in the case law, the court interpreted circuit precedent as requiring a fourth balancing step. It explained, however, that such a step will ordinarily not entail a heavy analytical lift. Rather, the court indicated that in "most instances," the rule of reason's fourth step will "require nothing more than . . . briefly confirming the result suggested by a step-three failure: that a business practice without a less restrictive alternative is not, on balance, anticompetitive." Because the district court's analysis included a sentence offering such confirmation, the Ninth Circuit held that the failure to expressly reference a fourth step was harmless.
The Ninth Circuit then addressed Epic's Section 1 tying claim. While the panel held that the district court erred in determining that app distribution and in-app payment processing are not separate products, the majority concluded that this error was harmless. In particular, the majority concluded that the modified per se rule against tying does not apply to ties involving platform software products, which often involve iterative innovations that bundle and unbundle various functionalities. (In reaching this conclusion, the court adopted one of the key holdings from the D.C. Circuit's 2001 decision in United States v. Microsoft.) Because Epic's tying claim was governed by the rule of reason rather than the modified per se rule, the court determined that it represented a repackaging of Epic's other Section 1 claim, which was unsuccessful for the reasons discussed above.
In concluding its review of the federal antitrust claims, the Ninth Circuit affirmed the district court's rejection of Epic's Section 2 allegations on the grounds that the court did not clearly err in finding that Apple lacks monopoly power or in determining that Apple's conduct was not anticompetitive.
The Ninth Circuit also affirmed the district court's conclusion that Apple's antisteering requirement violated California competition law, rejecting Apple's cross-appeal of that aspect of the decision.
The parties have until June 7 to petition for rehearing by the panel or the full Ninth Circuit.
Legislative Options
Congress has considered legislation that would have aimed to prohibit the restrictions targeted by Epic's lawsuit. In the 117th Congress, the Open App Markets Act (S. 2710, H.R. 5017, and H.R. 7030) would have required firms that own or control app stores with more than 50 million U.S. users to permit users to download apps through methods other than their app stores. The legislation also would have prohibited covered firms from tying their app stores to their payment processors and restricting communications between developers and app users regarding "legitimate business offers, such as pricing terms and product or service offerings."
Other bills in the 117th Congress would have made broader changes to the competition laws governing certain large technology firms. The American Innovation and Choice Online Act (S. 2992 and H.R. 3816), for example, would have imposed a variety of conduct rules—including rules regarding self-preferencing, tying, and interoperability—on designated tech platforms.
Congress also retains the option to leave the existing legal regime unchanged.