Northern District Of California Finds That Antitrust Claims Against Technology Platform Fail While California’s Unfair Competition Law Supports Limited Injunction
09/21/2021On September 10, 2021, Judge Yvonne Gonzalez Rogers of the United States District Court for the Northern District of California issued her post-trial decision in Epic Games, Inc. v. Apple Inc., No. 4:20-cv-05640-YGR (N. D. Cal. 2021). Plaintiff claimed that defendant’s developer policies violated Sections 1 and 2 of the Sherman Act and the Cartwright Act, California’s analogue to the Sherman Act, as well as California’s Unfair Competition Law (“UCL”). The Court, in a 185-page opinion, found that plaintiff did not meet its burden to show that defendant’s policies violated the antitrust laws and denied plaintiff the broad injunction that would have required substantial changes to defendant’s App Store business. However, the Court held that plaintiff was entitled to a limited injunction under the UCL as to defendant’s anti-steering restrictions. The Court also granted contract damages for defendant’s counterclaims against plaintiff.
This case’s factual and legal background is extensive. In brief, defendant distributes plaintiff’s and other developers’ games on its iOS App Store platform for use on defendant’s branded mobile devices. Defendant’s terms and conditions for distribution of plaintiff’s and other developers’ games are governed by defendant’s developer product licensing agreements (“DPLAs”). Among other conditions, defendant requires developers to use defendant’s payment processing system, pay defendant a 30 percent commission on any “in-app” purchase, and refrain from “steering” customers to a payment processing site outside of the app. “In-app” purchases are purchases of add-ons to the game, such as extra points, characters, or features, which are typically purchased after the initial purchase or downloading of the game. In summer 2020, defendant refused plaintiff’s request to allow users playing plaintiff’s popular Fortnite game on defendant’s mobile device to use alternative in-game payment options outside of defendant’s IAP system. Undeterred, plaintiff introduced a covert “hotfix” into the code of Fortnite’s new version which allowed players to bypass the IAP system. Upon discovery, defendant removed Fortnite from its App Store, and both parties suited up for the antitrust litigation battle, leading to a 3-week bench trial before Judge Gonzalez Rogers this spring.
A threshold question for all of plaintiff’s Sherman Act claims (both Sections 1 and 2) is the definition of the relevant market. After assessing both plaintiff and defendant’s proposed markets, the Court declined to accept either party’s proposed markets and instead crafted its own definition of the relevant submarket within the broader market of all video game transactions: global (sans China) digital mobile gaming transactions.
The Court found that on the record presented, plaintiff failed to prove its proposed relevant market(s). Plaintiff proposed three relevant markets: (1) Smartphone Operating Systems; (2) iOS App Distribution; and (3) iOS IAP Solutions, with the latter two as derivative aftermarkets (an aftermarket is a market for a second product whose demand is derived from a first product in the foremarket). Before addressing the merits of plaintiff’s proposed markets, the Court observed that plaintiff was incorrect in its approach to market definition. First, iOS IAP Solutions could not be a viable product market for antitrust purposes since IAP was not a product with its own market. Second, the App Store operates as a two-sided transaction platform that connects developers to consumers, and plaintiff failed to define the market to address these two-sided issues.
Despite these fundamental flaws, the Court went on to analyze and ultimately reject plaintiff’s theory of a foremarket with a “single-brand” aftermarket structure. Here, plaintiff’s market theory was that defendant had market power in Smartphone Operating Systems and based on its power in that market, “locked in” consumers and developers into its two aftermarkets (iOS App Distribution and iOS IAP Solutions). Since these latter two markets were defined only to include defendant’s brand, plaintiff argued that defendant was a complete monopolist of those markets. The Court observed that these kinds of single-brand markets are rarely appropriate as all companies would be monopolists if the market were defined to include only their branded products. To establish a single-brand aftermarket, plaintiff must prove, among other things, that consumers had no or limited advance knowledge of restrictions in the aftermarket(s), and were therefore unwittingly locked into those aftermarkets (iOS App Distribution and iOS IAP Solutions) when consumers made a purchase in the foremarket (here, the purchase of an iOS product). The Court found no evidence on the record that consumers were unaware that the App Store was the sole means of digital distribution on the iOS platform. Further, plaintiff failed to quantify any supposedly “hidden” costs of downloading apps. Also, any such costs were mitigated by increased cross-platform functionality and apps that make switching platforms and devices more convenient. The Court held that neither consumers nor developers were therefore locked-in to the App Store for their digital mobile game transactions. The Court also rejected defendant’s proposed market definition of all video gaming transactions, finding the relevant market for the purpose of antitrust analysis was the global submarket for digital mobile gaming transactions, excluding China.
Market and Monopoly Power
The Court assessed and ultimately found insufficient plaintiff’s showing that defendant had monopoly or market power as requisite for each of its antitrust claims under the Sherman Act. To prove market power (under Sherman Act Section 1) or the higher threshold of monopoly power (Section 2), plaintiff may present direct evidence (such as the ability to control prices or output market-wide) or indirect/circumstantial evidence (such as high market shares coupled with high market entry barriers).
According to the Court, defendant’s market share of digital mobile gaming transactions during the relevant period fluctuated between 52-57 percent, which was below the 65 percent needed for a prima facie case of monopoly power under Section 2 (although the exact threshold percentage adopted by different courts have varied). Further, in light of the increased output of mobile games transactions in the marketplace, the Court found that plaintiff had not proven that output had been reduced below competitive levels. The Court also pointed to evidence of competition both within the defined submarket from other mobile game platforms (i.e., Android) and within the broader video game market from other game platforms more broadly (i.e., consoles, PCs).
Even though the Court held that plaintiff had failed to define a market or prove monopoly/market power, it nevertheless analyzed the various forms of allegedly anticompetitive conduct plaintiff asserted. The Court ultimately found that plaintiff had failed to meet its burden.
Plaintiff alleged that defendant violated Section 1 and 2, including by requiring in its DPLA a 30 percent commission on all in-app purchases. Plaintiff also alleged that defendant violated Sections 1 and 2 by engaging in tying and depriving plaintiff and other developers of access to an “essential facility” needed to compete with defendant.
As to the Section 1 challenge to the DPLA conditions, the Court held that the DPLA did not meet the agreement element of Section 1 as it amounted to no more than defendants’ unilateral policy that was imposed on developers. Notwithstanding this finding, the Court went on to analyze the DPLA’s terms under the rule of reason framework. Under this framework, the Court found that defendant’s policies did have some anticompetitive effect in limiting the competitiveness of other app stores and reducing pressure to reduce commission rates. However, the Court also held that defendant had shown valid procompetitive reasons for the DPLA’s restrictions, namely that the restrictions improved iOS security and increased inter-brand competition with Google’s Android platform. The Court also found that protection of its intellectual property rights and the ability to obtain compensation for their use also provided a valid justification for the restrictions, though not necessarily related to the level of commission. In the final step of the rule of reason analysis, the Court found that plaintiff failed to rebut defendant’s justifications because its two proffered alternative models that defendant hypothetically could use for its App Store were not proven to be as effective in the procompetitive goals of ensuring security and reliability on the defendant’s platform. The Court used a similar framework and reasoning for rejecting the Section 2 claims related to the DPLA.
The Court also disposed of plaintiffs’ tying and essential facilities claims. The plaintiff’s tying theory was that defendant violated the antitrust laws by tying together app distribution and IAP. The Court rejected the tying claim because IAP was not a standalone product, as required to maintain a tying claim, but rather a feature that was integrated into the broader iOS ecosystem. The Court held that defendant’s App Store was not an “essential facility” because it was not “essential” to distribute apps to consumers, and multiple other ways exist to distribute apps to consumers, including Android, other game stores, and websites.
California’s Unfair Competition Law
In contrast to plaintiff’s antitrust claims, which require a showing of harm to competition in a properly defined market, the Court found that a California UCL claim should be analyzed under a broad and more amorphous “unfairness” standard where even conduct that merely “threatens an incipient violation of an antitrust law” can violate the UCL. Under this standard, the Court held that defendant was not entitled to prevent developers from providing prompts (either in-app or outside of the app) to users directing them to alternative payment methods or from communicating directly with users when the user has given permission. Finding that a less restricted flow of information could increase competition, and that, in contrast to the defendant’s policies with regard to distribution, the defendant did not have substantial justifications for the anti-steering restrictions, the Court found that plaintiff had met its burden for an injunction under the UCL as to the defendant’s anti-steering provisions.
Plaintiff has already filed its notice of appeal in this action. Also, while a similar putative class action by developers reached a settlement shortly before the Court issued its opinion, a putative class action by consumers based on similar theories remains active, with the parties currently briefing class certification.
This decision is an important addition to the growing body of antitrust and competition law relevant to analyzing competition in platform and two-sided markets, especially in the technology space. The Court’s rejection of the “single-brand market” theory proffered by the plaintiff here is consistent with well-established principles and consistent with allowing marketplace participants to determine whether they prefer an “open system” platform or ecosystem or a tightly controlled “walled garden” approach. Similarly, the Court’s recognition that courts are not price setters, and only in the most unusual circumstances can a plaintiff maintain an antitrust claim simply by alleging that a defendant’s prices are too high, even when that defendant has significant market share and profits, is also consistent with well-established principles. Further, the Court was concerned with entering a broad injunction that had the potential to undermine the important interests in security and privacy that defendant offered to justify its tight control of the platform, to fundamentally alter the business model of a successful and innovative company, and perhaps to stifle, rather than to promote, competition. As the Court succinctly held, “Success is not illegal.”
With regard to the UCL injunction, this is by far the most prominent use of the vague “threatens an incipient violation of the antitrust laws” standard since the California Supreme Court articulated it more than two decades ago. In entering the injunction, the Court appeared to be largely driven by its finding that the prohibitions on communications relating to lower-priced alternatives served no legitimate purpose but the restriction of competition. In this regard, this conduct differed in a fundamental way from the other challenged restrictions that the Court found served important procompetitive interests, primarily in promoting the security and functioning of the platform, even if this conduct also had the effect of restricting competition. How far the UCL can be stretched to attack alleged anticompetitive conduct that does not violate the antitrust laws is a question that will be tested in future cases.