The Epic v. Apple: The Ninth Circuit Should Look at Some Fundamentals

The District Court opinion contains fundamental infirmities: product market definition, scope of the injunction, and reliance on an unduly vague state law. Before deciding whether to restructure a global industry, the 9th Circuit should consider these fundamentals.

It is unclear whether these points will be raised by the parties.

Apple states that it requires in app purchases (IAP) in order to enhance customer security and privacy and to collect a commission for use of its intellectual property, products, tools, and services. It contends the record contains no evidence that IAP had anticompetitive effects in any relevant market or on Epic Games, Inc., and that IAP must be viewed as part of a two-sided market, citing the Supreme Court’s decision in American Express.

Epic wants to notify customers in the App Store that they can pay a lower price by not using Apple’s payment system, and to link the app in the App Store to  alternative facilities.

Is the Relevant Product Market Properly Defined?

Most of the facts relied on by the District Court do not go to how an antitrust product market is defined.

Antitrust product markets are defined by interchangeability of use, i.e., cross-elasticity of demand. If a seller were to raise price, make terms more onerous, or degrade quality, to what alternatives could a customer switch?

The court stated the standard of law correctly: a relevant product market includes “the product at issue as well as all economic substitutes for the product.” “Economic substitutes have reasonable interchangeability of use or sufficient cross-elasticity of demand with the relevant product.” Economic substitutes are sellers “who have actual or potential ability to deprive each other of significant levels of business.” Opinion at 119-120. In other words, to what alternatives can a customer turn on approximately the same economic terms, and do those alternative sellers produce enough to meet the demand?

Most of the factors relied on by the District Court are not logically related to this standard. Administrative groupings of products in a store, quantities of revenue from that group of products, profitability, tabs in the store, types of technology used in the product, characteristics of the customers, and nature of pricing structures do not address the question of where customers would turn if prices were to be raised, or terms or quality degraded.

The court found that gaming apps are lucrative and constitute a majority of app store revenue; the public recognizes a distinction between gaming and non-gaming apps; game developers employ specialized technology; there is a subset of developers who make only gaming apps; game apps use distinct pricing structures and are sold by specialized vendors and in general are subject to unique and emerging competitive pressures. There is no information about switching, ability to switch, inclination to switch and to what.

The opinion lacks a logical nexus between (i) the factors pointed to and (ii) the standard of law for defining an antitrust product market.

Analogize this to a brick and mortar pharmacy. The pharma products sold at the back of the store might be lucrative (revenue, % of profits), but they are not substitutes for one another. Store design puts the pharmacist at the back of the store in order to draw customers in (special tab for product type). Many drugs are developed using analogous research and testing techniques; packaged consumer goods at the front of the store may use similar market research, focus group testing, etc. (specialized R&D, technology, product design). Pricing systems for pharma are unique, influenced by insurance and government reimbursement programs (pricing structures, generalized competitive pressures). Some companies only make certain types of pharmaceuticals; some make only certain types of packaged consumer goods (specialized developers and vendors). The public recognizes pharmacies as a distinct type of store (games are different from non-games). But none of this informs us about what competes with what. What are a customer’s real world alternatives?

Apple may not raise this issue because there are very few other mobile app stores one can go to. Epic may not raise the point if it did not put actual switching evidence in the record, not only between stores, but between games that customers might be inclined to use instead of Fortnight, for example, games on which Apple lowered its commission to 15%. Epic seems to have focused instead on its argument that the App Store payment system by itself could be a relevant market.

But product market definition is so fundamental that the Ninth Circuit might want to examine this issue before mandating a global industry to be restructured.

Does the UCL Injunction Purport to Regulate Foreign and Interstate Commerce?

The court found the relevant geographic market is global except China. ”Apple’s engagement in that market does not change based on national borders.” Opinion at 133. Yet the court enjoined the anti-steering provisions in the nationwide U.S. only.

Given the court’s geographic market definition, the injunction necessarily purports to use state law (the California Unfair Competition Law) to regulate foreign and interstate commerce. As the court found, one cannot draw a line between nationwide and global commerce in this industry.

Under Article I, Section 8 of the U.S. Constitution, the power to regulate commerce with foreign nations and among the several states lies with Congress, not the individual states. Congress has exercised its power to attempt to define the foreign reach of federal antitrust law in the Foreign Trade Antitrust Improvements Act of 1976. The District Court’s opinion omits any mention of this statute, of Article I, Section 8, or of the inconsistency between the injunction and the geographic market finding. The injunction may be an impermissible reach of state law into foreign and interstate commerce.

The UCL’s Vague Standards Aggravate the Problem

The burden on interstate and foreign commerce is aggravated by the vagueness of the UCL. Concepts of incipiency and tethering to unwritten policies of a state legislature are extremely vague. Finding Epic, a multi-billion dollar corporation to be a “quasi-consumer,” the District Court determined that conduct not prohibited by any statute could still be unlawful as “unfair.” Conduct could be “unfair” if it is an “incipient violation” of some other law or violates the “policy or spirit” of another law, an issue that is equitable in nature and on which courts have “broad discretion.” 

Perhaps the court’s prohibition against the anti-steering provision was right, maybe not. The problem is the standard was not knowable in advance and the decision de facto applies the standard globally given the court’s geographic market definition and nationally, not just in California.

Regulating an industry, especially one of this size, should have clear standards of law knowable without years of litigation, so that business can be planned. A global business should not be expected to ascertain the standards of consumer protection law (some of which are knowable in advance, some not) of 50 states plus the District of Columbia and conform to all of them globally. The state law standards should only govern conduct in or affecting each respective state.

I have no relationship to, receive no compensation from, and have no financial interest in any party to this case, nor to any potential complainant or any government law enforcement agency. I write as an observer with 46 years of U.S. antitrust experience to contribute to dialog on the matter.