New paper on the App Store shows why antitrust intervention is needed

Last April Damien and I published a paper titled “The Antitrust Case Against the Apple App Store,” where we explored various concerns raised by Apple’s App Store practices under Article 102 TFEU. We focused in particular on the obligation of apps offering “digital goods or services” (think e.g., of a game app) to use Apple’s proprietary payment solution In-App Purchase (IAP) to accept user payments and pay a related 30% commission on the transaction value (note: the obligation to use IAP is at the core of the Commission’s ongoing investigation launched this summer in response to Spotify’s complaint).

Based on publicly available information, we considered that the App Store could be a separate market for app distribution in itself, given the lack of competitive constraints from other distribution channels (Android app stores, side-loading, preinstalling or web apps). We then considered that Apple’s practices raise both exploitative and exclusionary concerns (as well as concerns over discrimination).

• Exploitative concerns: starting from the fact that the 30% commission is charged only on 16% of apps distributed on the App Store, we considered that such commission may be charged only for the additional services rendered to such apps, namely payment processing and related services. That led us to conclude that the 30% commission is excessive, given that payment processing fees typically do not exceed 5% of the transaction value. We also considered that Apple’s conduct could be unfair in breach of Article 102(a) TFEU, since (a) IAP results in the confiscation of the app developer’s customer relationship; and (b) Apple seems to apply its policies in a capricious and arbitrary fashion (which could raise concerns under the Autorite’s recent decision in Google Ads).

• Exclusionary concerns: observing Apple’s aggressive push into services to source additional revenue, we noted that Apple may use its role as operator of the App Store to distort downstream competition and favor its own apps. To illustrate the concerns raised by Apple’s dual role, we used four case studies: (1) the obligation to use IAP (which results in raising rival apps’ costs); (2) the removal of rival screen time / parental control apps from the App Store on privacy grounds; (3) concerns over diminished interoperability with rival apps (again on privacy grounds); and (4) concerns over self-preferencing in App Store search results.

We concluded our paper offering a range of remedies that would address the concerns identified, chief amongst which would be the removal of the obligation to use IAP.

A couple of months later Sven Völcker and Daniel Baker from Latham & Watkins published a long rebuttal of our initial paper, arguing there is no antitrust case against the App Store. The rebuttal is well written and I encourage you to read it if you are interested in the matter. Völcker and Baker contribute to the debate, as they provide the other side of the story, which up till now may be found only in Apple’s court filings in its ongoing litigation with Epic Games in the US. To cut a long story short, Völcker and Baker challenge us on the following issues:

• On the facts: there is a clear and principled distinction between apps that have to use IAP and those that do not. IAP is not a payment processing solution; rather, it is Apple’s “digital checkout” for the App Store, a monitoring mechanism which allows Apple to observe transactions between app developers and users as well as calculate and collect its 30% commission. If there was no obligation to use IAP, then it would be impossible for Apple to collect its commission. No commission is charged on apps offering “physical goods or services” (e.g., Uber or Lyft) as in that case Apple cannot verify the underlying transaction (e.g., whether the driver picked the user). The authors stress that the 30% commission reflects the value of the App Store as a distribution / customer acquisition channel, and is thus not a payment processing fee.

• On market definition: Völcker and Baker argue that a proper market definition should include at the very least Android app stores, as Google Play and the App Store compete vigorously for app developers. The authors argue that app developers primarily single-home when it comes to app stores, and thus do not have to be present on both app stores. A proper market definition should also include alternative customer acquisition channels, since apps may also sell subscriptions on the web. The conclusion is that Apple cannot be dominant in such a market.

• On exploitation: since the 30% commission is for distribution / customer acquisition, it cannot be excessive. The authors point out to other marketplaces charging similar rates and refer to the Analysis Group study commissioned by Apple. Völcker and Baker also dismiss our concerns over unfair trading conditions under Article 102(a) TFEU.

• On exclusion: Völcker and Baker make the point that Apple has no incentive to distort downstream competition, since that would diminish the value of the App Store as a vibrant marketplace with third-party apps. In turn, that would lead to users valuing less the App Store, to the effect that they would buy fewer iPhones (or would be willing to pay less for them). Given that hardware sales are still Apple’s primary revenue source, any exclusionary conduct would eventually backfire on Apple. The authors then criticize our analysis of the four case studies but you will have to read it for the details!

While thought-provoking, the rebuttal did not convince us. So Damien and I wrote a (long) reply, which we published last week. If you don’t have the luxury of time to read it, here is a quick summary:

• On the facts: we disagree that the distinction between apps that have to use IAP and those that do not is clear and principled. All apps admitted to the App Store benefit from the same services, yet only 16% have to pay a commission. This does not seem particularly principled. As for the view that IAP is the “digital checkout”, we noted that there are alternative methods to monitor transactions, such as APIs informing Apple each time a payment takes place (actually Apple already uses APIs to inform app developers each time a user makes a purchase through IAP). So the argument that without IAP there is no way for Apple to get a commission is simply wrong. As for the argument that Apple cannot verify transactions concerning “physical goods or services“, we noted that its ability to verify transactions concerning “digital goods or services” is also severely limited (Apple cannot see whether the in-app feature is released to the user), yet that has not prevented it from seeking a commission.

• On market definition:
  • Android app stores are unlikely to constrain Apple. In fact, the Commission’s decision in Google Android dismissed Google’s argument that Google Play competes vigorously with the App Store for app developers. The Commission also held on the basis of internal Google data and third-party reports that app developers generally multi-home, i.e. they are present on both app stores.
  • Völcker and Baker could in theory argue that any attempt of Apple to increase its IAP fees could be counteracted by app developers disabling IAP and / or steering their users to the web to purchase a subscription. However, this is unlikely to happen in practice, either because it would be impractical (e.g., for a game app the user would have to pause the game in her phone, go to the web to make a purchase and then return to the app) or because of Apple’s marketing restrictions which prohibit app developers from steering users to the web. While some apps can disable IAP, this exception is only limited to certain categories of apps (and importantly does not cover game apps, which are responsible for the majority of Apple’s IAP revenues).

• On exploitation:
  • Even if the 30% commission is not charged for payment processing, the benchmarks put forward by the authors and Analysis Group miss the mark. Going beyond excessive pricing, the way Apple has structured its commission, whereby 16% of apps in the App Store pay for 100%, could be problematic under Article 102(a) TFEU. Since the 16% of apps are called to finance the App Store for everyone (this is essentially a cross-subsidization scheme among app developers), they are called to pay for services not delivered to them. This may be unfair in light of the Commission’s decision in DSD – Duales System Deutschland, where it was held that a dominant firm cannot charge its customers independently of the costs for providing its services (DSD had structured its license fees in a way that customers would pay regardless of whether they actually made use of DSD’s waste collection services).
  • The obligation to use IAP may amount to an unfair trading condition, as it does not appear to be necessary and proportionate to achieve Apple’s proclaimed aims (a safe and friendly user experience can be offered by third-party payment solutions; IAP is not necessary for Apple to collect its commission, since technical alternatives exist). IAP results in Apple becoming the merchant of record and handling the customer relationship with the app user, even though it has not provided the underlying good or service. This creates a host of inefficiencies when it comes to billing and customer support – and may even create switching costs dissuading users from switching to Android.
  • Apple’s arbitrary definition and application of its policies (e.g., there is no clear and principled distinction between apps that have to use IAP and those that do not) could amount to exploitation. Much like Google in the Autorite’s decision in Google Ads, Apple has a special responsibility to define and apply its rules in an objective, transparent and non-discriminatory manner, which we doubt whether it has discharged.

• On exclusion: While Völcker and Baker’s incentives argument is interesting, it is rather simplistic and theoretical. The proposition that vertically integrated platforms have no incentive to exclude downstream competitors (a view echoing the “single monopoly profit” theorem of Chicago School) is subject to multiple exceptions. And Apple’s ability to extract value from third-party apps by charging high prices on the device is getting limited as the average replacement cycle of iPhones is getting longer (i.e. users purchase new iPhones less often), thus putting pressure on Apple to maintain its $ 2 trillion valuation by pushing aggressively into services. After all, distorting downstream competition does not necessarily mean Apple will eliminate third-party apps; it may simply mean weakening rivals and taking market share away from them. Will users respond by purchasing fewer iPhones? We doubt they would.

In light of the above we concluded our paper sticking to our original remedies proposal, but with some slight modifications: first, Apple should be mandated to remove the obligation to use IAP, but that would not mean it would have to forgo the payment of a commission (alternative methods to monitor transactions such as APIs could be used). Second, Apple should be mandated to charge apps offering “digital goods or services” only for the services actually delivered to them, and not for services delivered to the 84% of apps.

For more, read the paper and let us know what you think!