The App Store has a ‘Too Big To Fail’ problem

The term ‘too big to fail’ is commonly associated with the various bank bailouts during the global financial crisis of 2008-2010, but the phrase actually predates that. The term’s first recorded use in a financial context is from 1984, in reference to the federal bailout of the Continental Illinois National Bank and Trust Co. (Continental), which had inhaled debt products produced by another bank, Penn Square, over the latter half of the 1970s.

Penn Square issued risky loans to energy companies, keeping only a small portion of those loans on its balance sheet and selling the rest to other banks in an early form of securitization. For a number of reasons* — the “oil glut” of the early 1980s and a reversal of easy money Fed policy under Fed Chairman Paul Volcker — many of the loans issued by Penn Square entered default and, despite petitioning the Federal Reserve and the FDIC for emergy loans, the bank was allowed to fail in 1982. But Continental, which had purchased $1BN in loans from Penn Square and held $40BN in assets on its balance sheet, was saved by the Federal Reserve with a $3.6BN loan and a $4.5BN line of credit organized for it from some of the nation’s largest banks.

Continental was allowed to live because its collapse would create catastrophic, systemic shocks that endangered the broader economy. In a 1984 congressional hearing on the Continental bailout, Stewart McKinney, a Republican Congressman, said to Paul Volcker:

Mr. Chairman, let us not bandy words. We have a new kind of bank. It is called too big to fail.

We are witnessing the emergence of a similar phenomenon on the iOS App Store. The presence of certain apps is so imperative to the App Store’s broader consumer appeal that Apple has no choice but to allow those apps to defy its various platform rules and restrictions. These apps are too big to fail: if Apple were to enforce its rules against these apps and block them from the App Store, it would suffer extreme commercial consequences. These apps tend to be:

  • Ubiquitous. These apps feature massive, globe-spanning user bases numbering in the hundreds of millions or billions. These apps are mostly household names, and many have existed for years and in some cases, predate Google Play and the App Store;
  • Multi-platform. These apps are functionally accessible from across many different form factors and not just on mobile. This multi-platform characteristic allows many of these apps to process payments on the web in a way that only dubiously aligns with existing platform payments policy (although this is changing). The multi-platform property of these apps also provides users with many different opportunities for interaction and reduces dependency on any one platform (but especially mobile);
  • Habitual. These apps benefit from habitual, daily engagement and sit prominently in their users’ lives. Deep user affinity and regular use means consumers demand that their computing devices feature these apps and may make purchasing decisions with that in mind.

I hinted at the idea of apps that are too big to fail in my three-part series on the future of mobile content platforms: I argued that the ubiquitous nature of many high-profile services puts the onus on mobile platform operators to support those apps and not on the app developers to hew closely to the specific platform policies. I describe this idea in the series as the “content dependency hierarchy”: a new or mobile-only product is dependent on the mobile platform operators for distribution, but a multi-platform product with an existing, internet-scale user base isn’t.

At some point in a product’s growth trajectory, assuming it can be interacted with across many different hardware form factors and not just mobile, a mobile platform operator becomes more dependent on the presence of that product in its store catalog than the app developer does, which lends extraordinary leverage to the developer in complying with platform rules. Many of these such apps exist already, and they are too big for the mobile platforms to exclude from their storefronts over platform guidelines compliance.

The canonical example of an app that is too big to fail is Roblox, as I describe in Why is Roblox allowed to stream games on iOS?. Roblox’s mobile app is a catalog of games that can be instantly loaded and played within the app itself. In September 2020, Apple updated its App Store developer guidelines to include new restrictions related specifically to games streaming services:

4.9.1: Each streaming game must be submitted to the App Store as an individual app so that it has an App Store product page, appears in charts and search, has user ratings and review, can be managed with ScreenTime and other parental control apps, appears on the user’s device, etc.

4.9.2: Streaming game services may offer a catalog app on the App Store to help users sign up for the service and find the games on the App Store, provided that the app adheres to all guidelines, including offering users the option to pay for a subscription with in-app purchase and use Sign in with Apple. All the games included in the catalog app must link to an individual App Store product page.

Games featured within Roblox are not offered as stand-alone downloads on the App Store, which would ostensibly violate guideline 4.9.2. It also seems unrealistic that the games are screened by Apple before appearing in the Roblox store, given that the games are developed and published to the Roblox store by users as user-generated content. This would violate guideline 4.9.1.

Despite these very obvious contraventions of App Store policy, Roblox has not been removed from the App Store; rather, Roblox declared that the individual pieces of content it makes available via its app are not ‘games’ but are rather ‘experiences’ that don’t trigger either of the new guidelines around games streaming. During the Epic Games v. Apple trial, when asked why Roblox was allowed to exist in the App Store, Apple’s Senior Director of Marketing argued that Roblox’s content doesn’t constitute gaming because games must have well-defined beginnings and endings. Revealed during the discovery phase of the trial was that this same person expressed surprise at Roblox’s approval to the App Store, given that it was initially rejected because Roblox offered, “streaming games which we don’t like, since these games don’t come into review.”

Why is this awkward construction of Roblox’s content offering as ‘experiences’ allowed to satisfy the App Store policy guidelines above? Because Roblox is too big to fail: not only is the game a perpetually Top Grossing app on iPhone, but critically, over half of all American children under the age of 16 played the game in 2020. Apple needs to be present in the lives of this cohort in order to ensure its future platform lock-in. If Apple requires from Roblox full compliance with its guidelines and Roblox abandons (or underinvests into) the platform, Roblox’s users — which demand access to the product across desktop, tablet, and smartphone devices — could potentially flee to Android, introducing systemic risk to the iOS economy. Google, of course, feels the same pressure, although Google operates the Stadia streaming service and, likely as a result, has not prohibited games streaming services from being published to Android.

A second obvious example of a product that is too big to fail for mobile platforms and which has been given meaningful leeway to skirt platform policy by both Apple and Google is Netflix. Last Summer, Netflix announced that it would begin bundling mobile games into its subscription offering, and in November, those games saw a staggered release on Android first, followed by iOS one week later. Netflix’s games are discoverable in a dedicated icon row within the Netflix app. Note that this doesn’t violate the “no app stores within the app store” App Store guideline, as all of Netflix’s games — while developed by third parties — are published under the Netflix label.

Clicking on a game icon within the Netflix home screen takes the user to a native Netflix storefront that appears designed to mimic an App Store product page. Clicking the “Get” button on that Netflix storefront takes the user to the App Store, where they may download the Netflix game. Once the game is loaded, the user is automatically logged into that game via their Netflix account (see the Netflix logo at the top right of the screen at the end of the video below).

There are a few interesting points about this discovery process to consider:

  • The first is that Netflix sends users to an intermediary product page before sending them to the App Store — presumably because Netflix assumes very few users will discover these games outside of the Netflix app (although the games are discoverable from within the App Store);
  • The second is that a user is automatically authenticated into a game with their Netflix account after downloading it;
  • The third is that none of the games published by Netflix, as of this writing, offer in-app purchases or feature advertising, meaning they are monetized exclusively through the Netflix subscription;
  • And the last detail of note is that these games are not playable without an active Netflix subscription.

In June 2020, a controversy erupted around the treatment of an email app called Hey. Hey saw its app updates rejected from the App Store because it didn’t offer functionality to users that didn’t already have an active subscription; upon initial app open, the user was confronted with only a login form. Apple’s Phil Schiller explained at the time that the app update rejection owed to the fact that, “you download the app and it doesn’t work, that’s not what we want on the store.” Schiller further explained that some apps were exempt from this standard by the Reader App rule, which allows “reader apps” — apps that offer content that was either previously purchased by a user or is available on a subscription basis (eg. Netflix!) — to gate functionality behind a login. Schiller further clarified that email apps aren’t classified as Reader Apps. Hey’s app updates were ultimately approved when the company began offering a free trial that could be instantiated from within the app.

Games clearly aren’t reader apps. Perhaps Apple justifies its treatment of Netflix with respect to its games because Netflix does offer free trials, and in all likelihood, Netflix’s games can be downloaded during an active free trial. But upon downloading any of Netflix’s games when not logged into a Netflix account, the games are unusable: the only content presented to the player is a login screen.

Why does Apple allow Netflix to offer subscriber-only games on the App Store that aren’t functional for non-subscribers? Because Netflix is too big to fail. In strictly enforcing App Store policy against Netflix’s product design choices, Apple would introduce systemic risk to the iOS economy because consumers demand that Netflix be present on all of their devices. Netflix is either being treated differently with respect to App Store policy than was, or Apple is accommodating an extremely tenuous, if not flimsy, defense of Netflix’s behavior (free trial with the Netflix app) as proof of compliance. In either case, outwardly, the App Store guidelines appear to be enforced inconsistently, especially for large, multi-platform products. Other examples of this type of treatment abound.

Edit: it’s worth pointing out that Netflix does allow users to subscribe to the Netflix streaming service with an IAP within its games. That this IAP satisfies the Multiplatform Services carve-out is tenuous, given that the subscription is for a separate service, Netflix, that merely unlocks use of the games. Nonetheless, it could be pointed to as compliance with App Store policy on a technicality.

Apple’s App Store developer policy is currently under a tremendous amount of scrutiny, both from developers themselves as well as from governments and regulatory bodies across the globe. I’ve attempted to chronicle the App Store policy changes catalyzed by this pressure. But new changes emerge almost on a monthly basis, as with Apple’s recent disclosure that it will allow for alternative payments for dating apps published to the Netherlands’ App Store incarnation, as ordered by the Dutch competition authority, but that it will nonetheless collect a platform fee from any payments processed outside of iTunes. As Apple makes exceptions for products that are too big to fail, it also cedes ground — albeit slowly and incrementally, mostly propelled into the future by regulation and legislation — on its App Store payments policies.

But where Apple has not relented is with mobile-only games, which, as was revealed in the Epic Games v. Apple lawsuit, generate the vast majority of the App Store’s revenue. This is somewhat tautologically true, given that, really, only mobile-first games are strictly subjected to App Store payments policies, as I detail in The App Store is the Games Store (for instance, Netflix has not offered a subscription option within its app, across iOS and Android, for many years. Google went so far as to cancel all active YouTube Red in-app subscriptions before asking those users to re-subscribe on the web).

This is an important consideration because it helps to articulate why the counterpoint to the notion of too big to fail on mobile doesn’t invalidate the general thesis: Fortnite. Fortnite is a scaled, multi-platform game with a rabid fanbase. Why wasn’t it “bailed out” (to continue the banking metaphor) when it flaunted the App Store’s rules? I think there are a number of reasons for this.

First, Epic did not approach the situation diplomatically, effectively taunting Apple into a lawsuit with Project Liberty.

But second and more importantly, Apple simply can’t afford to create exemptions for mobile games because mobile gaming is the revenue base for the App Store. Apple can look the other way on App Store policy for all other categories of apps — and it mostly is — but retaining its App Store gaming revenue is critically important to the company. Other initiatives from the company, such as advertising, can cover revenue leakage from non-gaming app categories, but if pure-play, mobile-only games are allowed to monetize outside of the App Store, or to operate in ways not sanctioned by Apple, then Apple’s services revenue is put at risk.

This creates a very fine line for Apple to navigate, especially in the case of Netflix. Apple must allow the products that are too big to fail to skirt (or appear to skirt) its App Store policies while preserving the agency and legitimacy to enforce those policies for the developers of mobile-only games. What will Apple do when a traditional mobile gaming studio attempts to monetize its portfolio of games exclusively through web subscriptions?

*This introduction was originally far longer and more detailed before I edited it down substantially. I recommend that anyone interested in understanding asset bubbles resulting from easy money Fed policy, and the financial crises they can cause, to read Belly Up: The Collapse of the Penn Square Bank

Photo by Daniel Lloyd Blunk-Fernández on Unsplash