Apple, Its Control Over the iPhone, and The Internet
Chapter One: The Creation of Today’s Internet and the Needs of Tomorrow’s
One of the neat things about the internet is who created it, why, and how.
Throughout the 1960s to 1990s, the foundation of today’s internet was built through a variety of consortiums and informal working groups composed of government research labs, public universities and independent technologists. These typically not-for-profit collectives typically focused on establishing common standards that would help them share information from one server to another (i.e. messages or files), and in doing so, make it easier to collaborate on future technologies, projects and ideas.
The internet’s quirky provenance is responsible for many of its best modern attributes. Today, everyone can create content on the internet, everyone is technically capable of accessing everything on the internet, and every web page on the internet can connect to another without the user needing to change browser, device or client. This flexibility, interoperability and universality isn’t by decree - there’s no Head of the Internet mandating the right to create, host or access/connect to a website.
Instead, this is a byproduct of the internet’s open standards, programming, markup languages and so on. These ensure that a user doesn’t need to pay for a web browser or to load a website, nor does the owner of a website need to pay for the code used by their website. Services like Zoom also work because they leverage actively maintained standards that are free to use and designed to support any device. Every device maker needs to support these standards in order to have happy customers. Some of us now recoil against aggressive data collection, ad insertions and tracking, but this is partly because we don’t need to give these up. The use of standards, such as HTML, means that browser extensions can block ads or tracking. We may be willing to give up data for a free service, but the Internet’s makeup means we don’t have to.
The benefits here are hard to overstate. But just imagine, for example, how the internet might differ if it had been created by multinational media conglomerates in order to sell things, serve ads, harvest user data for profits, or control your end-to-end experience (something AT&T and AOL both tried and failed to pull off). Downloading a .JPG could cost money, with a .PNG costing 50% more. Teleconferencing software might have required the use of a broadband operator’s app or portal (e.g. Welcome to your Xfinity Browser™, click here for Xfinitybook™ or XfinityCalls™ powered by Zoom™). Websites might only work in Internet Explorer or Chrome - and need to pay a given browser an annual fee for the privilege. Or maybe users would need to pay their broadband provider extra fees to read certain programming languages or use a given web technology (“This website requires Xfinity Premium with 3D Rendering”). Microsoft was sued, in part, due to the bundling of Internet Explorer with the Windows OS. If a corporation invented the Internet, would they have even allowed competing browsers (the literal gateway to the web)? Would they allow the user to do whatever they wanted on these browsers, or access (and modify) whichever sites they chose? Regardless of the specific differences, it’s likely internet penetration would be lower, as would usage and associated commerce/value.
The web’s cross-platform, standards-based and non-profit origins are inseparable from the internet’s rapid growth, the trillions of dollars in companies that have been founded over the past 30 years, and the positive societal impact of these companies (e.g. a drop in the cost and increase in the quality of communications, the reduction in gatekeeper power, lowered transaction fees, etc.).
Today’s heavily conglomerated internet giants remain mindful of the fact that open APIs, common standards, exportable data, etc., all help grow both the internet technology acceptance model and, in most cases, their own bottom lines. But these companies are less concerned with how the overall market grows than their share and control of this growth. Technology companies, almost by definition, prefer that the market build on top of or through them than have new entrants build around or in competition with them. As a result, the same companies that emerged thanks to openness tend to reject these principles where they might undermine their strategic position.
This isn’t an unusual disposition for a for-profit company, but the implications are particularly powerful in digital markets. In the offline world, free market economics enable robust competition that typically delivers the best products, variety and prices for consumers, and moderates the power of the strongest market players (if only due to diseconomies of scale). Online, incredibly powerful network effects and zero-cost marginal revenues/distribution has enabled dominant platforms to push back against the open nature of the internet, forcing consumers and creators to use them as universal intermediaries, and subduing standard market forces.
Right now, we are on the cusp of the next internet. The terms used for this future vary and the degree to which you believe in one label or vision is not particularly relevant. And the technologies to design, enable and support the fullest version of this are as far from the capabilities of 2021 as the 1990s Internet is from us today. But what matters is that a growing share of our time will be spent within virtual spaces and with virtual goods — for education, work, health, politics and leisure. Sometimes these spaces and goods will be purely virtual, other times virtual twins of physical ones, and sometimes augmented reality. For related reasons, a growing percentage of our income will be spent on virtual assets, goods, experiences — many of which we’ll be able to sell, trade, share, use or improve. And of course, enormous new industries, marketplaces and resources will emerge to enable these opportunities, with novel types of labor, skills, professions and certifications invented to serve them.
There’s no way for this future to be developed as the original internet was. The US government and public research institutions led the development of the information superhighway in part because few in private businesses understood the commercial potential of a World Wide Web, but also because these non-profits were essentially the only entities with the computational talent, resources and ambitions to build it. Conversely, all of today’s mega-tech companies are deep believers in this future internet, and are best positioned (via resources and talent) to build it.
The most important impediment today is Apple. Although no company has done more to propel the last 15 years of the internet, its policies are unlikely to produce the most prosperous overall ecosystem and do not lay a strong groundwork for the “next Internet”. Instead, Apple is inhibiting this future Internet. And it does so via tolls, controls, and technologies that not only deny what made and still makes the open web so powerful, but also prevents competition, and prioritize Apple’s own profits.
Chapter Two: There’s Apple, then GAFM
Apple’s adverse effects stem from three interconnected and increasingly powerful elements.
First, Apple believes it knows better than consumers, the developer community, and the market at large. This includes which technologies should be adopted by the internet community, the role of user agency (and the extent to which users are capable of informed consent), and common business practices around user data, privacy and security.
This disposition led Apple to limit the role of the marketplace on its device (something the company did not do for the Mac). Specifically, Apple mandates that on its devices:
All applications would need to be distributed by the Apple App Store, and could not be directly downloaded from the app maker or a third party app store
All applications distributed by the App Store would need to be approved by Apple, and approval would be contingent upon an extensive list of policies and requirements
All applications distributed by the Apple App Store would need to conform with the App Store’s billing policies, which typically meant that Apple was the exclusive in-app biller for iOS applications
These three layers are technically different — the ability to use an operating system driver has nothing to do with the ability to distribute an app to the consumer or charge for them for a given piece of content or function. However, Apple’s iOS devices forced them together. If a developer wants to make an app, they need access to native drivers. And if a developer wants access to native drivers, their apps need to be distributed by the App Store. And if they want to use the App Store, they must adhere to Apple’s policies, which means using Apple’s billing systems.
Apple’s control and integration allowed it to offer a best-in-class mobile experience that also helped repel the most pernicious aspects of the online world; onboard and engage less technically savvy users; and develop a richly monetized app ecosystem. This, in turn, led to unprecedented success. Today, the iPhone has 66% market share in the United States, 75% of U.S. App Store revenues, and over 80% of time spent on the mobile internet (iOS’s share of physical e-commerce transactions is likely somewhere in the middle of this range). And this dominance is also growing. Eighty percent of U.S. teenagers have iPhones and the device held 90% of smartphone activations in the week after Christmas 2020.
While Apple’s closed approach is why the company’s products are so successful, the enormity of this success is what makes this approach so problematic. There is no proprietary, closed system that affects more lives on a daily basis than that of iOS. Due to this fact, Apple has become a de facto regulator for the internet; a single for-profit body that wields enormous soft, hard, and even accidental power.
There are almost no large companies in the world that can live without a mobile app (these apps are sometimes available only to a company’s employees). Due to iOS’s share of users, spend and time, this means having an iOS app. And having an iOS app means complying with all of Apple’s policies and requirements.
Consider Apple’s smackdown of Travis Kalanick-era Uber. To prevent fraud and identify users even after they deleted the app, Uber had been using a technique known as fingerprinting. According to a report in The New York Times, Kalanick immediately ended this practice after being threatened by Apple CEO Tim Cook. The point is not that this is a bad outcome. Instead, it’s remarkable that Uber was brought to heel by Apple after having spent years flaunting real world regulations globally, often actively campaigning against them and even mobilizing its users to the cause.
Another example is Apple’s upcoming changes to the Identifier for Advertisers (IDFA), a tool that has enabled advertisers and app developers to identify and track a user by a unique device signifier, without the user providing any account information. In a unilateral decision, Apple announced in 2019 that it would shift its IDFA solution from “opt out” to “opt in” in 2021. Whether or not you agree with the policy, the effects will be seismic. Between 70 to 90% of users are expected to “opt out” when prompted later this year - a move estimated to reduce the 2021 revenues of Facebook and Google by between $5 billion to $20 billion. Crucially, Apple’s policy adjustment was not prompted by any real world law or push. Indeed, it's not clear legislators have either the power or inclination for such a move. Instead, it reflects part of Apple’s new privacy initiatives and distaste for ad-targeting (it’s notable that Apple collects 0% of advertising-based mobile app revenues, but substantial portions of payments).
Apple’s policies also govern how apps are designed and operated off iOS, too; although developers can partly “fork” their apps so that there’s an iOS edition and a non-iOS edition, this is technically, financially, and operationally impractical for almost all developers. And the iOS ecosystem is now so popular and lucrative that entire markets and technologies (e.g. cloud gaming, 5G, AR) have only made it once they’re embraced by Apple, which means the company also defines how they’re deployed.
Apple’s regulatory role frequently leads to widespread good. The company’s aforementioned efforts to suppress excessive tracking and data collection is particularly commendable and worth highlighting. And despite (and sometimes because) of its control, there are many vibrant and competitive secondary markets on top of iOS, such as those of streaming video and direct-to-consumer e-commerce.
However, Apple has a vested financial interest in its regulation, and in many cases, it makes decisions that obviously represent those interests over and at the expense of its users and/or the ecosystem at large. Section 3 enumerates these decisions, which include which technologies and standards should exist, the monetization models that should be used and the profits collected, and thus also which businesses are built and not built, and when. Collectively, this power can prevent, or at minimum confine the next generation internet — often to Apple’s benefit.
Chapter Three: The Harms from Apple’s Regulatory Power and Legislations
(It isn’t strictly necessary to read the below dropdowns, but I strongly recommend them. Just click to expand)
↪ A: Apple effectively controls whether specific products/businesses can have an app
Though counterintuitive, the best starter example here is the 2008 app “I Am Rich". This poorly-designed, typo-ridden application cost $999 (Apple’s maximum allowed price) and described itself as a “work of art” whose sole purpose was enabling a purchaser to “show other people they were able to afford it.” A day after its release, Apple terminated the app and refunded all purchases without providing the developer with a rationale. The following year, the developer of “I Am Rich” submitted “I Am Rich LE” to Apple’s App Store. This sequel cost $10, not $999, and included a basic calculator function so that the app had “definable content.” Apple approved this app, which is still available twelve years later.
Although the pricing of these apps is quite different, Apple’s approach suggests a clear approval methodology. In its first iteration, Apple decided it did not have enough value to the user. In its second iteration, “I Am Rich” no longer suffered from this problem due to its price reduction and/or (perfunctory) functional enhancement. And thus it could live. (Apple told the Wall Street Journal its original decision was a “judgement call.”)
“I Am Rich” was a dumb app, but there was nothing illegal about it and it was hardly harmful (and would be less so under iOS’ current refund policies). Apple likely denied the app because it was a “bad look” for its fledgling app ecosystem and thus it was not allowed to exist.
The stakes with “I Am Rich” were low, as were those of the app economy in 2008. But today, with Apple controlling nearly 80% of American App Store revenues and more than two-thirds of app users, the stakes are incredibly high. Apple determines whether an app, business model or company is allowed and/or can monetize.
As a society, we believe that whether something has value and should be bought is best left up to the marketplace (with exceptions for danger, fraud, etc.). For example, Wal-Mart can choose not to stock poor quality products that wastes consumer money — or consumers can choose not to buy it, leading Wal-Mart to stop stocking it. But in either case, the maker of that product can pursue other retailers or sell direct-to-consumer. But when it comes to apps, there’s only one retailer. And the market only decides when it comes to purchasing multi-purpose hardware every two-to-four years, which has no material relationship to a given app (which might not even exist at the time of purchase).
This issue is increasingly important as the virtual world grows. For example, there’s an entire economic category for products that are priced well above their intrinsic value because of their scarcity or brand: Veblen Goods. What if Damien Hirst wanted to make a single edition artwork app for $1,000; would that be allowed? The answer is probably, because of his name. But what about the next would-be Damien Hirst? Or Damien Hirst under pseudonym? And what if Damien priced at $1 million? We don’t know. (And would Apple deserve 30%?)
According to former a16z partner Benedict Evans, some bans “seem to be just performance preference, or taste… whatever it is, there’s no safety, security or privacy issue — Apple just doesn’t like those apps.” It is problematic enough that consumers can only determine an app’s viability by purchasing a $1,000 smartphone. It’s worse still when Apple can’t (or won’t) clearly articulate the rules or rationale behind its approvals — especially when it comes to emerging products or categories that have few precedents.
In some instances, we know exactly what Apple doesn’t like. An example here is pornography. Companies such as YouPorn and OnlyFans are not allowed to have apps, only websites. Certainly, a number of these sites have serious issues relating to their upload capabilities for user-generated content. As such, many are happy Apple deprives these companies of applications. But the issue of uploader consent and legality isn’t the reason for Apple’s ban. Even if a single person were to develop their own app and offer notarized consent for every upload, this app wouldn’t be allowed. “We do believe we have a moral responsibility to keep porn off the iPhone,” Jobs wrote to a user in 2010. Apple’s porn ban is not about the law or safety, but instead protecting Apple’s brand or proselytizing the personal philosophy of its executives. As a result, an entire industry is prohibited from having an application. This issue isn’t limited to adult content: Apple has similar policies on alcohol and drug culture. But the pornography issue is particularly unfortunate as the virtual world should make sex work far safer.
Parler is another interesting example. Although many were pleased with Apple’s decision to ban the app from its App Store, and therefore three quarters of American phones, the fact that Apple could unilaterally (and unstoppably) choose to do so within 24 hours (or less) was alarming. It also doesn’t seem to have been necessary.
After the 2021 storming of the U.S. Capitol, many of Parler’s software vendors, such as AWS, served Parler with a termination notice. Unlike Apple, losing AWS doesn’t kill Parler, as the company could shift to any number of competing providers. However, Parler’s CEO eventually announced that “every vendor from text message services to email providers to our lawyers all ditched us too on the same day,” and that they could not find viable substitutes for any of them. This is a far more healthy market mechanism than “does the company that operates the phone most people have like the product?” It’s also more impactful as the loss of these vendors means not just the end of Parler’s iOS app, but all apps and its websites, too.
These specific examples might seem small, but their implications are enormous, compounding, and particularly consequential when foraging into an unknown future.
↪ B: Apple can and does hold up entire industries and business models
For close to a decade, consumers have had the ability to play video games via cloud stream, rather than local installation and processing, just as they have enjoyed streaming audio or video without first downloading the file.
Today, a substantial portion of audio and video industry revenues (and almost all growth) is in streaming. Not so in video games. And a key issue has been Apple, which prohibited real cloud gaming services from the App Store. “Netflix for Games” apps such as Google Stadia or Microsoft xCloud were nominally approved… but only if they didn’t play games. Google’s iOS Stadia app, for example, was only approved as a demoroom app of sorts; the Stadia app could show the games Stadia included and sold, but even a paying subscriber couldn’t access or buy them via the Stadia app.
Because cloud game streams are video streams, and the Safari browser supports video streaming, cloud gaming is still technically possible on iOS devices even if the App Store prohibits it (though Apple doesn’t allow a service’s app to tell users this). This means running into the aforementioned limitations of Safari-based experiences. UI/Xs are clumsier, crash more frequently, have fewer capabilities and so on. There’s a reason most people use the iOS Netflix app rather than Netflix via the Safari browser. Furthermore, Safari’s prohibition on browser-based notifications means a player can’t see a friend’s play invitation unless they’re in the game. These may seem like surmountable problems, but just imagine trying to sell consumers on “the future of gaming” while also falling short of historical service expectations and table stakes features.
As mentioned earlier, Apple long argued game bundle apps (cloud streamed or not) were banned in order to protect users. This is because Apple would not be able to review/approve all titles and their updates, and thus users could be harmed by inappropriate content, privacy violations or substandard quality. However, this argument was inconsistent with other content categories (e.g. video bundles, which include content Apple doesn’t approve but which could do harm), other game bundles (e.g. Roblox) and its other moderation policies (Apple requires robust, but not perfect, moderation).
Given this, many have argued that Apple’s policies were motivated by the desire to protect its own business. For example, close to three quarters of App Store revenues and even greater share of profits come from games. As a result, the mere existence of high value all-you-can-eat subscriptions threatened to reduce overall consumer spending on the category. What’s more, Apple would find itself disintermediated from developers (who would sell their games to a service like Xbox Game Pass rather than via the App Store) and players (who would subscribe to these services through other channels, such as via their Xbox).
Apple may see the rise of music streaming as a cautionary tale. In 2012, iTunes had a nearly 70% market share in digital music revenues in the U.S. and operated at nearly 30% gross profit margins. Today, Apple Music has less than a third of streaming music share and is believed to operate at negative gross margins. Spotify, the market leader, doesn’t even sell itself through iTunes. Amazon Music Unlimited, which ranks third, is almost exclusively used by Prime customers and thus nets Apple no revenue.
Cloud game streaming might also have a negative impact on Apple’s hardware revenue. One of the key benefits of cloud game streaming is the ability to play a game on a device that can’t run that very game. This is because the device isn’t doing the work. Instead, it’s just receiving a Netflix-like video stream sent out by a remote, highly powerful server. This doesn’t matter for games like Candy Crush, which an LCD-equipped toaster can run, but it means that suddenly, any device with an internet connection and video screen can run the most technically capable games in the world, such as MS Flight Sim. As a result, cloud game streaming reduces the benefit from buying new, let alone top-of-the-line iPhones and iPads. In fact, it means that the original 2007 iPhone could play, via cloud, a “better version” of Fortnite than the 2020 iPhone 12 Pro via local processing. And it wouldn’t take up a single byte of local storage space, either.
Apple’s policy also meant that Apple was the only company that effectively could bundle app-based games on the iPhone, as any app offered in Apple Arcade was de facto approved by the App Store. More on this in 4F.
In the summer of 2020, Apple finally revised its policies so that services such as Google Stadia and Microsoft’s xCloud could exist. But these policies are inarguably byzantine and were widely described as anti-consumer. For example, cloud gaming services (henceforth referred to as “Stadia” for simplicity) would need to first submit every single game (and future update) to the App Store for review, and then maintain a separate listing for the game in the app store. This policy requirement has several implications. First, Apple would effectively control the content release schedules for Stadia. Second, Apple could unilaterally deny any title (which would happen only after Stadia had licensed it, and Stadia would have no direct ability to modify the game in order for it to meet Apple’s requirements). Third, user reviews would be fragmented across Stadia’s app and the App Store. Fourth, Stadia, a game distribution service, would need its developers to form a relationship with the App Store, a competing game distribution service.
Apple’s policies also stated that Stadia subscribers would still not be able to play Stadia games through the Stadia app (which would remain a catalogue). Instead, users would need to download a dedicated Stadia app (which would really just be a thin web client) for every individual game they wanted to play. This would be like downloading a House of Cards Netflix app, and an Orange is the New Black Netflix app, and a Bridgerton Netflix app, with the Netflix app itself serving only as a catalogue/directory for rights management, rather than a game service. This meant that even though Stadia would bill the user for their gaming subscription, curate the content inside that subscription, and power it’s delivery, Apple would distribute the cloud game (i.e. via the App Store) and iOS customers would access the title through the iOS homescreen (not the Stadia app). This policy also results in inevitable consumer confusion. If a game was offered by multiple services, for example, the App Store would end up with multiple listings (i.e. there would be Cyberpunk 2077 — Stadia, Cyberpunk 2077 — Xbox, Cyberpunk 2077 — PlayStation Now, etc.) And every time a service removed a title from their service (e.g. Stadia removed Cyberpunk 2077), users would be left with an empty app on their device.
Apple also declared that all game streaming services would also need to be sold through the App Store. This contrasts with other media bundles, such as those Netflix and Spotify, which have their apps distributed by the app store but can (and choose) not to offer iTunes billing. Finally, Apple said that every subscription-based game must also be made available a la carte purchase through the App Store. This, again, differs from its policies with music, video, audio and literature. Netflix, for example, does not need to (and doesn’t) make Stranger Things available on iTunes for purchase or rental.
Thus far, none of the major cloud gaming services have taken advantage of Apple’s revised policies. And with good reason. To quote The Verge, “Apple is effectively saying these companies cannot build a “Netflix of games” [via apps] on the iPhone. That holy grail of cloud gaming is outlawed. They have to squeeze the entire idea, one game at a time, into an App Store-shaped hole.” Microsoft felt similarly, stating “This remains a bad experience for customers. Gamers want to jump directly into a game from their curated catalog within one app just like they do with movies or songs, and not be forced to download over 100 apps to play individual games [that stream] from the cloud.” Again, it’s understandable that Google, Amazon, Facebook, and Microsoft have decided against contorting their future-facing gaming services into unfriendly and antiquated delivery models.
↪ C: Apple can, does, and will wipe out existing businesses and technologies with little-to no-notice
In 2010, Apple announced that iOS would never support flash. Specifically, Jobs argued that Flash was a clumsy, insecure software platform and that a better experience was available via apps. Jobs wasn’t wrong in his critique (and he had argued as much for years). Flash performed poorly and had been promising better compatibility with mobile devices for years, as well as various security improvements. Nearly every game was better off Flash, as were users and developers.
At the same time, Apple’s decision is seen to have unilaterally destroyed Flash. And this is notable for several reasons. First, Apple had far less power in 2010 than it does 2021, and it was still able to defeat a widely deployed standard. Second, Apple achieved this without destroying current businesses (i.e. extant Flash games and reach), but foreclosing a critical growth opportunity and in doing so, signaling to developers that they needed to move on. Third, Apple’s decision was only a death sentence because of Apple’s control over third party browsers, as this meant Chrome and Firefox couldn’t offer Flash support on iOS even if they wanted to. Fourth, Apple’s move helped push Flash developers to apps, which in turn meant these developers would need to form a B2B relationship with Apple, be intermediated from the customer by the App Store, and hand over a portion of player spending, too. Fifth, Apple explicitly told developers that if they wanted to continue making web-based games, they should use Apple’s WebKit standard.
A decade later, a similar and yet very different event occurred. In August 2020, Epic Games updated its Fortnite app such that users could use Epic’s in-app payment solution, rather than the App Store’s. Within hours, Apple removed the Fortnite app for the App Store, claiming policy violations, and Epic Games in turn sued Apple arguing monopoly power (among other issues).
Not long after, Apple announced it would suspend Epic’s access to iOS’s development tools and, as a result, Epic would no longer be able to update the Unreal Engine. This put thousands of active iOS developers (i.e. businesses) in immediate jeopardy — even impacting those whose games were licensed by Apple for its Arcade subscription service. At any moment, a new bug could emerge — potentially one resulting from a standard iOS update — and Epic would be unable to patch it. And as long as Epic’s license was suspended, the iOS implementation of Unreal would never be able to improve or be updated for compatibility with newer iOS devices. This meant that every Unreal-based developer had to make a choice: wait and hope that Apple and/or Epic would settle before their game broke, or begin an entirely unnecessary, time consuming, costly, and difficult process of rebuilding their games in another engine. And, in stark contrast to Flash, Apple had long touted Unreal and Unreal-based games to developers and players when unveiling its newest devices.
Ultimately, the courts blocked Apple from suspending Epic’s developer access and effectively terminating all Unreal based games. But this block stemmed not from the court’s desire to protect Unreal developers, but from finding Apple’s move to be retaliatory and unrelated to Epic’s Unreal business.
This situation can be read in two ways. One, that Epic is uniquely confrontational and that Apple reacted in kind, albeit perhaps to an extreme. Or two, that the reason why there are not many more companies taking action to address Apple's uniquely constraining business terms is that its available mechanisms for retaliation are so potentially lethal. The second interpretation seems likely when you consider that Epic is one of the only companies who could even attempt to push back.
There are scores of companies that believe Apple holds illegal monopoly power, but they (rightfully) fear retaliation and thus remain silent. Independent developers are particularly scared to speak out against Apple’s policies (even when Apple offers no explanation for rejecting their apps), but even Apple’s most powerful and public critics are cautious. Spotify and Match Group, for example, have long complained about Apple’s abuses, but they’ve filed little litigation (all of which is in the European Union) and instead spent years prodding regulators. This is because the majority of their revenues come from iOS users and a business line or app. As a result, their businesses and brand would be crippled by a delisting, and they'd have no recourse other than months of uncertain litigation.
In this regard, the Epic v. Apple case isn’t unique because of its circumstances, but instead because Epic is one of the few powerful companies that can sue Apple. Unlike the major FAANG giants, iOS doesn’t drive the majority of Epic’s revenues. Unlike Spotify or independent developers, Epic is highly profitable. Unlike Microsoft, Fortnite is for leisure not critical productivity work and can thus survive being delisted without harming third parties and partners. And unlike almost all other large companies, Epic is private and majority controlled by one person. A world in which only companies like Epic can sue or even critique Apple is not healthy.
This problem stems from Apple’s unilateral control over all iOS apps. The company is under no obligation to stock apps it dislikes in its own store. But if it doesn’t, then the app can’t exist. And while Apple benefits from maintaining developer trust, it has also shown that it will break this trust — even with its own partners — to protect its overall strength and weaken its competitors. And there’s no recourse for those that suffer collateral damage.
↪ D: Apple’s policies result in higher consumer prices and/or lower developer profits
Over its first few years, both Steve Jobs and Apple downplayed the profitability and commercial purpose of the App Store. Two months after the store’s July 2008 launch, Jobs reviewed the nascent business with the Wall Street Journal. In its report, the paper stated that “Apple wasn’t likely to derive much in the way of a direct profit from the business… Jobs is betting applications will sell more iPhones and wireless-enabled iPod touch devices, enhancing the appeal of the products in the same way music sold through Apple's iTunes has made iPods more desirable.” To this end, Jobs told the Journal that Apple’s 30% fees were intended to cover credit-card fees and other operating expenses for the store. He also said "[The App Store] is going to crest a half a billion, soon… Who knows, maybe it will be a $1 billion marketplace at some point in time."
The App Store passed this $1B mark in its second year, with Apple noting the store now operated “a bit over break-even”. A decade later, the store had gross sales of $73B and direct profits that would place it within the Fortune 15.
While Apple incurs some marginal costs from additional App Store sales (e.g. credit card fees, bandwidth, expanding its app review team), these costs are fractional. As a result, App Store profits will continue to grow rapidly. Mobile time spent is still increasing, corporations are moving more of their businesses to mobile applications. In addition, Apple continues to launch new endpoint devices (e.g. Watches, TV pucks and, soon, AR glasses), and it’s difficult to imagine users decamp from iOS en masse for competing smartphones (See Section 4B). And last, but not least, the App Store’s profits are protected by the fact it faces no competition in in-app billing.
Most transactions today occur via credit card, in which case the credit card provider (e.g. American Express, MasterCard, Visa) charges the merchant 1.5 to 4% of the purchase price (the upper range stems from a minimum fee applied a low purchase price, such as a $0.05 fee on a $1.15 cup of tea). These fees are kept low due to intense competition. If American Express charged 20%, but Visa only 2%, a coffee shop wouldn’t accept AmEx. In addition, these credit card companies contend with merchants who can just require cash, which has no processing fee. In this sense, they must “earn” their cut through convenience, revenue reporting, and the ability to increase sales through faster transactions and more valuable customers. AmEx, as an example, typically charges the highest fees of any credit card specifically because its customers are wealthier and thus typically spend more, too.
Compared to credit card transactions, Apple’s fees look incredibly high. In addition, they’re likely several times its cost base. Epic Games, which operates the second largest PC game store, says it’s marginal costs from a game sale range from 5 to 7% of revenue. Given mobile app downloads are significantly smaller than games, and Apple is likely to have the world’s best terms with the credit card companies, it’s likely Apple’s marginal costs are even lower. Google Play is similarly instructive. Although the Play Store takes a similar share of in-app payments to the App Store, Google shares the majority of its commission with wireless carriers and device manufacturers. As Google says Play is a profitable business, this means it too is collecting several times its cost basis.
The problem here isn’t that 30% is “wrong” and another specific number is “fair.” Instead, it’s that there’s no direct competition for these rates. If a developer wants to have an iOS app, they must use the App Store, and if they use the App Store, they must accept Apple’s rates (which Apple set with the explicit purpose of not “leaving money on the table”). As a result, Apple’s rates face competition only at the hardware level and/or in the rare case a developer believes it can exist without an iOS app.
We also have some evidence on how competition would lower these non-market rates. Only days before the official launch of Epic Games Store, which took just 12% (7% net if a game already used Epic’s Unreal Engine) Valve’s Steam announced it would roll back a more than 15-year-old policy of charging a flat 30% on all game sales. Instead, it would take only 25% after $10 million in lifetime revenue and 20% after $50 million. The platform even announced it would backdate eligibility by two months.
If Apple’s fees are above competitive rates, there are necessary consequences: consumers prices are inflated and/or developer profits are depressed. If competition forced Apple to a 10% rate, for example, a developer could offer consumers 20% lower prices and still have the same profits (in all likelihood, profits would go up as lower prices would drive more sales). Alternatively, a developer could maintain current pricing and enjoy a 28 percentage point increase in net revenues (and likely a greater increase in net profits).
This is a critical point because it’s easy to look at the enormity of Apple’s payouts to developers and size of the ecosystem and consider it a success. And it is an enormous success. It’s also clear that enormous businesses have been built despite these fees (and companies such as Epic Games and Match Group don’t need a 15% reduction in cost of goods sold). But we don’t know how many more businesses would exist at a lower rate, or how many of the businesses that do exist would have better products and more employees under different terms. But we do have some idea.
Almost no IRL companies have 30% operating margins. This means that the vast majority of these businesses would not be able to exist under Apple’s fees, or they’d have to push the store fees to consumers (which, of course, few businesses can do anyway). The scope of this problem is enormous. More than half of U.S. GDP and over 60% of employment resides within small- to medium-sized businesses, where the median net income margin is 8 to 12%. Even a partial fee reduction can be transformative.
We saw this with Apple’s recently announced “App Store Small Business Program”, in which the App Store drops its commission from 30% to 15% for companies grossing less than $ million per year. An extra $150,000 is a very meaningful sum - one that can pay for a good mid-level software developer job (including all benefits). And certainly, it seems economically healthier that an additional $150,000 goes to an SMB for their work rather than Apple’s historic cash reserves. But this perk is tiny in its scope, highly encumbered, and doesn’t substantially affect an SMB’s ability to grow or reduce its need for outside capital.
To point, 98% of SMBs generate more than $1 million per year (after all, $1 million in pre-tax revenue can’t support many employees). And for Apple, less than 5% of App Store revenues came from developers making less than $1 million. This means Apple is giving up only 15% of 5% of App Store revenues and aiding an even smaller number of developers.
In addition, a business that exceeds this $1-million threshold will return to a 30% commission in the following year — even for their first million in sales. This has a number of important consequences. For example, Apple counts all of a developer’s apps against that million. Therefore a developer who pays 15% on a $900,000 app would make less money the following year if they decided to launch a second app with ‘only’ $200,000 in revenues. Similarly, a developer whose app crosses the $1-million line would need to grow revenues by more than 15% in order to offset the end of Apple’s App Store Small Business Program. This is neither a generous system, nor one businesses can safely plan around.
Finally, it should be noted that Apple’s high fees are partly responsible for the state of today’s websites. Even companies with successful subscription media businesses, such as the New York Times, load their websites with ads, pop-ups, trackers, and cookies — even for paid subscribers — because they need the extra margin, not because they want to. And, of course, Apple doesn’t take even a single basis point of ad revenues.
↪ E: Apple unilaterally controls monetization of applications, and the results are inconsistent and problematic
The issue of Apple’s in-app purchasing is much broader than whether 30% is a fair market rate or not. Apple actually has a dozen different rates and policy exceptions depending on the specific business, their size, and influence. As a basic summary:
iOS takes a 30% cut of in-app transactions for in-app content/experiences, such as an extra life, extra level, TV episode or song, package of Tinder swipes, etc.
iOS takes a 30% share of a monthly subscription fee until a subscriber reaches the 13th consecutive month, in which case only 15% applies. However, select video services pay 15% from day one as long as they also offer Apple TV apps. In addition, select video services (e.g. Amazon and Altice) that are individually approved by Apple pay only 5% on in-app transactions related to their core business (e.g. a movie rental)
All digital subscription apps can offer their subscriptions direct-to-consumer outside the app store, but they cannot tell consumers this inside their iOS apps. In addition, “Read Only” subscription services, such as Spotify or Netflix, can choose to only offer subscriptions outside the app store, but they still cannot tell consumers that inside their iOS apps.
When the iPhone is used to buy physical goods (e.g. shoes on Amazon) or load a gift card used for physical goods (a Starbucks gift card) Apple takes 0%
These variances seem arbitrary, but are based on Apple’s self-perceived leverage circa-2010 (again, they were designed to prevent the company from “leaving money on the table”). As a result, they have unequal and inconsistent implications, produce consumer-unfriendly results, and generate perverse incentives.
As an example, video and audio-based subscriptions do not need to pay Apple’s 15-30% commission. And many of the leading services, such as Netflix and Spotify, don’t. Meanwhile, video gaming subscriptions don’t receive this exception and incur an arbitrary net revenue ceiling. Apple differentiates between video gaming subscriptions and video/audio ones on the basis that the latter are “read only” (i.e. interactive). However, Netflix is now offering a suite of interactive stories, such as Bandersnatch, while Spotify is prepping its own interactive podcasts. Most video services are now offering watch parties, interactive chats and polls now, too. “Read only” is not just a false description, it’s an undefinable one.
To a similar end, no one quite knows how/why Roblox is compliant with Apple’s App Store policies. The platform operates 40 million games (more than 10x iOS’s total number of apps), none of which have been approved by Apple or have separate app store listings. In addition, Roblox’s highest value Robux subscription is only available online, even though it can be used on the iOS version of Roblox. And if Roblox is in-policy, it’s telling that the reason why is unclear (the most common explanation is that Apple overlooked Roblox for too long and is now too popular to deny and therefore had a unique and unreplicable grandfather clause of sorts).
Although Apple claims that app “pricing is up to [the developer]” and "Apple cannot make specific recommendations", the company is nevertheless directly involved in setting the price of iOS apps, too. This happens in a few ways. First, the App Store team also routinely rejects apps that they consider to have “irrationally high [prices]" or which are “clear rip-offs”. While Apple’s goals here are admirable, the execution is inconsistent, non-specific, and untestable.
For example, one anonymous developer had their app rejected in February of 2021 on the grounds that $7.99 week was too high. Apple did not disapprove of the app, and did support it being a paid app, and one sold on a weekly basis. The issue was that an app store reviewer considered it to be worth perhaps $3.99 or $4.99 or $5.99 or $6.99 per week and no more. Exactly which price was acceptable, or why, is entirely unclear. As is how a developer might convince Apple that $7.99 was “rational” (comps? target margin? return on investment? operating costs?). Regardless, the developer could not let the market decide what’s rational, nor could they appeal to any market data at all (e.g. retention rate, usage), since Apple wouldn’t permit the app for sale in the first place. (Of course, there’s also an absurdity in Apple’s philosophy here given that it also sells $700 wheels for the MacBook Pro and a $1,000 stand for its $5,000 computer).
In addition, Apple requires most developer to pick a global pricing tier for their apps. This means that when a developer sets a price based on their core market (e.g. the US), Apple determines its pricing in all other ones. The consequences here are profound. For example, a developer with a market leading app in Italy must price their German app as though they lead the market there, as well, even if they’re a new entrant or laggard (or alternatively, they’d have to drop their German price). Furthermore, it makes little sense to force the same pricing rubric to all app categories in all markets. The pricing/market dynamics of mobile games and productivity software aren’t the same, least of all as applies to India v. Mexico v. the US v. Norway, etc.
Lastly, it’s important to highlight how Apple’s policies create whipsaw economics and payments. A game that’s ad based, for example, owes no fee to Apple. A transaction-based game pays 30%. A game with a monthly subscription pays 30% until the first year ends, in which case it becomes 15%. A toys-to-life game, though, can pay nothing even if it involves significant in-game virtual goods. This is not an effective model for the new digital economy, especially when it’s devoid of direct market forces.
↪ F: Apple’s policies frequently benefits its own services and harms those of its competitors
Apple’s control of app approval policies, distribution, and payments also strongly advantage its own apps to the detriment of those they compete with (i.e. independent developers on iOS).
As mentioned above, Apple Arcade operated for more than a year before Apple even allowed other companies to bundle third party games. And while Apple now permits these bundles, the company also imposes laborious compliance rules that make operating such a service cumbersome and allows Apple’s Arcade service to move quickly (it also means Apple Arcade is the only bundle that can have games only available by subscription). What’s more, there’s no digital media subscription bundle in the marketplace today (mobile or otherwise) which operates at 15 to 30% profits margins. As a result, these services cannot afford to pay Apple 15 to 30% per subscribers fee per month. This cost doesn’t really matter for Apple’s own media subscription bundles, such as Apple Arcade or Apple Music. They might lose money on a direct basis, but they’d be large revenue drivers for Apple’s App Store or multi-subscription bundle, Apple One.
The inability to afford Apple’s store fees means that Spotify and Netflix do not support in-app billing (again, Apple does not allow gaming bundles to skip in-app payments). This means Spotify and Netflix must acquire customers on their websites (and their apps aren’t allowed to tell users this). This isn’t life-ending, but it means that it’s far easier for a consumer to subscribe to Apple Music than to Spotify, and that some would-be Spotify subscribers won’t even know why they can’t subscribe in app or where to do so. Apple Music also benefits from being preloaded on iOS devices, and it’s participation in the (deeply discounted) Apple One bundle.
Collectively, this is a powerful advantage - especially in a category broadly where players have limited content-based differentiation. By 2019, Apple Music had reportedly passed Spotify in the United States, even though it launched four years after the Swedish service. Pundits have long over-hyped the threat of a late-entering FAANG giant destroying an independent market leader (e.g. Facebook Dating v. Match Group). However, this threat is far more serious when that company has a 15 to 30% COGS advantage and doesn’t need to spend money on installation.
We can see another example of the preload in mapping. When Apple launched Apple Maps in 2012, it was far worse than Google Maps. And it should have been! At the time, more than 7,000 Google employees worked on Google’s mapping tech and apps, while Apple had only 14,000 non-retail employees total. And yet three years later, Apple boasted that Apple Maps was opened 3.5 times more frequently than Google on iOS.
Mobile payments offers another great, and incredibly lucrative example. Apple does not allow third party payment providers to access the NFC chipsets on the iPhone and Apple Watch. As a result, Apple Pay is the only payment solution that can use the tap-and-pay feature on iOS devices. Apple claims this is to protect iOS users. However, the company also allows hotel and automotive companies to unlock doors via NFC. Notably, Apple doesn’t operate in the hotel or automotive industry. It does, however, take an estimated 0.15% of every Apple Pay transaction. It’s also notable that during the COVID-19 pandemic, Apple’s policy means that if users don’t switch to Apple Pay, they need to use physical credit cards or cash in order to pay. That’s hardly safer.
Finally, we can return to the earlier example of web browsers. Because all iOS browsers must use the Safari WebKit engine, they can’t differentiate in speed, standards support, or capability. Instead, browsers such as Chrome only offer users features such as ability to synchronize tabs, bookmarks and search history with non-iOS devices. This is a nice, but fairly marginal advantage - especially given third-party browsers used outdated versions of WebKit and are thus technically inferior to iOS Safari. And for those whom a smartphone is their primary computing device, synchronization with a secondary device is not a particularly compelling value proposition. Note, too, that Apple doesn’t allow extensions on third-party browsers. That means the experience and capabilities a Chrome user receives on a PC or Mac can’t be replicated on mobile. What’s more, iOS Safari does have selective extension support, including ad blocking.
This year, we’re likely to see yet another case study: by some reports, Apple is planning to launch its own natively integrated search engine to compete with Google.
Chapter Four: Apple’s Deficient Monopoly Defenses
Apple typically defends itself from monopoly allegations and its supposed harms using one of five arguments. But given its unprecedented power and controlling policies, our standards for these claims must be high. Apple should either be able to show that its rules are consistently and primarily aimed at the best outcome for users and developers or that users and developers can practically escape those rules. They are unable to do either.
A: Developers and Users Can Always Leverage the “Open Web”
Apple correctly argues that developers do not need to make an app to reach iPhone users. Instead, they can create websites accessible via the iPhone’s Apple Safari browser, or one developed by a third party, such as Google Chrome. In either of these latter two cases, Apple does not review, approve, or deny the content of these websites, nor does it require the use of the App Store for payment. However, this argument is misleading.
Websites are at a deep disadvantage on the iPhone. Because apps use native device drivers, they typically run much better and more efficiently than web pages and web apps, which uses “heavier” code that isn’t optimized for the user’s specific device and instead require a “translator” to leverage the device’s capabilities.
In addition, the iPhone UX is intentionally designed around apps and not websites. Note, for example, how much easier it is to navigate, manage and sort apps than browser tabs (and even if you bookmark a site or progressive web app to the homescreen, they open into your browser and then get lost or duplicated in tabs). Clearing your web history, cache or cookies means logging out of all your browser experiences and deleting login credentials, but not those of your apps. Crucially, this is a design choice. Other OSes, most notably Palm’s WebOS, were conceived to facilitate Web apps.
These two reasons - one technical, the other experiential - explain why users download Netflix apps versus access the service via browser, and why Apple tells developers that their businesses will be more successful via app.
There is no open web on the iPhone, either — only the “iPhone web.” Five years after the launch of the iPhone, Apple revised its App Store policy to allow for third party browsers such as Google’s Chrome and Mozilla Firefox. But this was only a surface level compromise: Apple does not truly allow for alternative browsers. To quote Apple expert John Gruber, the iOS version of Chrome “does not use the Chrome rendering or JavaScript engines - the App Store rules forbid that. It's the iOS system version of [Safari] WebKit wrapped in Google's own browser UI.” In other words, Chrome on iOS is simply a variant of iOS Safari that syncs to a user’s Google/non-iOS Chrome accounts and usage. And notably, Apple forces third party browsers to use older, and thus slower and less capable, versions of WebKit than iOS Safari.
This approach also means Apple’s technical decisions for Safari affect the open web for iOS users. For example, Safari doesn’t support much of WebGL, a JavaScript API that allows for complex browser-based 2D and 3D rendering via local processing and without plug-ins. Safari also doesn’t allow websites or progressive web apps to perform background data sync, access the camera (thus no face-based logins, AR-experiences, light sensor usage, etc.), access many BlueTooth devices and functions, pay with NFC, etc. And because Safari opts against these capabilities for web-based experiences, they’re strictly off limits to iPhone developers and owners, no matter the browser
Many of these policies decisions are intended to protect users. Allowing browsers unlimited access to device drivers and folders poses a security risk, for example. However many seem specifically designed to protect Apple’s App Store and billing - especially for games, which drive 75% of App Store revenues.
WebGL, for example, might not run as well as device-specific code, but today’s hyper-powerful iPhones are capable of running a large number of WebGL games without crashing or disappointing the user. Battery is still a challenge, but that’s true when playing Call of Duty Mobile or PUBG anyway (both of which Apple frequently promotes in its App Store). Remember, too, that Apple doesn’t require developers to use the “best” tech for its games, nor the most efficient code. So it’s just a policy being enforced in this context. And even when a user manually saves PWA to their home screen, these web apps are prohibited from sending push notifications or performing background sync. This doesn’t “protect” the user, but it does effectively prohibit a game that is deeply reliant on friend notifications. Plainly put, Apple’s blocking of rich WebGL of stymying of cloud game streaming has ensured that its App Store is the only way a developer can distribute a premium game on iOS, and the only way an iOS user can access one.
The rejection of web-based NFC, meanwhile, helps drive mobile payments through apps distributed by the App Store, and its security could be bolstered through any number of secondary requirements (e.g. thumbprint verification or FaceID).
Ultimately, Apple cannot reasonably argue developers and users can freely leverage the open web when Apple distributes the browsers used to access it, determines which standards and capabilities these browsers afford, and how a web app can interact with the user. Especially given the ways in which it unnecessarily limits the “open web”.
Let’s apply Steve Jobs’ own definition of openness:
“Adobe claims… that Flash is open, but in fact the opposite is true. Let me explain. Adobe’s Flash products are 100% proprietary. They are only available from Adobe, and Adobe has sole authority as to their future enhancement, pricing, etc. While Adobe’s Flash products are widely available, this does not mean they are open, since they are controlled entirely by Adobe and available only from Adobe. By almost any definition, Flash is a closed system.”
By almost any definition, but especially Jobs’, the iPhone web is a closed system. As a result, the vast majority of the US mobile internet is “closed” and subject to Apple’s decisions. In truth, this feels less like a defence (Section 4) and more like an additional harm (Section 3).
B: Consumers Can Buy Other Phones and/or Developers Can Flee To Other Devices
For the foreseeable future, iOS will be the dominant access pathway, passport, monetizer and platform for not just digital life, but virtual ones. Apple holds this role because it makes best-in-class hardware, offers the best apps, and operates the most lucrative app store. This is a reflection of its success in the competitive mobile phone marketplace for close to 15 years and under a broadly consistent playbook.
And while consumers can buy other phones, it’s incredibly difficult to imagine a substantial re-platforming. The number two operating system, Android, is supported by nearly every other smartphone manufacturer, Google previously bought one of the leading OEMs (Motorola) in order to better compete with Apple, and continues to make its own proprietary “iPhone killers” to little success. There is no real third competitor, and more broadly, it’s not clear what could convince an iPhone user to leave for Android or this mythical runner-up to the runner-up.
Part of the problem here is how, exactly, a competing smartphone can successfully differentiate from the iPhone given its lucrative customer base. For example, Android or a third mobile OS could try to attract mobile developers through better policies and permissions. But there are almost no companies on earth that could leave iOS outright as this would mean leaving behind two thirds of their users and 75% of revenues. Even Google typically prioritizes the iOS builds/releases of it’s apps, rather than those of Android, because it fears losing users of its iOS-based apps. As a result, it’s impossible to imagine developer decampment forcing iOS to change its policies or leading enough iPhone users to switch platforms (which requires hundreds of dollars of spend and years).
Alternatively, a competing OS or device manufacturer could seek out technical differentiation. However, almost all innovation today comes not from the hardware itself, but how it’s used to produce differentiated software and services. It doesn’t matter what a phone can do, but how these capabilities are used by developers. In this case, we run back into the iOS-first developer problem: Apple’s dominance is such that new technology (e.g. 5G or AR) isn’t really a “thing” until it comes to Apple devices.
One indication of Apple's control over developers is the fact they stay despite their many complaints. Returning to Benedict Evans, he notes that Apple’s inconsistent App Store policies/approvals have “done real damage to Apple’s brand amongst developers.” But this obviously has not led to boycotts or resistance; it’s not clear they would work, at least on any timeline that a major developer could endure. Contrast this with the example of Facebook. The social giant's history of changing API and monetization policies in the late 2000s/early 2010s left it without a developer ecosystem.
The iOS ecosystem is also becoming stronger, more tightly controlled, and competitively exclusionary. The Mac, which has historically been an open platform, is now locking down by pushing users to the Mac App Store and standardizing its chipsets and policies with those of iOS. In the coming decade, many analysts and technologists believe the iPhones will take on the role of “edge servers” for local computation. This would mean even more of the world around us will run off, be powered by and governed through the iPhone (e.g. our glasses, television, bikes). This also reduces the likelihood of iOS being supplanted by a “new OS” (as iOS displaced Windows).
In addition, Apple continues to launch more Apple-only software and services (e.g. Family iCloud Sync, Apple Fitness, iOS Apps for Mac) which build both service-level and family-based lock-in. Over the past five years, the number of Apple devices owned by an iPhone user has grown from 1.45 to 1.7. And since 2020, Apple has required every iOS app that uses cross-platform accounts log-ins such as Facebook, Twitter, or Google, to use Apple IDs. Users don’t need to use their Apple IDs, of course, but many will - which means as a result, app makers such as the New York Times will need to support Apple ID on all devices and endpoints, including the PC web and Android. And it’s much harder to leave a hardware and services ecosystem when one of these services (i.e. identity) is your passport to the web.
C: The iPhone is Apple’s Phone
This is the first time consumer rights and laws enter the discussion, and where the answers become less clear and more path dependent.
A purchased iPhone is the physical and personal property of the purchaser. In most cases, this gives the purchaser the right to use this product in any way, shape, or form they please (as long as laws aren’t broken). The same is technically true with the iPhone. However, the operating system that runs it remains Apple’s and is needed to run the phone. The laws here are unclear at best, but are more commonly understood to favor Apple’s nearly-unlimited controls. Accordingly, it’s helpful to look to analogy and consider what’s societally desirable, acceptable, and tolerable.
The Ford Motor Company could mandate which types of tires are used on the F150. It could also take a cut of the tires sold for all F150s, place controls around what sorts of roads an F150 drive down and the speed that could be used, while also requiring all car-related purchases (e.g. gas or drive through coffee and food) use a Ford payment service. No one would accept these limitations today, obviously, but it’s only recently technologically possible. Had this model been available in the early 20th century, it’s doubtlessly true that Ford (and later, others) would have tried it. And this would certainly have increased consumer prices, even if this integration led to a better running car and prevented users from reckless behaviour.
But while the technology required to create such a bundle has emerged over the past 20 years, a number of laws designed to curb these sorts of controls have emerged. For example, the Motor Vehicle Owners' Right to Repair Act, which is now used in all 50 states, required automotive manufacturers to provide the same information to independent repair shops as they do for dealer shops. This was particularly important as onboard computer systems began to track more complex performance and diagnostic data. Many of these acts also prohibited automakers from invalidating car warranties if an independent dealer was used.
In this regard, we should ask what rights the consumer should have, rather than which rights Apple might prefer and be owed under current laws.
D: Apple Can’t Have a Monopoly On Its Own Product
Apple’s strongest defence is that of precedent. Decades of case laws suggests (1) a product cannot itself be a market (i.e. trucks are a market, not Ford F150s); and (2) a company cannot have a monopoly over its own product.
It’s irrelevant that Apple has 100% market share of iPhones; the iPhone isn’t a market but instead a product in the mobile device category. Here, Apple’s share is a still-impressive 66% in the US (25% globally). Apple correctly notes there are many other competing devices, some of which are far cheaper and/or have superior specifications. What’s more, smartphones are typically replaced every 2-4 years, which means Apple must consistently win these customers (unlike, say, railroads, telephone poles, or electric grids). It is also difficult to prove that Apple forces current iPhone customers to remain iPhone customers.
But this is where analysis gets tricky. It’s wrong to call the iPhone a product. The iPhone is a platform. It’s also a bundle of hardware + an operating system + distribution system + payment solution + services.
To use a simple analogy, the Phone is not a product, like a Ford F150, or even a market, such as the truck market. Instead, the iPhone (or more appropriately, iOS) operates a lot more like the US Interstate Highway System. The IHS is not a dominant share of highways (30%) and especially not all roads (1.7%), but it dominates the largest and most important roads and is the backbone of US commerce and trade. And in this specific case, it would be as though the IHS also had its own proprietary cars and on-platform credit card program, license plate system and passport, owned and rented all the land along the highway to private businesses (which, from time to time, it decided to compete with), and operated its own police. And because of its importance and popularity, the iOS Interstate’s technical decisions also informed the construction of all competing highways and roads, as well as the products, business models, and architecture of all road-based businesses (e.g. gas stations, drive throughs, etc.). The analogy obviously isn’t perfect, but it sounds more like government than a market, let alone a product.
Apple is hardly the only company that insists on total control over its user experience; the above descriptor fits Disneyland just as well (if not better). What’s more, Apple’s tight, end-to-end control is part of why consumers buy Apple’s devices.
However, we (i.e. society) recognize the importance of interrogating how economies operate, what they prioritize, and how they benefit consumers, sustain competition, and drive innovation. It is clear that the digital/virtual economy is only going to grow in importance. As a result, we need to watch for default outcomes and drive to the ones we aspire towards. iOS is far more consequential than Disneyland and Apple intends for it to be ‘the’ platform of the digital economy.
E: Apple’s Policies Are Designed to Protect Its Users
Apple argues its control over app distribution, the policies of the app store, and payments are required to protect user data, avoid viruses, preserve safety of child measures and services like screentime, and ensure their devices run properly. There’s some truth here. Curation, store guidelines, and installation controls do help limit malware, spyware, and other shady practices - fewer apps are downloaded, fewer permissions are allowed, disclosures are increased, and developers face enormous downsides (i.e. bans) if they trick Apple or users. And Apple’s control of app distribution also enables it to control web browsers; there may not be an open web on the iPhone, but Apple’s policies throttle malcontents.
Security is important because of how important our devices are to day-to-day life and what we store on them, but Apple's arguments include ample exaggeration. For example, corporations are allowed to operate their own private app stores on iOS devices and manage all app submissions, updates, and approvals without Apple. There is no evidence that this destroys the user experience or puts their privacy at risk (and notably, this corporate data should be far more valuable and thus sought out). Apple also allows developers to distribute beta versions of their apps via TestFlight, rather than the App Store. TestFlight reaches only a fraction of iOS users, most of whom are sophisticated users, but there’s no evidence of harm here either.
Apple has also used the argument of security when denying apps such as Microsoft’s Xbox Game Pass, a service that bundles 100+ Xbox games into a single Netflix-like service. Specifically, Apple argued that these bundles meant Apple could not individually review each game (and update) to ensure its content, quality, and data practices are above board. This is a flimsy rationale. As specific examples, the games bundled by Microsoft Xbox and Sony PlayStation are unlikely to include inappropriate content or contain secret, data harvesting code. And if they did, Apple could then remove the services. After all, Apple doesn’t mandate flawless moderation, just robust efforts. Apple has also given policy exemptions to a number of major telecoms and technology companies.
This safety argument also ignores the fact that many of the App Store’s approved games, such as Minecraft, already contain inappropriate UGC content (e.g. phallus-shaped buildings with fountains on the top) and suffer from targeted harassment via audio chat. In addition, Apple does not apply this “individual approval” policy to other content bundles. Netflix, for example, need not submit all of its titles for approval, nor does the Fox News App (or Roblox, which is effectively a game bundle). There’s also no evidence that security explains Safari’s refusal of rich WebGL-based gaming.
Lastly, it’s notable that while MacOS does not have iOS’s restrictions on software/app installations, it remains secure and safe to use. This is because the majority of security is held at the kernel/OS-level. To this end, malicious apps (and updates) have lengthy history of making it through the App Store review. This is inevitable given the volume of submissions, time spent reviewing code, and human error rates, but crucially, these bad actor apps don’t destroy the device or devour private files (as is often the case with Windows malware). This is because of iOS’s system and API-level security, which is a stronger and more scalable solution - and one that offers consumers more choice and developers more capabilities.
Chapter Five: The Importance of Prioritizing Overall Prosperity
Why does this matter? How much should we really care about how apps are made and who gets paid for them? Or the platforms that are used or allowed? Or who can advantage which identity solution, service, or standard? The answer is that we have to care. In 2021, almost every company is a digital company, and the scope, significance, and complexity of the digital/virtual world continues to expand.
To prosper, new winners must be able to emerge at every layer of the “new economy” and without being constrained by what a single company envisions or wants. The current situation isn’t terrible — great things are happening in terms of users, creators, and innovation. But the very decisions that have made Apple so successful today also limit the number of creators that participate in the virtual economy, handicap the creativity of their products and business models, and make every transaction more expensive. Most important, it is blocking the organic evolution of the overall Internet.
Apple’s store fees are a clear example. We need businesses to be rewarded for the creation of value and able to invest accordingly. However, average profit margins in the US are between 10-15%. Apple’s 15-30% cut of revenue therefore means Apple collects more profit from the creation of a new digital business or digital sale than those who invested and took risk to build it. As Apple is the most valuable company in history, generates more operating cash flow than any company in history, and operates the most successful product in history, Apple’s uncontested toll system is probably not an ideal outcome.
Apple’s fees are particularly steep for businesses that focus on virtual experiences — a group which is quickly growing in strategic importance and commercial value. As analyst Ben Thompson notes, Apple’s “current organizing principle is digital versus analog; anything that is digital has to have in-app purchase, while anything that is analog — i.e. connected to the real world — can monetize however it pleases. That is why Amazon or Uber can ask for your credit card, and Airbnb can do the same for rooms but not for digital experiences (according to Apple).” This is a terrible system that conflates iOS’s control of the smartphone audience with its role creating all smartphone-based markets, and penalizes the very businesses that are most in need of investment: new ones.
If Nike decides to build a ‘virtual sneaker’ store in 2021, is iOS responsible for this opportunity? Why should Apple take 30% of these transactions but 0% if Nike builds a new physical sneaker brand that can only be purchased via app? Apple’s role (and investment) here is indistinguishable, but its take is (literally) infinitely higher and non-negotiable. And while Nike might have the legacy cashflows to fund this nascent business, a start-up faces a more challenged environment. In addition, Apple consistently affords its biggest partners (e.g. Amazon or Vivendi) exceptions to its otherwise standard payment terms. In this case, $170-billion Nike Inc., might get a discount but start-up NewCo would not.
Apple’s byzantine rules are based on its bargaining power from a decade ago, which differentiated between extant businesses (in which Apple had limited leverage), nascent ones (in which Apple was a key growth partner) and those yet-to-be-created (in which Apple was a gatekeeper). Today, these principles are maintained by Apple’s influence — nearly every business is now a digital one, and there can be no scaled Western business that doesn’t go through the iPhone. And given Apple’s overall competitive position, its rules are essentially uncontestable.
The prosperity of the virtual world is similarly impeded by Apple’s control over which new businesses, business models and technologies exist and when. Cloud gaming is a clear example, as is WebGL and browser support. The best example, though, is Roblox, which encapsulates many of the key issues with the App Store and is explicitly pioneering our virtual future.
Bloxing In Roblox
Although Roblox is typically described as a game, it’s better described as a platform that operates an unlimited number of user-generated worlds that are accessible and integrated across all platforms, supported by an identity and payment system, and which grows through an economy of trading, creation, and scarcity. No one seems to know why this company, which is valued at more than $30 billion and has over 150 million monthly users, is allowed on the App Store.
It’s also notable that Roblox highlights policy changes at the App Store as its sixth biggest risk factor, writing: “The owners and operators of these mobile application platforms… have approval authority over our platform’s deployment on their systems and offer consumers products that compete with ours… [It also has] broad discretion to change and interpret its terms of service and policies... If we were to violate, or an operating system provider or application store believes that we have violated, its terms of service or policies, [it] could limit or discontinue our access... In some cases these requirements may not be clear or our interpretation of the requirements may not align with [its] interpretation... which could lead to inconsistent enforcement of these terms of service or policies against us, and could also result… [in] limiting or discontinuing [our] access…”. Roblox also mentions that these policy risks span “data collection and privacy practices, business models, operations, practices, advertising activities [and] application content.”
This may seem alarmist, but Apple’s recent decision to deprecate IDFA is estimated to cost Facebook and Google - two of Apple’s three FAANG competitors — billions each in 2021 revenues. If Apple can so easily harm the largest companies in the world, it’s also no surprise that while analysts such as Thompson have spoken to dozens of developers with valid App Store grievances, none of which “were willing to go on the record for fear of angering Apple.” And while Roblox slipped through Apple’s approval policies, its growth is nevertheless limited by the App Store’s lack of direct competition for payment.
Roblox’s platform is full of happy users and talented creators. But few of these creators are making money. Although Roblox has $2 billion in revenues, three billion monthly hours of playtime and more than 160 million users, only 29 developers (i.e. companies) netted over $1 million in 2020 and three crossed $10 million. This is bad; more developer revenue means more developer investment and better products for users, which in turn drives more user spending.
Unfortunately, however, it’s hard for developers to increase revenues given Roblox pays developers only 24.5% of every dollar spent on their games, assets, or items. While this makes Apple’s rates 70-85% payout rates seem generous, the reverse is true.
Consider the illustrative $100 in iOS Roblox revenue (an estimated 75-80% of all revenues). $30 goes to Apple off the top, while $31 is consumed by Roblox’s core infrastructure and safety costs, and another $11 is taken up by overhead. This leaves a total of $28 in pre-tax gross margin dollars for Roblox to reinvest in its platform. This reinvestment spans three categories: research and development (which makes the platform better for users and developers), user acquisition (which grows network effects, value for the individual player, and revenues for developers), and developer payments (which leads to the creation of better games on Roblox). Today, Roblox reinvests 23% of revenues in R&D, 7% on sales and marketing, and the aforementioned 24.5% on developer payments. As a result, it currently operates at a roughly -25% margin.
Roblox has doubtlessly enriched the digital world and led to hundreds of thousands of new digital creators. But for every $100 it creates, it loses $25, developers collect $24.5 in net revenue (i.e. before all of their development costs), and Apple collects roughly $30 in pure profits despite putting nothing at risk. The only way for Roblox to increase developer revenues today is to deepen its losses or halt its R&D, which would in turn harm both Roblox and its developers over the long-term.
Roblox’s economics should improve with scale. Overhead, sales and marketing should grow more slowly than revenues. However, this would unlock only a few percentage points to cover significant losses or marginally increase developer revenue shares. R&D should offer some scale-related margin improvements, too, but fast-growing companies shouldn’t be achieving profitability through R&D operating leverage. The company’s two largest costs, which comprise roughly 61% of revenues, are essentially fixed. Infrastructure, which largely scales with usage and will become more expensive as the platform expands its concurrency capacity and expands into VR. And Roblox doesn’t control store fees; that’s exclusively up to the platform.
No Platforms on my Platform
This (i.e. word 12,775 - lol) is where we get to the crux of the problem: the Roblox situation is a feature, not a bug. Apple's default position is that all the products created in and on platforms that are distributed by iOS should be instead individual 'apps' that can be purchased on its App Store. As a result, Apple is always the platform a consumer uses to access an app and that a developer uses to build, distribute and monetize their apps. Consider, for example, “professional” game developers like EA, rather than independent Roblox hobbyists. EA would never make a game on Roblox, where they can collect only 25% of consumer spending, when they can instead make an iOS game and collect 70%. Similarly, Apple’s cloud gaming policies prevent each of its Big Tech competitors - Google (via Stadia), Amazon (via Luna), Facebook (via Gaming), and Microsoft (via xCloud) - from operating on iOS and therefore coming between Apple and both game developers and game players.
To return to Jobs’ 2010 “Thoughts on Flash” memo:
We know from painful experience that letting a third party layer of software come between the platform and the developer ultimately results in sub-standard apps and hinders the enhancement and progress of the platform. If developers grow dependent on third party development, libraries and tools, they can only take advantage of platform enhancements if and when the third party chooses to adopt the new features. We cannot be at the mercy of a third party deciding if and when they will make our enhancements available to our developers. This becomes even worse if the third party is supplying a cross-platform development tool. The third party may not adopt enhancements from one platform unless they are available on all of their supported platforms. Hence developers only have access to the lowest common denominator set of features. Again, we cannot accept an outcome where developers are blocked from using our innovations and enhancements because they are not available on our competitor’s platforms.
Flash is a cross-platform development tool. It is not Adobe’s goal to help developers write the best iPhone, iPod and iPad apps. It is their goal to help developers write cross-platform apps. And Adobe has been painfully slow to adopt enhancements to Apple’s platforms.
Our motivation is simple — we want to provide the most advanced and innovative platform to our developers, and we want them to stand directly on the shoulders of this platform and create the best apps the world has ever seen. We want to continually enhance the platform so developers can create even more amazing, powerful, fun and useful applications. Everyone wins — we sell more devices because we have the best apps, developers reach a wider and wider audience and customer base, and users are continually delighted by the best and broadest selection of apps on any platform.
Jobs is very clear: developers should not use cross-platform development tools, but instead iOS tools. They should make the best iOS apps, not the best apps. And this will be better for users, and thus better for the developers. This logic doesn’t just apply to Flash, Unreal or Unity, but also Roblox Studio, Fortnite Creative Mode, and Minecraft. When common web standards exist, Jobs says to use the ones Apple creates and operates: “Apple even creates open standards for the web. For example, Apple began with a small open source project and created WebKit, a complete open-source HTML5 rendering engine that is the heart of the Safari web browser used in all our products. WebKit has been widely adopted. By making its WebKit technology open, Apple has set the standard for mobile web browsers.”
This position is a kind of circuitous monopoly logic. The only way developers are better off developing only for Apple is if Apple’s OS runs all relevant devices, its standards power all experiences, and the company successfully develops the devices or services for every possible category.
This isn’t true and won’t be true. And the mobile gaming ecosystem on iOS is massively bigger because developers use Unity, even when iOS drives the majority of their revenues. This is because Unity, as a cross-platform engine, allows a developer to easily reach the entire global market with their game, rather than just iOS users. With more users comes more revenues and a better game, which in turn benefits both App Store revenues and iOS users. And because Unity and Unreal are focused on being the best possible cross-platform game engines and in turn, benefit from a wide range of customer innovations, the entire game industry benefits from lower prices and better capabilities. In addition, cross-platform games like Fortnite, Roblox and Call of Duty are so powerful because they connect players across every device, rather than just the devices Apple makes.
The success of cross-platform tools and experiences may suggest Apple’s control isn’t a problem. But this assessment ignores Apple’s attempt to ban Unreal, its success stymying web-based rendering engines and cloud games, the fact no one knows if Roblox is actually allowed on iOS or just a grandfathered accident, and how Apple’s commissions constrain platform-like apps such as Roblox and Snapchat (which doesn’t offer its own “in-app app store” for this very reason - even though this model is incredibly popular and lucrative in Asia).
In effect, the only way a Roblox developer could collect a substantially greater share of their game revenues would be if (1) Apple built its own Roblox-like platform; (2) all eligible users had and only wanted to use their iOS devices to play Appleblox; and (3) Apple operated Appleblox at break-even (which the App Store was intended to do but doesn’t) or didn’t pay fees to the Apple App Store (which all Apple services do).
Apple’s efforts to avoid platform intermediation are wide-ranging and ever-growing. In 2020, the company revised its App Store policy such that (with a few exceptions) any iOS app that used third party identity systems (e.g. log-in using your Facebook or Gmail account) would also need to support the Apple account system. In order words, Apple announced that “if you want your apps on iOS, and your apps support the account networks of our competitors, you must also deploy our account solution”. Every other account provider must earn (or purchase) this business. Apple, meanwhile, can and does legislate it. And in doing so, it is able to occlude its horizontal platform competitors.
Collectively, Apple’s anti-platform policies and philosophy do more than prevent competition on today’s Internet - they also impede the development of a new one. Apple does not want a digital world built and innovated upon interoperable standards, device/endpoint agnosticity, and without Cupertino. Where virtual items or currencies are freely exchanged between applications and virtual worlds. With marketplaces for labor, creation, and consumption that that disintermediate Apple or control the never ending process of disaggregation and re-aggregation. Instead, Apple believes the next iteration of the web should just be a “virtual app store”, with your avatar just a virtual version of your Apple ID, your bank that of Apple Pay, and your experiences based on what Apple envisions and enables, when, and how.
This is no surprise. The “next” Internet the greatest opportunity for market growth and happier consumers, but also to displace today’s market leaders (just as Apple’s iOS supplanted Microsoft’s Windows). Apple can’t stop the forever, but that doesn’t mean there’s no harm. Every additional use case and capability means consumers will spend more in the digital/virtual world, and companies will invest more, make better products, and be more successful.
Peace, Prosperity and Good Governance
The worst part of the ‘Apple Problem’ is that everyone knows Apple’s policies are a bottleneck to business creation, new business models and new products — and the dominant response is just to wait for them to change. Everyone knew, for example, that Apple was unnecessarily holding up the cloud gaming industry and would need to revamp its rules. And thus nine years after the first service emerged, and some two months after declaring the services were security risks, Apple allowed for cloud gaming apps. Yet everyone still considers these policies unsatisfactory, so the industry has returned to waiting. Similarly, everyone knows Apple won’t be able to keep charging its store fees to services they compete with, such as Spotify; the question is when they’ll crack. Even a few more months of +15%, or exclusivity in game bundling, can produce billions of cash and elongate first mover advantages.
After Apple revised its cloud gaming policies, The Verge wrote “Arguing over whether Apple’s guidelines did or didn’t include a thing is kind of pointless, though, because Apple has ultimate authority. The company can interpret the guidelines however it chooses, enforce them when it wants, and change them at will.” This is not the right foundation for the future.
Chapter Six: Solving the Apple Problem
It’s likely that a combination of accumulating policy contradictions and mounting legal pressure will push Apple to make important concessions in the coming quarters and years. And every single significant opening will lead to new products, companies and perhaps even a new mainstream platform.
At the same time, we should also recognize that the core design principles that underpin the iOS/App Store platform are nearly twenty years old and grew out of a store for songs, and then for simple apps. And despite the enormous change in the digital world over this time - including the addition of billions of new Internet users, millions of new digital-only businesses and scores of new technologies and ideas - Apple’s principles have never been comprehensively overhauled. Accordingly, the company’s concessions are likely to be overly complex, onerous, and insufficient. And given the importance and influence of the iOS economy we should not rely only on slow-to-cook pressures or voluntary concessions.
Apple has the right to run its own store, offer its own standards, and develop services that are exclusive to its hardware. The problems arise from Apple’s forced bundling of hardware, an operating system, distribution system, payment solution and services. As a result, there is a straightforward remedy — forcing Apple into competition in app distribution and payments. Specifically, regulators should require Apple to:
Allow iOS users to download apps from any source (as they do on Windows and Mac computers), including direct from the software maker
Allow iOS users to use third-party app stores on iOS devices
Allow developers to use payment solutions other than those of Apple’s App Store, whenever they choose and even when distributed via Apple’s App Store
This partial unbundling would benefit even those who continued to download their apps exclusively from Apple, and pay for them via Apple’s billing system whenever possible. This is because the App Store would need to compete directly for the business of every app user and app developer, rather than win this business via its iOS devices and then control it via iOS policies.
This doesn’t mean Apple wouldn’t be able to charge higher prices to consumers, or collect above-average fees from developers. But just like any retailer, the Apple App Store would need to earn its consumers via their brand, curation, ease of use, reliability, safety, and earn suppliers (i.e. developers) through the ability to increase their sales net of store fees.
Apple would still have near-hegemonic advantages, too, such as its pre-installation on iOS devices, its 13+ year head start, world-class catalogue, and most of all, unmatched user loyalty. Furthermore, the company would likely replicate the software installation policy it deploys on the Mac — i.e. telling users it has not signed for apps downloaded from outside the Mac App Store. This would help it maintain market share and dissuade a substantial portion of its users from considering alternatives.
But given the chance, many developers would begin distributing and monetizing their apps directly, and/or through third party stores. This pathway would offer consumers net lower prices, while also maintaining net unit revenues for developers and likely leading to greater unit sales too. Or it would enable developers to increase their gross profit under existing prices. This, in turn, would put pressure on Apple to lower and standardize its store fees in order to retain app users and developers.
More broadly, the ability to opt-out of Apple’s App Store would mean that Apple’s control over industry technology and standards would slightly decline, and hopefully, encourage the company to prove which of its limitations are truly for “security”. Businesses such as Spotify, Prime Video, and Game Pass would also be able to match the gross margins of Apple Music, Apple TV+, and Apple Arcade. That doesn’t seem like a bad thing.
It may feel unfair to force Apple to loosen the controls that led it to such unprecedented success and adoption. Yet problems arising from Apple’s controls are becoming larger every day, as is the company’s unprecedented strength. The future of the global economy is digital and virtual. Broad prosperity depends on platforms that compete to create value for developers and users, and that give birth to new platforms that do the same. Apple is not meeting the moment. The defenses it provides for the controls it demands are not convincing. They neither demonstrate that its policies primarily benefit customers, nor that these benefits outweigh their downsides or anti-competitive side-effects.
Matthew Ball (@ballmatthew)
