Mastering One’s Domain: Some Key Principles of Platform Capitalism

The disruption brought by platforms hides some of their other effects: Platforms make possible a form of total control. It […]

The disruption brought by platforms hides some of their other effects: Platforms make possible a form of total control. It is therefore crucial to see how platforms are combined, effectively creating media ecosystems.

In the Internet economy, the key to economic clout is to maintain exclusive control over the surface on which the market exchange takes place. While the value of individual bits strives towards zero, the value of the infrastructure approaches dizzying sums. This model has, increasingly, been applied to physical markets as well, as part of what is sometimes called the “sharing economy”—but should more aptly be called “platform capitalism,” since it is facilitated by stable intermediaries (like Facebook and Google) that provide storage, navigation and delivery of the digital content produced by others. These intermediaries—digital platforms—can be seen as technological and material “stages” that provide leverage, durability, and visibility. The problem is that the underlying principles of this platform capitalism seem to beget an inherent form of monopolism.

As “everything” is rapidly becoming digitized, let’s keep in mind that the Internet has,rather rapidly, become privatized and is now dominated by a handful of corporations (global giants like Apple, Google, Amazon, Facebook etc.), each with considerable market capitalization. Although many of the individual entrepreneurs behind these emerging actors may be partially independent of these giants, the basic premise is that each platform has to reach a dominant position in its respective market. In addition, any new platforms will generally be dependent on already established platforms. Ultimately, those platforms that do become truly successful tend to be bought up by the established giants.

In an increasing number of markets where the major platform actors have come to dominate, it is nowadays practically impossible for new entrants to be able to acquire market share. While it is hardly possible to offer as fast, flexible and affordable a cloud service as those of Amazon, Google, or Microsoft without enormous financial muscle, it is not lack of capital that seems to be the reason Europe is struggling to compete with the incumbent platform giants. While the world’s leading digital platform businesses have a combined market capitalization of $4 trillion, only 4 % of this value has been generated by European firms.

It is when considering the nature of digital infrastructure, and viewing it as a power principle, that platformization really gets interesting. I would argue that platformization is, in many respects, the key principle of economic control in our time, even comparable to established paradigms of corporate management like Fordism and Taylorism. While these paradigms, typical for the 20th century, are generally associated with serialization, later developments like post-Taylorism are primarily concerned with coordination and “lean production,” where value-chain production becomes intensified, more elaborate, sophisticated, complex, and more coordinated in time and space. However, these late-20th-century intensifications were still based on value chains rather than value networks. Platform capitalism is an even deeper intensification of capitalism, based directly on two things: the nature of digital infrastructure and the nature of networks.

All hail the platform?

As established media powerhouses worldwide—be it the BBC, Springer, The New York Times, or Schibsted—are challenged by international platform giants, one key solution has, strikingly, been to build their own platforms (or so their publicly stated intentions go, at least). Alternatively, another solution has been to simply begin collaborating with the tech giants, as Facebook’s budding Instant Articles and Google’s Digital News Initiative show.

While the digital nature of platforms is truly new, the imaginaries of real-time “matching,” “feedback” and “automation,” fueling a lot of current developments, echo a form of “information idealism” reminiscent of the late 20th century and themodern planning of that era. Further, when seen as an overarching structural principle, platformization has a worrisome tendency towards monopolistic control. Much as the binary nature of data makes things either one or zero and never anything in-between, the corporate tendency to full market dominance, already familiar to capitalism, becomes even more uncompromising.

My own approach comes from years of studying the file-sharing culture and its economic implications. The file-sharing ecology is characterized by the freedom to tinker and the ability to index, upload and modify files and file listings on surfaces that are owned and controlled by the users themselves. One of my main insights has been that The Pirate Bay serves as a de facto platform—a singular registry and search engine, charting some of the otherwise nebulous file-sharing sprawl. For many people, The Pirate Bay has come to symbolize file sharing in its entirety.

Many things could be learned from The Pirate Bay, but one of the key insights is that massive popularity and market dominance can be generated by reaching critical mass of utility through concentrating access to all the assets (database of objects/subjects) in one single place. In addition, if what is provided is more than mere access, but the actual hosting of content itself—much like another innovative Swedish tech stalwart emerging from the file-sharing culture, Spotify, managed to do—then the dominance is additionally solidified, and future competitors are discouraged even further from viable market entry.

“Actually, capitalism and competition are opposites,” Peter Thiel has convincingly pointed out. That firms seek market dominance is nothing new; what is new is the “all-or-nothing” nature inherent to digitization, where services and platforms tend to become totalitarian. In an era of platform capitalism, firms have incentives to create what could be called “creative monopolies”; conditional systems within which new services are constantly offered and new markets emerge—provided that the platform owner maintains a monopolistic control over what happens.

At the same time, it’s important to admit the huge innovative potential that comes with information empires. In modern history, such potential has become realized many times, during various different techno-capitalistic reigns, such as that of AT&T and Bell Labs during the mid-20th century, as Tim Wu has brilliantly recounted. It is often said to be in the interest of consumers that platform-based providers cover as much of the available market as possible: It often brings both convenience and cost-efficiency for consumers when the entire range of choice is gathered in one single app. Still, we might be fooling ourselves in the long run, as the glorious supremacy of digital “one-stop shops” seems to have deep societal costs that are hidden at first glance, such as an impending erosion of the middle class.

Platforms incrementally enabling yet more platforms?

The idea that we’re living in a platform economy is sometimes labeled old news. Thereally novel news would be the emergence of structures that operate through complex, nested arrangements of “platforms-of-platforms”—Alphabet being the quintessential example.

Once upon a time, Google and Facebook were digital platforms simply providing search and social networking; Amazon was a retailer; and Apple a digital hardware manufacturer. However, these actors have long since diversified into various other markets. Technically, they also work as platforms on which other platforms are, in turn, built. Platforms have to be situated in ever-wider systems of mutual interplay, co-dependence, and productivity. To begin with, platform companies are directly dependent on either venture capital (privately traded companies) or stock market valuations (publicly traded companies). Facebook, for example, can make stunning investments thanks to credit-based liquidity resulting from the company’s extraordinarily high valuation on the stock market. Moreover, none of the global Internet companies would be what they are today if the leading telecom corporations hadn’t allowed a relatively free flow of data in their cables and radio links. Budding platform companies like Uber would never have been possible if larger platform companies like Apple and Alphabet hadn’t allowed a certain degree of freedom of innovation in the new economic ecosystems that are generated through app stores (allowing companies to plant software on mobile Internet-connected appliances) and map applications (allowing geographical coordination of software and devices). Spotify, which has historically relied on its own servers, has recently announced that the company is migrating its infrastructure to the Google Cloud Platform. Similarly, Netflix, while accounting for 37 % of all Internet traffic in North America during peak viewing times, is arguably more dependent than ever on Amazon Web Services for its hosting and traffic.

When Apple enables an ecosystem of apps, the degree of prediction is low in regards to which new markets and business opportunities can be built on top of such a software platform. In that sense, global platform companies enable new establishments of lesser platform companies, such as Uber and Airbnb, as these latter actors combine mobile operating systems platforms for apps (App Store, Google Play) with the mapping services offered by the same platform giants (Apple Maps, Google Maps). Yet, as we shall see, one should not conflate a sprawling information infrastructure (an ecosystem) with a platform (a particular surface).

Despite the strong form of local, platform-specific technical control inherent to digital code, some of the academic literature highlights the relative lack of control over the ways in which platforms develop over time and interrelate with other platforms. As technical architecture, a platform allows for large sets of IT capabilities to be crammed into a relatively well-bounded and controlled system, which can be continuously re-designed and expanded. While the design of a platform often starts off with a bounded set of closed specifications, it often grows in complexity over time, as platforms are expected to meet varying user needs and facilitate various forms of compatibility.

However, it is important to distinguish singular platforms and applications from the much wider, more complex, and more dynamic information infrastructures that they make part of. Sometimes, when people praise the relative openness and flexibility of the Apple or Google ecosystem, in the same breath they confuse this with the singular platforms in question. While companies like Apple and Alphabet are, in effect, complex arrangements of interrelated platforms, each such platform might, however, be rather restricted in terms of sheer functionality. An iPhone forms the nexus of a diverse information infrastructure, yet some of its constituent platforms (such as the iTunes interface) might in fact be highly constrained, path-dependent, and not at all flexible.

A key insight, nevertheless, is that the nested “platform hierarchy” of our current global information infrastructure makes many of the lesser platforms quite dependent on the top providers in (i.e. the global giants), arguably further solidifying the dominance of these latter actors.

Exclusive control

Platform control could be defined as exclusive control over the surface on which the exchange takes place. This does not mean that whatever happens on Facebook is determined by Facebook, but it does mean that Facebook has the irrevocable and absolute sovereignty to boot you out if you break the rules. While monopolies, consolidation, and market dominance are familiar phenomena within capitalism, I would argue that what is genuinely new with digitization is the concept of total control that is implemented when law and norm is crystallized into code.

I don’t intend to associate platformization with Taylorism, as if these were idealized “logics” that market actors invoke, regardless of the actual behavior of these actors. Rather, platformization, being highly dependent on technology, is a material organizing principle as regards the division of labor and resource efficiency in contemporary society. Of course, a proviso must be added that the specific effects of the respective platforms (not to mention the combined effects of entire ecosystems of platforms) may be highly unpredictable, especially when considering the externalities created. Platforms are charged with a “paradoxical tensionbetween the logic of generative and democratic innovations and the logic of infrastructural control.” Apple and Alphabet currently have to allow for quite significant degrees of freedom of innovation among the app startups crowding both App Store and Google Play, but there is nothing absolute to this degree of freedom. As some of Apple’s App Store policies make clear, the rules of engagement are sometimes rather arbitrary: The company has banned “content that ridicules public figures” and “sexually suggestive apps,” yet allows for titillating material from mainstream publishers like Playboy. It prohibits all services that mimic the functionality of existing Apple services. More disturbingly, the company has rejected apps for reasons it never warned developers of in advance, and reserves the right to change its policies at any time, without prior warning. Platforms become like opaqueblack boxes: We don’t know what they do, and you’ll be punished for peeking inside.

Regardless of the circumstances of their inception, the competitive goal of any platform company, Uber and Airbnb included, is to seek a monopoly in its respective niche. Digitization enables a hardcore form of standardization and tracking, in regards to measurements of what takes place, where and when. In order to generate revenue from the many, sprawling, seemingly chaotic interactions taking place on a platform, certain standardized rules and metrics have to be imposed: All transactions must be billable. Further, in order to make possible some kind of economy of scale, compliance has to be automated: Hardware setups and software algorithms automatically assign billing orders to your transaction. Moreover, none of the participating nodes on the platform should be allowed to abuse their relative freedom by using loopholes or glitches: The system has to be “water-tight.” No free lunches by mistake.

Due to reasons such as these—virtually all of them stemming from the existential fact that digital code by its very nature is inexorable —platform-based control is a form of total control. This is why so many people have come to see the algorithmic mode of management as, for good and for bad, “better than law.” It is thought of as an inhuman, perfect form of institutional functionality where individuals are freed from subjective decision-making. My closing word on the matter will therefore be one of warning: We have walked down similar paths before, in terms of the ways in which power is arranged in society. It might appear “disruptive” at first, as various actors battle for the platform real-estate. However, over time, platform capitalism risks becoming the polar opposite, helping to further consolidate those power structures that are already in place.