This past July, MIT’s Initiative on the Digital Economy held its 10th annual Platform Strategy Summit. The hybrid Summit included a number of panels and keynotes throughout the day, a few of which I attended online. In particular, I’d like to discuss the very interesting presentation by Boston University professor Marshall Van Allstyne as part of a panel on emerging trends in which he raised and answered five fundamental questions about platforms:
- Structure: Why do platforms have such high market caps but so few assets or employees?
- Strategy: Why does traditional product strategy fail in platform markets?
- Regulation: Why does traditional antitrust fail in platform markets?
- Fake News: Why is misinformation such a hard problem?
- Can Decentralized Autonomous Organizations dethrone platforms?
In his presentation, Van Alstyne, aimed to articulate a single framework that will help answer all five questions, that is, a consistent way of thinking about platforms based on the research that he and his colleagues have been conducting over the past decade. Let me summarize Van Alstyne’s presentation by first describing his unifying framework, followed by how the framework applies to each of the five key questions.
Platforms have long played a key role in the IT industry. IBM’s System 360 family of mainframes, announced in 1964, featured a common hardware architecture and operating system. The ecosystem of add-on hardware, software and services that developed around System 360 helped it become the premier platform for commercial computing over the next 25 years.
In the 1980s, the explosive growth of personal computers was largely driven by the emergence of the Wintel platform based on Microsoft’s operating systems and Intel’s microprocessors, which attracted a large ecosystem of hardware and software developers.
The commercial success of the internet in the 1990s drove platforms to a whole new level. Internet-based platforms connected large numbers of PC users to a wide variety of web sites and online applications. The economic power of platforms has grown even more dramatically over the past decade, with billions of users now connecting via smart mobile devices to all kinds of cloud-based applications and services.
The internet’s universal reach and connectivity has led to increasingly powerful network effects and to the rise of platform economies. Network effects are all accompanied by externalities, that is, an indirect cost or benefit to an uninvolved third party. In a social network, for example, the connection benefits you get attracts and bring benefits to other individuals, each of which then attracts others making the network more valuable for all.
Scale increases a platform value. In a two-sided commerce platform, for example, the more products or services the platform offers, the more consumers it will attract, helping it then attract more offerings, which in turn brings in more consumers, which then makes the platform even more valuable for everyone. Moreover, the larger the network, the more data is available to personalize recommendations, further increasing the platform’s value.
What is often missed, said Van Alstyne, is that leveraging network effects to benefit other people or to attract third parties requires the orchestration and design of the platforms’ algorithms, and these don’t happen by accident. In the digital economy, translating network effects into economic advantage requires careful attention to the governance of the platforms.
This is the essence of his unifying framework. Let’s now see how it applies to each of the five platform questions he posed.
Why do platforms have such high market caps but so few assets or employees? Van Alstyne showed a slide that compared the market cap per employee of several platform versus traditional companies in a variety of industries. They ranged from 4X for Twitter (market cap $28 billion, 7,500 employees) vs NY Times (market cap $5 billion, 5,000 employees), to 27X for Airbnb (market cap $61 billion, 6,000 employees) vs Marriott (market cap $46 billion, 120,000 employees). In the Forbes global 2000, platform companies have higher market caps ($21,726 vs $8,243), higher margins (21% vs 12%), and half the employees (9,800 vs 19,000) than traditional companies.
The reason for the difference is that in traditional companies the production of value is done by its internal employees. Platform companies have successfully inverted the firm, so that the production of value is not just done by their internal employees but by leveraging their much larger external communities of users and clients,- e.g., those who post content on Twitter and Facebook, rent rooms and houses on Airbnb, and drive cars for Uber.
Why does traditional product strategy fail in platform markets? The industrial economy of the past two centuries has been driven by supply-side economies of scale. Because of the massive fixed costs of physical assets, firms achieving higher volumes have a lower cost of doing business, which allows them to reduce costs and further increase volumes. Market power is thus achieved by increasing efficiency, becoming more profitable, and fending off competition.
But the nature of competition and strategy are quite different in a platform-based business, where the driving force is demand-side economies of scale. The community of users and providers that a platform company is able to attract, retain and grow is its most important asset. You’re not just valuing the assets of the firm, you now have to value the assets of the entire community.
Traditional companies are focused on controlling their internal assets and building a moat around the business to keep potential competitors out. In a properly orchestrated platform business, - that is one where you control the information and monetary flows, - even your competitors can become unique sources of value by bringing them onto the platform, and thus helping attract an even larger community of users and consumers. Apple and Google’s Android want as many developers as possible on their app stores; Amazon wants as many merchants as possible on its platform.
Why does traditional antitrust fail in platform markets? Van Alstyne argues that while the federal government has properly identified that there are issues with the market power of the largest platform companies, their solutions are wrong. How do you define market share dominance when the markets of platforms companies are so hard to define. Is Amazon in books, cloud, e-commerce, entertainment, home devices, or groceries? Is Google in search, email, maps, home devices, or self-driving cars?
Traditional economics teaches us that one way companies become monopolies is by restricting their output so they can charge more for their products and services. But, Google is not restricting search, Facebook is not restricting posts, and Amazon is not restricting purchases. Another classic test for monopolies is their pricing, whether it’s too high and consumers are being gouged, or too low to drive out the competition. But those tests don’t work with platform companies because they are in so many markets that their business model is to subsidize or give stuff away in one market in order to grow the size and increase the value of another market. In addition, classic antitrust remedies, like breaking up companies don’t work so well with inverted firms like platform companies. Since the central power of platform companies stems from their ability to leverage huge amounts of consumer data, new antitrust remedies, like improving the portability of consumer data are needed.
Why is misinformation such a hard problem? The network effects discussed so far are all examples of positive externalities, where the activities of one individual or group create value for others. But misinformation is a negative externality that causes damage to others, such as conspiracy theories that lead to bad medical decisions, political polarization, or insurrections.
This is one of the most challenging problems our society faces. Why is fake news so difficult to control? The government cannot do it because freedom of speech, including misinformation, is protected by the first amendment. Federal, state and local governments are thus barred from taking action to correct the problem. As a result, negative externalities like misinformation have to be corrected by markets, and so far, we don’t have market remedies for dealing with and correcting negative externalities.
Can Decentralized Autonomous Organizations dethrone platforms? Decentralized Autonomous Organizations (DAOs) are a major part of Web3, a promising set of technologies and applications that aim to replace today’s corporate mega-platforms with blockchain-based decentralized networks, thus leading the way to a more open, entrepreneurial internet and a middleman-free digital economy.
Can DAOs dethrone platforms, asked Van Alstyne as he concluded his presentation, by removing middlemen and reducing transactions costs? Yes, he said, “but that’s only half the story. The other half of the story is again the positive orchestration of externalities.” And this cannot happen if all you do is decentralize. You have to have governance to create and orchestrate the necessary positive externalities. “DAOs that don’t have governance will never displace platforms,” he said in conclusion. But DAOs that implement governance, such as adding smart contracts, could potentially be a threat to platforms with very interesting new implications.
Comments