Platform Power Examined: How Digital Platforms Control Markets, Shape Behavior, and Resist Accountability
In January 2021, Apple removed Parler--a social media application popular among right-wing users--from its App Store. Google removed Parler from the Google Play Store on the same day. Amazon Web Services terminated Parler's hosting contract shortly afterward. Within 48 hours, Parler had been effectively eliminated from the internet. The company had not been convicted of a crime. No court had ordered its removal. Three technology companies, exercising their contractual rights as platform operators, had made a unilateral decision that a social media service used by millions of people should cease to exist.
Whether Parler's removal was justified is a matter of intense debate. What is not debatable is what the incident revealed about platform power: the extraordinary control that a small number of digital platforms exercise over who can participate in the digital economy, what they can say, what they can build, and how they can reach their audiences.
Platform power is not an abstract concept. It is a concrete, daily reality for billions of people and millions of businesses. When Google changes its search algorithm, businesses gain or lose customers overnight. When Facebook adjusts its News Feed algorithm, publishers gain or lose audiences. When Apple changes its App Store policies, entire business models become viable or unviable. When Amazon adjusts its marketplace rankings, sellers' livelihoods rise or fall. These platforms do not merely participate in markets--they are the markets, and the rules they set determine who wins and who loses.
Understanding platform power--how platforms acquire it, how they exercise it, what harms it produces, and how it might be constrained--is essential for anyone seeking to comprehend the structure of the modern digital economy.
What Is Platform Power?
Platforms as Intermediaries
A digital platform is an intermediary that connects two or more groups of users and facilitates interactions between them. Amazon connects buyers and sellers. Google connects searchers and websites. Uber connects riders and drivers. Airbnb connects travelers and hosts. App stores connect users and developers.
What distinguishes platforms from traditional intermediaries is the scale and nature of their control. A traditional marketplace (a farmers' market, a shopping mall) provides a physical space where buyers and sellers interact. The marketplace operator sets some rules (opening hours, booth fees, sanitation requirements) but does not control the interaction between buyer and seller at a granular level.
Digital platforms operate differently. They do not merely provide a space for interaction; they mediate and shape every interaction through algorithms, interfaces, terms of service, and data collection. A seller on Amazon does not simply display goods on a shelf; their products are ranked, priced, promoted, and presented according to Amazon's algorithms. A website does not simply exist on the internet; its visibility is determined by Google's search algorithm. A driver on Uber does not simply pick up passengers; their fares, routes, ratings, and work conditions are determined by Uber's platform.
This granular control over interactions gives platforms a form of power that has no direct precedent in economic history. Platforms are simultaneously market makers (creating the conditions for economic activity), regulators (setting and enforcing rules), competitors (offering their own products alongside those of third parties), and infrastructure providers (operating the technical systems on which economic activity depends).
Dimensions of Platform Power
Platform power operates across multiple dimensions:
Market access: Platforms control who can participate in the markets they mediate. An app that is rejected from the App Store cannot reach iPhone users. A seller that is banned from Amazon loses access to the platform's hundreds of millions of customers. A website that is de-indexed from Google effectively disappears from the internet.
Algorithmic visibility: Within platform markets, algorithms determine which products, content, and services are visible to users. A change in Google's search algorithm can move a website from the first page of results to the tenth--effectively from visibility to invisibility. A change in Facebook's News Feed algorithm can increase or decrease a publisher's audience by orders of magnitude.
Data asymmetry: Platforms collect vast amounts of data about the behavior of users, sellers, and developers on their platforms. This data provides platforms with insights that no individual participant can match, creating information asymmetries that platforms can exploit for competitive advantage.
Terms of service: Platforms set the terms under which participants can operate, and they can change those terms unilaterally. These terms often include provisions that would be considered unconscionable in other contractual contexts: mandatory arbitration, intellectual property assignments, liability limitations, and rights to terminate service without cause.
Pricing power: Platforms set fees, commissions, and pricing structures that determine how economic value is distributed between the platform and its participants. Apple's 30% commission on App Store purchases, Amazon's various seller fees, and Uber's take rate from drivers are all exercises of platform pricing power.
How Do Platforms Gain Power?
Network Effects
The primary mechanism through which platforms acquire power is network effects: the phenomenon in which a product or service becomes more valuable as more people use it.
Network effects operate in two forms:
Direct (same-side) network effects: The value of the platform to each user increases as more users of the same type join. A social network becomes more valuable to each user as more of their friends join. A messaging app becomes more useful as more contacts adopt it.
Indirect (cross-side) network effects: The value of the platform to one type of user increases as more users of a different type join. A marketplace becomes more valuable to buyers as more sellers join (and vice versa). An operating system becomes more valuable to users as more developers create applications for it.
Network effects create winner-take-most dynamics: once a platform achieves a critical mass of users, the value it provides to each user exceeds what any competitor can match, because the competitor's smaller user base makes its network effects weaker. This dynamic explains why most platform markets are dominated by one or two players: Google in search, Facebook in social networking, Amazon in e-commerce, Uber in ride-sharing, Airbnb in short-term rentals.
Data Accumulation
Platforms accumulate data from every interaction on their systems. This data serves multiple purposes:
- Product improvement: More data enables better algorithms, better recommendations, better fraud detection, and better user experiences, which attracts more users, which generates more data (a self-reinforcing cycle)
- Competitive intelligence: Platforms can observe what products are selling, what content is popular, what features users want, and where market opportunities exist--information they can use to develop competing products or services
- Advertising targeting: User data enables precise advertising targeting, which is the primary revenue source for platforms like Google and Facebook/Meta
- Barrier to entry: The data accumulated by established platforms cannot be replicated by new entrants, creating a durable competitive advantage
Switching Costs
Platforms create switching costs that make it difficult for users to move to competing platforms:
- Data lock-in: Users' data (photos, messages, documents, purchase history, social connections) is stored on the platform and may be difficult or impossible to export
- Social lock-in: Users' social connections are on the platform, and leaving means losing access to those connections
- Ecosystem lock-in: Users who have invested in a platform's ecosystem (purchased apps, learned workflows, integrated with other services) face significant costs in switching to a different ecosystem
- Contractual lock-in: Business users may be bound by contracts, licensing agreements, or terms of service that restrict their ability to switch
| Power Mechanism | How It Works | Example |
|---|---|---|
| Direct network effects | More users make the platform more valuable for each user | WhatsApp: your contacts are there |
| Indirect network effects | More of one user type attracts more of another type | App Store: more users attract more developers |
| Data accumulation | Usage data improves the product, attracting more users | Google Search: more searches improve results |
| Switching costs | Leaving the platform is costly or impractical | Apple ecosystem: locked into iCloud, apps, devices |
| Ecosystem control | Platform controls tools, APIs, and infrastructure others depend on | AWS: businesses built on AWS services |
| Brand and trust | Established reputation deters users from trying alternatives | Amazon: trusted for delivery, returns, reviews |
How Do Platforms Exercise Power?
Algorithmic Control
The most pervasive exercise of platform power is through algorithms: the automated systems that determine what users see, what content is promoted, what products are recommended, and how information is organized.
Algorithmic control is powerful precisely because it is invisible and unilateral. When Google changes its search algorithm, websites experience dramatic changes in traffic, but they cannot see the algorithm, understand why their ranking changed, or appeal the decision to an independent authority. When Instagram changes its recommendation algorithm, creators experience changes in reach and engagement, but they have no visibility into the algorithm's logic and no recourse when the changes are harmful.
The opacity of algorithmic control creates a power asymmetry that is qualitatively different from other forms of market power. A traditional monopolist exercises power through pricing, supply control, or exclusive contracts--mechanisms that are visible and subject to legal challenge. A platform exercises power through algorithmic decisions that are invisible, proprietary, and practically unaccountable.
Fee Extraction
Platforms extract economic value from their participants through fees and commissions that participants have limited ability to negotiate or avoid:
- Apple and Google charge a 30% commission (reduced to 15% for small developers in some cases) on most App Store and Play Store purchases. Developers who want to reach iPhone or Android users must accept these terms; there is no alternative distribution channel for most mobile applications.
- Amazon charges sellers various fees (referral fees, fulfillment fees, advertising fees, storage fees) that can consume 30-50% of a product's selling price. Sellers accept these fees because Amazon controls access to hundreds of millions of customers.
- Uber and Lyft take 20-40% of each fare. Drivers accept these rates because the platforms control access to riders.
- YouTube retains 45% of advertising revenue generated by creators' content. Creators accept this split because YouTube controls access to the largest video audience in the world.
These fee structures are exercises of platform power: the platform sets the fees unilaterally, and participants must accept them because the platform controls access to the market.
Self-Preferencing
Platforms use their position as market intermediaries to favor their own products and services over those of third parties:
- Amazon displays its own branded products (Amazon Basics, Amazon Essentials) prominently in search results alongside third-party products, using data gathered from third-party sellers to identify profitable product categories
- Google has been found by European regulators to favor its own comparison shopping service in search results over competing services
- Apple pre-installs its own applications (Apple Maps, Safari, Apple Music) on iPhones and makes it difficult or impossible for users to set third-party applications as defaults for key functions
- Meta has historically adjusted its News Feed algorithm to favor its own video service (Facebook Watch) over competing platforms like YouTube
Self-preferencing is particularly problematic because the platform is simultaneously the referee and a player: it sets the rules of the market and competes within that market, with access to data and control mechanisms that no other player possesses.
What Are the Harms of Platform Power?
Economic Harms
Platform power produces several categories of economic harm:
Rent extraction: Platforms capture an increasing share of the economic value generated on their systems, leaving less for the businesses and individuals who create that value. When Apple takes 30% of every app sale, developers receive less revenue for their work. When Amazon takes 30-50% of seller revenue through various fees, sellers' margins are compressed.
Market foreclosure: Platforms can prevent competitors from reaching customers by controlling distribution channels. If Apple refuses to approve an app, or if Google de-indexes a website, the affected business loses access to its market. This foreclosure power is particularly concerning when the platform uses it to protect its own competitive position.
Innovation suppression: Platform power can suppress innovation by creating a "kill zone" around the platform's core business. Venture capitalists have reported reluctance to fund startups that compete with platform companies, because the platform can replicate the startup's functionality and distribute it to its existing user base at marginal cost. If your startup's product can be replicated as a feature of an existing platform, your startup may be unfundable.
Social and Democratic Harms
Platform power extends beyond economic effects to produce social and democratic harms:
Information control: Platforms' algorithmic decisions shape what information billions of people encounter. Google's search results, Facebook's News Feed, YouTube's recommendations, and TikTok's For You page are among the most powerful information gatekeepers in human history. The decisions these algorithms make about what to show, what to promote, and what to suppress influence public opinion, political discourse, and cultural values at unprecedented scale.
Speech governance: Platforms make daily decisions about what speech is permitted, what is removed, and what is algorithmically suppressed. These decisions affect political discourse, artistic expression, journalism, and activism. The platforms' moderation policies--developed internally, applied inconsistently, and enforced without due process--constitute a form of private governance over public discourse that has no democratic accountability.
Surveillance: Platforms collect comprehensive data about users' behavior, preferences, social connections, locations, purchases, and communications. This surveillance enables targeted advertising (the primary business model of many platforms) but also creates risks of manipulation, discrimination, and abuse of sensitive personal information.
Labor market power: Gig economy platforms (Uber, Lyft, DoorDash, Instacart) exercise power over workers who depend on the platform for income but are classified as independent contractors without employee protections. The platform sets pay rates, work conditions, and performance standards unilaterally, while workers bear the costs and risks of the work.
Can Platforms Be Held Accountable?
Regulatory Challenges
Holding platforms accountable for the harms their power produces faces several structural challenges:
Jurisdictional complexity: Platforms operate globally but are regulated nationally. A platform headquartered in the United States serving users in Europe, Asia, Africa, and Latin America is subject to multiple, often conflicting regulatory frameworks. Platforms can exploit jurisdictional gaps by locating operations in favorable regulatory environments.
Technical complexity: Platform power operates through algorithms, data systems, and technical architectures that regulators often lack the expertise to understand. Regulating algorithmic decision-making requires technical knowledge that many regulatory agencies do not possess.
Regulatory capture: Platforms invest heavily in lobbying, political donations, and relationships with regulators. The revolving door between technology companies and regulatory agencies creates conflicts of interest that can compromise regulatory independence.
Speed of change: Technology evolves faster than regulation. By the time regulators understand and respond to a platform's practices, the platform may have already moved to a different strategy.
Emerging Regulatory Responses
Despite these challenges, regulatory responses to platform power are emerging:
European Union: The EU has been the most active regulator of platform power. The Digital Markets Act (DMA), which took effect in 2023, designates certain large platforms as "gatekeepers" and imposes specific obligations: interoperability requirements, restrictions on self-preferencing, data portability mandates, and prohibitions on certain anti-competitive practices. The Digital Services Act (DSA) imposes transparency and accountability requirements on platforms' content moderation practices.
United States: The US has pursued antitrust enforcement against major platforms. The Department of Justice filed an antitrust suit against Google in 2020, and in 2024 a federal judge ruled that Google had maintained an illegal monopoly in search. The Federal Trade Commission filed an antitrust suit against Meta (Facebook) in 2020. Congressional committees have investigated platform power and proposed legislative reforms, including interoperability mandates, restrictions on self-preferencing, and requirements for algorithmic transparency.
Other jurisdictions: India, Japan, South Korea, Australia, and other countries have implemented or proposed regulations targeting platform power, including requirements for fair dealing, data protection, and competition.
What Are Potential Solutions?
Structural Remedies
Antitrust enforcement can address platform power through structural remedies: breaking up platform conglomerates into separate companies. Proposals have included separating Amazon's marketplace from its private-label business, separating Google's search engine from its advertising business, and requiring Meta to divest Instagram and WhatsApp. Structural remedies address the concentration of power directly but face legal challenges and implementation difficulties.
Behavioral Remedies
Behavioral regulation imposes rules on how platforms can operate without breaking them up:
- Interoperability requirements: Mandating that platforms allow their users to communicate with users on competing platforms (as the EU's DMA does for messaging services)
- Data portability: Requiring platforms to allow users to export their data in standard formats, reducing switching costs
- Non-discrimination requirements: Prohibiting platforms from favoring their own products or services over those of competitors
- Algorithmic transparency: Requiring platforms to disclose how their algorithms work, enabling external scrutiny and accountability
- Fee regulation: Limiting the commissions and fees platforms can charge their participants
Alternative Models
Public and cooperative alternatives to private platforms offer structural solutions to platform power:
- Public platforms: Government-operated or government-funded platforms that provide essential digital services (search, communication, marketplace) as public goods rather than private monopolies
- Cooperative platforms: Platforms owned and governed by their users or workers rather than by private shareholders. Examples include Stocksy (a cooperative stock photography platform) and Up & Go (a cooperative home services platform)
- Decentralized protocols: Protocols like ActivityPub (which powers Mastodon and the broader Fediverse), Matrix (decentralized messaging), and various blockchain-based systems attempt to create platform-like functionality without centralized platform control
These alternatives face significant challenges: public platforms face political and funding constraints; cooperative platforms face governance complexity and scaling difficulties; decentralized protocols face usability challenges and network effect barriers. But they represent important experiments in organizing digital markets without concentrating power in a few private companies.
Will Platform Power Persist?
Structural Persistence
Platform power is likely to persist because the mechanisms that create it--network effects, data accumulation, switching costs, and ecosystem control--are self-reinforcing. Each advantage strengthens the others: more users generate more data, which improves the product, which attracts more users, which increases switching costs, which makes the network effect stronger.
Historical precedent offers some hope for constraint. Railroad monopolies, telephone monopolies, and media conglomerates were all eventually subjected to regulatory frameworks that limited their power. But each of these regulatory responses took decades to develop and implement, and platforms are evolving faster than the regulatory processes designed to constrain them.
Emerging Pressures
Several pressures may constrain platform power in the coming years:
Regulatory momentum: The EU's DMA and DSA, the US antitrust cases, and growing regulatory attention globally are creating constraints that platforms have not previously faced. The cumulative effect of these regulatory actions may be significant even if individual measures are limited.
User awareness: Public awareness of platform power and its harms is growing. High-profile controversies--Cambridge Analytica, the Parler de-platforming, algorithmic amplification of harmful content--have shifted public opinion toward greater skepticism of platform power.
Technological alternatives: Decentralized technologies, federated protocols, and open standards offer technical alternatives to centralized platforms. While these alternatives are currently marginal, their existence demonstrates that platform power is a design choice rather than a technological inevitability.
Competitive dynamics: Platforms face competitive threats from each other and from new entrants. TikTok's rapid growth challenged Meta's dominance in social media. AI-powered search tools challenge Google's search dominance. These competitive dynamics may limit platform power more effectively than regulation in some cases.
The most likely outcome is not the elimination of platform power but its partial constraint through a combination of regulation, competition, and user action--similar to how earlier forms of monopoly power were constrained without being eliminated. The question is whether this constraint will be sufficient to prevent the most serious harms, or whether platform power will prove more durable and resistant to accountability than previous forms of concentrated economic power.
References and Further Reading
Srnicek, N. (2016). Platform Capitalism. Polity Press. https://www.politybooks.com/bookdetail?book_slug=platform-capitalism--9781509504862
Zuboff, S. (2019). The Age of Surveillance Capitalism. PublicAffairs. https://en.wikipedia.org/wiki/The_Age_of_Surveillance_Capitalism
Parker, G.G., Van Alstyne, M.W. & Choudary, S.P. (2016). Platform Revolution. W.W. Norton. https://en.wikipedia.org/wiki/Platform_Revolution
Khan, L.M. (2017). "Amazon's Antitrust Paradox." Yale Law Journal, 126(3), 710-805. https://www.yalelawjournal.org/note/amazons-antitrust-paradox
Crémer, J., de Montjoye, Y.A. & Schweitzer, H. (2019). "Competition Policy for the Digital Era." European Commission. https://ec.europa.eu/competition/publications/reports/kd0419345enn.pdf
Stigler Center. (2019). "Stigler Committee on Digital Platforms: Final Report." University of Chicago Booth School of Business. https://www.chicagobooth.edu/research/stigler/events/antitrust-competition-conference
Rahman, K.S. (2018). "The New Utilities: Private Power, Social Infrastructure, and the Revival of the Public Utility Concept." Cardozo Law Review, 39(5), 1621-1689. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2820370
Cusumano, M.A., Gawer, A. & Yoffie, D.B. (2019). The Business of Platforms. Harper Business. https://www.harpercollins.com/products/the-business-of-platforms
European Commission. (2022). "Digital Markets Act." https://digital-markets-act.ec.europa.eu/
Rochet, J.C. & Tirole, J. (2003). "Platform Competition in Two-Sided Markets." Journal of the European Economic Association, 1(4), 990-1029. https://doi.org/10.1162/154247603322493212
Wu, T. (2018). The Curse of Bigness: Antitrust in the New Gilded Age. Columbia Global Reports. https://en.wikipedia.org/wiki/Tim_Wu
Pasquale, F. (2015). The Black Box Society. Harvard University Press. https://www.hup.harvard.edu/catalog.php?isbn=9780674970847
Kenney, M. & Zysman, J. (2016). "The Rise of the Platform Economy." Issues in Science and Technology, 32(3). https://issues.org/rise-platform-economy-big-data-work/