Platform Capitalism

Platform capitalism is the analytical framework Nick Srnicek advanced in his 2016 book Platform Capitalism to describe what he saw as a structural transformation in how capital accumulates in the digital economy. The central argument is that the platform — a digital infrastructure that mediates interactions between two or more participant groups — has become the defining business model of the era, and that its logic of accumulation is distinct from prior industrial capitalism in ways that have significant consequences for markets, labour, and regulation.

The key insight is that platforms do not primarily produce goods or services in the conventional sense. They produce infrastructure and sell access to it, while extracting value from the data generated by the interactions they facilitate. A ride-sharing platform does not manufacture vehicles or employ drivers in the traditional sense; it operates the matching and pricing infrastructure through which drivers and riders find each other, and extracts a rent from every transaction. An advertising platform does not produce content; it operates the attention infrastructure through which publishers and advertisers are matched, and extracts a rent from every impression. The common structure is intermediation plus observation plus monetisation of the resulting intelligence.

Data, in Srnicek's framework, is the new raw material of capitalism — not in a loose metaphorical sense, but in the precise sense that industrial capitalism required physical raw materials that were extracted, processed, and sold. Data is extracted from user interactions at minimal marginal cost, processed through machine learning systems into prediction products and operational intelligence, and sold or used to extract further rents. The platform that accumulates more data processes it into better predictions and more efficient matching, which attracts more participants, which generates more data. This is the compounding dynamic that makes platform incumbency so durable.

The structural tendency toward monopoly is therefore not incidental to platform capitalism but intrinsic to its logic. Network effects — the increase in a platform's value as more participants join — are familiar from earlier economic analyses. What Srnicek adds is the observation that data advantages compound network effects: a larger platform does not just have more users but has seen more transactions and therefore has better predictive and matching capabilities, which attracts still more users. The incumbent's advantage grows with scale in ways that conventional antitrust frameworks, designed for industrial markets, were not built to address.

The labour implications of this model have become increasingly visible. By classifying workers as independent contractors rather than employees, platform firms externalise the costs — benefits, downtime, equipment — that industrial employers internalised, while retaining algorithmic control over the conditions of work. The platform owns the infrastructure and the customer relationship; the worker owns the capital and the risk. This structure allows platforms to scale rapidly without corresponding growth in fixed labour costs, and it produces a form of work that is formally autonomous but practically governed by the platform's rating systems, pricing algorithms, and access controls.