Technofeudalism

Technofeudalism is the argument that the dominant digital platforms of the twenty-first century do not represent an intensified form of capitalism but a structural departure from it — a reversion to something resembling feudalism, in which lords extract rent from economic activity occurring on their land rather than competing to produce better goods and services.

The term was most systematically developed by economist Yanis Varoufakis in his 2023 book Technofeudalism: What Killed Capitalism. Varoufakis argues that traditional capitalism operates through market competition: firms produce goods, compete for sales, and accumulate profit from that competition. The characteristic activity of the capitalist is production. What distinguishes the major digital platforms — Amazon, Apple, Google, Meta — is that their primary activity is not production but the ownership and administration of digital infrastructure through which others must transact. Amazon does not primarily sell goods; it owns the marketplace through which millions of sellers must access buyers, and charges rent for that access. Google does not primarily produce content; it owns the search infrastructure through which the internet is navigated, and charges for placement within it.

This is the feudal structure precisely: a landlord who owns the territory through which economic activity must pass, extracting a percentage of everything without competing in the activity itself. The serfs — users — are not merely customers. They are unpaid labourers whose activity (searching, posting, clicking, reviewing) produces the data that makes the platform's targeting valuable, and therefore makes the rent it can charge advertisers and merchants higher. Every hour you spend on a platform is an hour of free data-generation that increases the platform's leverage over those who pay to access you.

The feudal analogy extends to exit costs. A medieval serf could not practically leave their lord's land because all viable agricultural land was controlled by lords, and leaving meant starting from nothing in hostile territory. A small business today cannot practically leave Amazon or Google because those platforms control access to the majority of online commerce and search traffic. The merchant who exits does not enter a free market; they enter obscurity. The exit is theoretically free and practically impossible.

What makes this analysis more than rhetorical is what it predicts about policy. Standard antitrust remedies — breaking up companies, mandating interoperability, fining anti-competitive behaviour — were designed for capitalist markets where the problem is one company preventing other companies from competing. They do not address the feudal structure, which operates not through blocking competition but through owning the terrain on which competition occurs. You can fine a feudal lord for specific abuses; you cannot restore a free market by doing so, because the problem is the existence of the fiefdom, not the lord's particular behaviour within it.

The implication for individuals is that their relationship to platforms is not that of a consumer to a service, but something closer to a tenant to a landlord — with the added feature that the rent is paid in data rather than money, which obscures the transaction entirely. Recognising this structural position does not immediately change it, but it changes what questions to ask: not "is this platform good or bad?" but "what conditions would need to hold for a relationship this asymmetric to be in my interest?"