Coasean Singularity

In the AI Story

Coase himself rejected the zero-transaction-cost assumption embedded in what became known as the Coase Theorem. "I never liked the Coase Theorem," he told Russ Roberts. "It's a proposition about a system in which there were no transaction costs. It's a system which couldn't exist." The frictionless world where parties negotiate costlessly to efficient outcomes is a theoretical construct useful for isolating the effect of property-rights assignment, not a description of any real economy. The Coasean Singularity commits the same error at organizational scale: treating the limiting case as an achievable endpoint rather than an analytical tool.

The concept's appeal lies in its apparent fit with observable trends. AI does reduce search costs (finding information, matching capabilities to needs), bargaining costs (natural language specifications replacing formal contracts), and monitoring costs (algorithmic oversight of distributed work). Platforms demonstrate that coordination at massive scale is possible without traditional organizational hierarchy — Uber coordinates millions of drivers, Airbnb coordinates millions of hosts, GitHub coordinates millions of developers, all through algorithmic infrastructure rather than managerial direction. Extrapolating these trends produces the Singularity vision: every worker an independent agent, every transaction algorithmically matched, every firm dissolved into a platform-mediated market.

Samuel Hammond's warning about power concentration exposes the Singularity's political flaw. Near-zero monitoring costs favor whoever does the monitoring. In practice, that tends to be the entity with the most data, the most compute, the most comprehensive surveillance infrastructure — not individuals or small firms but large platforms. The Coasean boundary might not contract the firm but expand it, because the platform that monitors cheaply can absorb activities that smaller entities perform less efficiently. The result would be not atomized individuals but concentrated platform power — a different organizational future, still Coasian in logic, producing centralization rather than the decentralization the Singularity narrative assumes.

Origin

The term appeared in Shahidi, Rusak, Manning, Fradkin, and Horton's 2025 NBER chapter on AI and the future of work, drawing on earlier speculation by Tyler Cowen and others about AI's effect on organizational boundaries. The concept synthesized Coasian transaction-cost logic with AI capability projections, producing a limiting-case scenario useful for identifying which transaction costs were genuinely collapsing and which remained stubbornly high. The Singularity framing was deliberately provocative — designed to force clear thinking about what firms actually do and what would happen if those functions became market-transactable.

Key Ideas

The unattainable limit. Zero transaction costs are as unreachable as frictionless planes in physics — useful for isolating variables, impossible in reality where information asymmetry, uncertainty, and social complexity persist.

Markets cannot provide trust. The social functions firms perform — building relationships, transmitting tacit knowledge, maintaining professional standards, providing belonging — are not reducible to transaction costs and do not disappear when transactions become cheap.

Power concentration as alternative trajectory. If monitoring costs approach zero, the Coasian logic may favor large platforms that can monitor cheaply rather than small individuals — producing centralization, not atomization.

The Singularity as diagnostic tool. The concept's value lies in identifying which organizational functions are transaction-cost-justified (threatened by AI) versus which are social-function-justified (persist after AI).