The Invisible Surplus

In the AI Story

Erik Brynjolfsson and his collaborators have estimated that consumer surplus from free digital goods in the United States alone may amount to hundreds of billions of dollars annually. None of it appears in GDP. The AI transition does not merely extend this problem — it transforms it. Consumer surplus from Google Search is passive: the user receives value by querying a system that already exists. The AI production surplus is active: the user creates value by directing a system that amplifies capability. The active surplus is larger per interaction, more heterogeneous in form, and even harder to measure.

Consider the scene Segal describes: a marketing manager building a custom tracking tool in an evening. She did not purchase the tool from a software vendor. She built it herself, using a tool that costs a hundred dollars a month. The value she created accrues to her organization, but it never generates the market transaction that the national accounts require. If she had purchased equivalent software from a SaaS provider at a thousand dollars per month, GDP would register twelve thousand dollars of annual activity. Because she built it herself, GDP registers approximately twelve hundred dollars and misses the remaining value entirely.

Coyle has framed the broader digital surplus problem by asking whether data is more like oil, air, fish, or wine — each analogy implying a different economics. Her conclusion is that data exhibits characteristics of all four, which means no single existing framework is adequate. The AI surplus problem inherits this complexity and adds a new dimension: the surplus is generated not merely by consuming but by producing.

The conceptual precedent is the imputation statistical offices already perform for owner-occupied housing. A homeowner consuming a housing service she also provides to herself has that service's rental value imputed into GDP. The same logic applies to AI-assisted personal production — but no statistical office currently imputes it, because the category barely existed before 2025.

Origin

Coyle's engagement with digital consumer surplus runs through her contributions to the Bean Review, her Daedalus essay on 'socializing data' (2022), and her collaborations with Erik Brynjolfsson on the GDP-B framework. Her October 2025 essay 'Measuring AI's Economic Impact' applied the framework directly to AI.

Key Ideas

Price captures zero. When services are free or nearly free, all value is consumer surplus and entirely invisible to GDP.

Active vs. passive surplus. Producing with AI generates larger, more heterogeneous surplus than consuming AI services passively.

The imputation precedent. Owner-occupied housing shows how statistical offices can count non-market value — the methodology exists but is not applied to AI.

Progressive distribution. Low-cost AI tools generate proportionally larger surplus for populations previously excluded from production — a benefit invisible to income-based inequality metrics.