Marilyn Waring's 1988 If Women Counted posed a question so fundamental it embarrassed an entire discipline: why does the system that measures economic activity systematically exclude the work that sustains human life? The answer was structural, not conspiratorial. National income accounting was designed by men in the 1930s and 1940s to measure the market economy. Household production — cooking, cleaning, childcare, eldercare, emotional labor, domestic management — was excluded not because anyone argued it was valueless but because it was not transacted through markets. The omission was technical in origin. Its consequences were political. What the accounts did not measure, governments did not discuss. What governments did not discuss, budgets did not fund. And the work budgets did not fund was, disproportionately, work performed by women.
Time-use surveys across OECD countries consistently show household production representing between twenty and forty percent of total economic activity, depending on methodology and boundary definitions. None of it appears in headline GDP figures. Satellite accounts that estimate its value are produced irregularly, receive minimal policy attention, and are treated as supplementary information rather than core economic intelligence. The structural omission persists because the institutional infrastructure persists.
The AI transition transforms the household production gap from chronic measurement failure into acute crisis. When AI tools make productive work supernormally stimulating — when the flow state is available on demand, when the imagination-to-artifact ratio collapses — the opportunity cost of household production increases. Every hour spent cooking, attending to a child's homework, or maintaining emotional infrastructure is an hour not spent in the measurable flow of AI-augmented work.
The viral 'Help! My Husband is Addicted to Claude Code' Substack post is a domestic production crisis expressed as relationship complaint. When the husband withdraws from household responsibilities to build with Claude, two simultaneous transactions occur. His market-visible productivity increases. His household production decreases. The national accounts register the first and ignore the second. GDP sees a more productive worker. It does not see a depleted household.
The destruction is not hypothetical. Research on the relationship between household production and child development is extensive and largely unambiguous: the quality and quantity of parental engagement in early years is among the strongest predictors of long-term cognitive and emotional outcomes. The household production that AI displacement erodes includes the sustained, friction-rich engagement with children that developmental psychology identifies as foundational.
The critique begins with Waring's If Women Counted (1988) and develops through feminist economic scholarship — Julie Nelson, Nancy Folbre, Diane Elson. Coyle's engagement runs through her contributions to the Bean Review and her essays on national accounting reform. The contemporary literature on AI-driven household displacement is emerging in sociological and economic journals from 2024 onward.
Twenty to forty percent. Household production represents a substantial fraction of total economic activity across all OECD countries.
Gendered asymmetry. Women perform approximately twice as much unpaid household labor as men — an asymmetry the AI transition may intensify.
Displacement pressure. AI increases the attractiveness of market production without changing household technology, creating structural pressure for time reallocation.
Developmental consequences. Displaced parental engagement has predictable effects on child cognitive and emotional outcomes that measurement systems cannot detect.
Whether household production displacement from AI is substantial and gendered remains empirically uncertain. Some researchers argue that AI-augmented workers will use freed time to engage more intensively with families. Others argue that task seepage will intensify displacement. The question is resolvable only through longitudinal time-use data that most countries do not currently collect with sufficient frequency.