What economic growth statistics don't count, and why those things are mostly owned by women
By Kwame Asante
Category: Economics & Policy
Tags: unpaid care work, gdp measurement, care economy, feminist economics, gender inequality, economic policy, national accounting, global south, labor market, development economics
In 1941, a 23-year-old economist named Phyllis Deane arrived in Northern Rhodesia with a job: apply the new British system of national income accounting to a few colonial territories. She would spend months in village surveys, watching, counting, recording. What she found stopped the project cold — not because the data was bad, but because the framework she'd been handed had no place for most of what she was seeing.
The work that kept households alive — collecting firewood, transporting crops, grinding corn — was nearly all done by women. Deane argued it was "illogical" to exclude this labour, that the convention existed precisely because it was viewed as women's work. She pushed for it to be included. The system ignored her call, and because UN technical assistance programmes later exported Stone's method globally, activities central to everyday life in low-income African countries simply did not appear in national accounts.
That framework, with the blind spot intact, became GDP. We still use it today.
This is not an argument that economists are misogynists, or that the people who designed national accounting were acting in bad faith. It is an argument about what a measurement, once institutionalized, decides to see — and who ends up invisible as a result.
The numbers are not small.
The International Labour Organisation estimates that unpaid care and domestic work by country is valued at between 10 and 39 per cent of GDP. The OECD puts the figure at approximately 15 per cent on average across member countries when using the replacement cost method — rising to 27 percent when the opportunity costs of workers at home are accounted for — and notes that "women create the majority of this economic value."
That upper figure is worth sitting with. Twenty-seven percent. In the world's wealthiest economies — the ones that produce the growth statistics that get reported each quarter, the ones whose percentage-point fluctuations move markets and topple governments — there is an activity equivalent to more than a quarter of all that measured output happening in parallel, uncounted, and performed mainly by women.
An Oxfam analysis found that 65 percent of women's working hours are unpaid every week, excluded from official measures of economic activity. Women carry out the majority of unpaid care globally — nearly 90 billion hours per week.
The gender disparity inside this is stark: women perform roughly 76 percent of all unpaid care work, averaging four hours and 25 minutes per day, compared to men's one hour and 23 minutes.
Valued at an hourly minimum wage across 64 countries, the ILO estimates this unpaid care would constitute 9 percent of global GDP — around $11 trillion annually. The real figure, using market replacement rates rather than minimum wages, would be considerably higher.
Here is the part that gets philosophically uncomfortable.
GDP was not designed to measure welfare. Simon Kuznets, the economist most responsible for its development, warned as early as 1934 that the welfare of a nation could not and should not be inferred from GDP. What it measures is market activity — things that are bought and sold. It was a wartime tool for assessing production capacity, not a report card on how well a society is doing. That it became the latter is one of the stranger institutional transformations in modern history.
The System of National Accounts, which governs what goes into GDP across most countries, has expanded over time to include non-marketed goods — since 1993, subsistence agriculture can appear in national accounts. But it has continued to exclude unpaid care services within the household, justifying this on the grounds that the value was difficult to measure in the absence of market prices.
That justification has always been weaker than it sounds. We price all sorts of things without direct market transactions — imputed rent, for instance, is the estimated value of living in your own home rather than renting it, and it does appear in GDP. The difficulty of measurement is real but not decisive. The question that doesn't get asked often enough is: what else explains the exclusion?
I've read several of the economic methodology papers on this and I don't have a clean answer. The honest account seems to involve a mixture of genuine statistical difficulty, disciplinary inertia, and — less charitably but not implausibly — the fact that the people building these frameworks were not the people doing the work.
What the exclusion does, concretely, is this.
Governments allocate resources based on measured economic activity. Infrastructure spending, labor market policy, tax design, social transfers — all of it flows from a picture of the economy that is missing a chunk between 15 and 39 percent of its actual size, depending on how you count and where you are. An estimated 708 million women worldwide are outside the formal labor force because of unpaid care responsibilities. That figure doesn't register as a policy problem when the framework you're working from treats those 708 million people as economically inactive.
The miscount compounds. When care work is invisible in national accounting, it is also invisible in labor market statistics, in retirement systems, in health policy design. Women who spend decades doing work that sustains the entire edifice of market production end up with thinner pension entitlements, lower lifetime earnings, and more economic precarity in old age. The invisibility in the ledger translates to invisibility in outcomes.
There is also a macroeconomic feedback that tends to go unremarked. The OECD has noted that in some economies, unpaid care contributes more to economic functioning than the manufacturing, commerce, or transportation sectors. These are sectors with lobbying arms, trade associations, seat at the policy table, and ministers whose portfolios depend on their performance. Care has none of that, partly because it produces no revenue, partly because it isn't counted, and partly because the two conditions reinforce each other.
The argument for including unpaid work in economic measurement isn't new — it has been a contested issue in debates about national accounting for decades, with economists, statisticians, and feminist scholars pushing for change. The response has generally been: satellite accounts. These are supplementary measures, maintained alongside official GDP figures, that attempt to value household production separately. The UK, Canada, Australia, and others have them.
Satellite accounts are better than nothing. They are not the same as counting.
A satellite account is a document that a government can — and routinely does — ignore when making decisions about spending, taxation, and growth targets. It is a spreadsheet attached to the main spreadsheet, produced by statisticians, read by researchers, and consulted by policymakers approximately never when the political pressure is to hit a quarterly growth number. The information exists. It sits outside the number that matters.
India's second Time Use Survey, published in 2024, showed women spending 305 minutes daily on unpaid domestic work — nearly six times more than men's 56 minutes. India does not include this in its GDP. The survey exists; the accounting doesn't follow.
There's a version of this argument that is essentially technical: GDP is a flawed measure, let's build a better one. Add unpaid care to the production boundary. Design comprehensive time-use surveys. Reform the System of National Accounts at the UN level. This version is correct and useful and almost certainly insufficient.
The more uncomfortable version goes like this: the people with the most interest in changing the measurement are the people with the least power to do it. National accounting frameworks are set by international statistical bodies that are staffed by economists trained in traditions that excluded this work from the beginning. The exclusion is not actively enforced — nobody is arguing that care work shouldn't count. But institutional inertia doesn't need active enforcement. It just needs to persist.
The World Economic Forum has estimated that a $1.3 trillion investment in the care economy would generate a $3.1 trillion GDP return and create more than 10 million jobs in the United States alone. That is a more-than-two-to-one return. It doesn't happen, in part, because the input doesn't register in the denominator that drives investment decisions.
Phyllis Deane spent the rest of her career trying to find ways to count the things that Richard Stone's framework had decided not to. Even after the UN declined to include unwaged labor in official national accounts, Deane continued researching ways to quantify women's work — including firewood collection and crop transport — with an eye toward social reform, because she believed that in order to address distributional concerns, the economic contributions of all producers had to be counted.
She was right in 1941. The frameworks moved around her, and the blind spot stayed.
The question now is whether the people who use GDP figures — not just economists, but politicians, journalists, voters — understand that the number describes only part of what an economy is. The other part is being done right now, by somebody who will not appear in the quarterly growth data, and who probably knew before opening this article that this was the case.
SOURCES
Oxfam GB, "Almost two-thirds of women's working hours excluded from GDP globally" — press release for Radical Pathways Beyond GDP report. oxfam.org.uk. URL: https://www.oxfam.org.uk/media/press-releases/almost-two-thirds-of-womens-working-hours-excluded-from-gdp-globally/
Luke Messac, "Women's unpaid work must be included in GDP calculations: lessons from history," The Conversation, June 2018. URL: https://theconversation.com/womens-unpaid-work-must-be-included-in-gdp-calculations-lessons-from-history-98110
NIESR, "Phyllis Deane and the limits of national accounting," October 2023. URL: https://niesr.ac.uk/blog/phyllis-deane-and-limits-national-accounting
UNDP Latin America, "The Missing Piece: Valuing women's unrecognized contribution to the economy," March 2024. URL: https://www.undp.org/latin-america/blog/missing-piece-valuing-womens-unrecognized-contribution-economy
Ontario Pay Equity Office, "Time to Care: Recognising the truth behind the economy of unpaid care," 2023. URL: https://payequity.gov.on.ca/time-to-care-recognising-the-truth-behind-the-economy-of-unpaid-care/
Naila Kabeer, "A foundational asymmetry: gender, unpaid care work, and the market economy," Oxford Review of Economic Policy, March 2026. URL: https://academic.oup.com/oxrep/article/41/3-4/702/8502727
World Economic Forum, "Improving care economy is vital to growth and well-being," October 2024. URL: https://www.weforum.org/stories/2024/10/caring-care-economy-key-growth-and-well-being/
Economics.Town, "The Role of the Care Economy in the National Economy: Beyond GDP," November 2025. URL: https://economics.town/gender-issues/role-of-care-economy-beyond-gdp/
ClearIAS, "The Economic value of Unpaid Work," November 2024. URL: https://www.clearias.com/unpaid-work/