The Care Economy
The hidden GDP. Who takes care of children, the elderly, and the sick - and what it costs when that work is not counted.
(ILO, 2018: 16.4 billion hours a day valued at the minimum wage, about 9% of world GDP)
(ILO, 2018; 3.2:1 worldwide and 4.7:1 in the Arab States)
(ILO projection made in 2018; 2.1 billion in 2015, most of them children under 15, with older people the fastest-growing group)
A note on framing. Care work - raising children, supporting elderly parents, attending to sick family members - is the largest category of human labor that is mostly invisible in economic statistics. It does not appear in GDP, it is not counted in labor force participation data, and it is overwhelmingly performed by women. This page describes how much care work exists, who does it, what happens when demographic change increases demand while social change reduces supply, and what different societies are trying. Sources include the International Labour Organization (ILO), the OECD Social Expenditure Database (SOCX), WHO, and World Bank gender data.
The scale of care work
The ILO estimated in 2018 that 16.4 billion hours of unpaid care work are performed globally every day. If valued at minimum wage, this would represent roughly $11 trillion per year - larger than the GDP of Japan and Germany combined. Women do about three-quarters of it. In high-income countries, women perform approximately twice as many hours of unpaid care work as men. In parts of South Asia, the Middle East, and North Africa, the ratio is far higher. This unequal division is a central explanation for a large portion of the gender pay gap, the pension gap, and the wealth gap across all societies for which data exists.
Paid care work - nursing assistants, home health aides, childcare workers, eldercare attendants - is among the lowest-paid work in virtually every economy, despite being essential. In the US, home health and personal care aides earned a median $17.21 an hour in 2025, yet the Bureau of Labor Statistics projects that no occupation will add more jobs by 2035. In England, the median care worker earned £12.60 an hour in December 2025, 39 pence above the minimum wage (Skills for Care). In Japan, care workers earned about a fifth less than the all-industry average in 2025, despite severe shortages. One explanation is that care long done for free by women within families is still priced against that unpaid alternative; another points to public buyers with fixed budgets and to work that is hard to make more productive.
The eldercare crisis
The demographic shifts described on the Demographics page - falling birth rates, rising life expectancy, aging populations - create a care demand curve that is highly predictable and already arriving. Japan, the most aged large country (29.6% of its population was 65 or older in September 2026), has run a public long-term care insurance system (kaigo hoken) since 2000. Its care workforce nonetheless shrank in 2023, and the health ministry expects to need 570,000 more care workers by 2040 than in 2022. The government has raised care pay repeatedly, most recently in June 2026, and widened visa routes, admitting foreign care workers to home-visit care from April 2025. South Korea started a national system of medical and long-term care at home in March 2026.
Germany's Pflegeversicherung (care insurance, introduced 1995) provides graduated benefits based on assessed care needs. Scandinavian countries fund some of the world's most comprehensive eldercare through high taxation, but face fiscal pressure as the ratio of taxpayers to elderly dependents declines. The US, by contrast, has no national long-term care insurance. A private nursing-home room cost a median $129,575 a year in 2025 (CareScout survey), and Medicaid covers nursing home costs only for those who have exhausted their assets, so middle-class families face catastrophic financial exposure if a member needs extended care. The 2025 budget law cuts projected federal Medicaid spending by $911 billion over ten years; supporters say it protects older and disabled people, while critics note that home care is optional for states, 15 of which planned new limits for fiscal 2026 (KFF). The pattern across all developed economies is the same: demand for eldercare is rising on a predictable curve, supply of caregivers is not keeping pace, and no society has solved the financing question sustainably.
Childcare economics
Childcare costs in many developed economies now rival housing as a share of family budgets. In the US, the average price of childcare was $13,184 in 2025, according to Child Care Aware of America, a network of referral agencies. That is more than the $11,950 average published tuition and fees at in-state public four-year universities. In England, working parents have been eligible since September 2025 for 30 funded hours a week from a child's ninth month. Families who do not qualify pay an average of £189 a week for part-time nursery care for a child under two (Coram, 2026). These costs fall disproportionately on women: when childcare costs approach or exceed one parent's take-home pay, it is overwhelmingly the mother who leaves the workforce, a pattern that compounds through reduced lifetime earnings, pension contributions, and career progression.
Some societies have made different choices. France's école maternelle system provides free, universal early childhood education from age three. Quebec's subsidized childcare program, introduced in 1997, charges parents C$9.65 per day (2026, indexed annually), with the government covering the remainder. Studies of the Quebec program show substantial increases in maternal labor force participation, though the effects on child outcomes are debated. Baker, Gruber, and Milligan found negative behavioral outcomes in the program's early years (2008) that persisted, with more crime as those children grew up (2015); other work finds the harm concentrated among children drawn into care by the policy. The Nordic model integrates parental leave, subsidized childcare, and flexible work arrangements into a system designed to keep both parents attached to the labor force through the early childhood years.
Care worker migration
One of the less visible global supply chains is the movement of care workers from lower-income to higher-income countries. The Philippines exports more nurses than any other country - Filipino nurses work in hospitals across the Gulf states, the UK, the US, Canada, and Australia. Indian nurses are a major workforce in the Gulf healthcare systems. Eastern European and Central Asian women provide much of the eldercare labor in Western Europe, often through informal or semi-formal arrangements. In the US, 30% of direct care workers in long-term care were immigrants in 2024 (KFF); stepped-up enforcement, a pause on immigrant visas for 75 countries and the end of Haitians' Temporary Protected Status all bear on that workforce.
This creates what sociologist Arlie Russell Hochschild called "global care chains." A woman from the Philippines cares for an elderly person in London, while her own children are cared for by a grandmother in Manila, who may in turn need care herself. Remittances from care workers are economically significant for sending countries, but the arrangement also drains sending countries of healthcare workers they need domestically. The World Health Organization has noted that low-income countries often face the paradox of training nurses who then emigrate to high-income countries that can afford to pay them more.
Singapore offers a distinct model. Its migrant domestic worker program allows households to hire live-in domestic workers, primarily from the Philippines, Indonesia, and Myanmar; no minimum wage applies. There were about 324,000 such workers in June 2026. The system provides affordable care but has faced criticism over worker protections, living conditions, and the dependency it creates.
Technology and care
Japan, facing one of the most acute care worker shortages, has invested heavily in care robotics. Assistive devices range from robotic lifting aids, which reduce back injuries among caregivers, to companion robots for isolated elderly people. The evidence on effectiveness is mixed. Physical assistance robots have measurable benefits for caregiver health, and Japanese nursing homes that adopted robots employed more care staff, not fewer, and kept them longer (Lee, Iizuka, and Eggleston, 2024). Companion robots show some positive effects on loneliness and agitation in dementia, but researchers caution that they do not substitute for human connection and raise questions about dignity. AI check-in calls, used by some Korean local governments, were followed by lower depression scores in a small uncontrolled study but often miss signs of distress.
Remote monitoring (wearables, smart home sensors, telemedicine) can help elderly people live independently for longer, deferring institutional care. But the technology raises surveillance and consent questions, particularly for people with cognitive decline. As the Technology page argues, who benefits, who is displaced and whose dignity is affected depend on institutional choices, not on the technology alone.
Where analysts disagree
Should unpaid care work be included in GDP? Feminist economists (Marilyn Waring, Nancy Folbre) have argued for decades that excluding unpaid care from national accounts systematically undervalues women's work and distorts policy. Others argue that including it would introduce measurement problems that would reduce GDP's usefulness as a comparable metric. Satellite accounts - separate estimates that complement rather than replace GDP - are the current compromise used by some national statistical agencies.
Is public or market provision better? The Nordic model (high taxes, public provision) produces high-quality care and high female labor force participation but requires sustained political consensus for high taxation. The market model (US, UK to some extent) keeps taxes lower and leaves more choice, but prices many families out. Mediterranean and East Asian models rely more heavily on family (especially female family) provision, which keeps public costs low but reduces women's labor force participation. Norway's 1970s childcare expansion barely raised mothers' employment, mostly replacing informal care, but improved the children's adult outcomes (Havnes and Mogstad, 2011). No model has solved quality, access and cost at once.
Can immigration solve the care gap? For destination countries, care worker immigration is a partial solution: in the US, more immigration has raised nursing-home staffing and improved residents' outcomes (Grabowski, Gruber, and McGarry, 2023). Britain, citing exploitation, closed its care worker visa to overseas recruitment in July 2025, and visas for care workers fell from 107,772 in 2023 to 344 in the year to June 2026. England's care vacancy rate kept falling, to 6.2% in 2025/26, though posts filled by British nationals also fell (Skills for Care). For sending countries, the brain drain in healthcare workers is a genuine problem. Whether bilateral agreements (the Philippines-Germany nursing corridor, for example) can balance sending-country and receiving-country interests is an active policy experiment.
Care is the hidden foundation of every economy. When it works - when children are raised, elderly parents are supported, sick people are tended - it is invisible, unpriced, and uncounted. When it fails - when childcare costs force parents out of the workforce, when elderly people die alone, when care workers cannot afford to live on their wages - the costs become visible everywhere else: in health systems, in labor markets, in family formation, in the demographic curves this site tracks.
The paths from here
Dated September 2026, with Britain closed to overseas recruitment of care workers and US states limiting Medicaid home care. Care demand is the most predictable variable on this site. The age structure that determines it is already fixed, and the only open questions are who does the work and who pays for it. Each path below is one society's answer, and most countries are already some blend of them.
The family absorbs it
Care stays unpaid and inside households, which in practice means it stays with women. Public costs remain low and female labor-force participation stays suppressed, along with the lifetime earnings, pension entitlements and career progression that follow from it. Southern Europe and much of East and South Asia sit closest to this pattern today.
Will it happen? It is the default wherever nothing else is built, and it is getting harder rather than easier. The families expected to absorb it are smaller, more geographically dispersed and more likely to have two earners than the families the arrangement assumed.
Public insurance for long-term care
Care is funded through a dedicated social-insurance contribution, as in Japan's kaigo hoken since 2000 and Germany's Pflegeversicherung since 1995. Costs are pooled and visible, entitlements are graded by assessed need, and the political argument is about the contribution rate rather than about who deserves help.
Will it happen? Where it exists it has proved durable, which is the strongest argument for it. Neither system has solved the workforce shortage, and both are under fiscal pressure as the ratio of contributors to claimants falls. Introducing one from scratch requires a political consensus that is easier to describe than to assemble; Washington State's version, worth up to $36,500 per person, began paying benefits in July 2026.
Imported labor fills the gap
Receiving countries recruit care workers from lower-income ones, formally through bilateral agreements or informally through migration. The care is delivered and the wage bill stays affordable. The sending country loses trained health workers it also needs, and the global care chain described above extends by another link.
Will it happen? It is already the main answer in the Gulf, much of Western Europe and Singapore, and Japan keeps widening its routes. The constraint is political in receiving countries and ethical in sending ones, and the political one now binds in places: Britain ended overseas recruitment of care workers in 2025, and South Korea scrapped its foreign caregiver pilot.
Technology extends independence
Remote monitoring, assistive robotics and better home design let more people live independently for longer, deferring institutional care rather than replacing it. Physical-assistance devices reduce injuries among caregivers, which addresses part of the workforce problem from the other end.
Will it happen? Partly, and the evidence is uneven. Lifting aids have measurable benefits. Companion robots show modest effects on loneliness and raise questions about dignity that the research does not settle. Nothing in the current pipeline substitutes for the human hours; it shifts when they are needed.
The gap simply widens
Demand rises, none of the above scales fast enough, and the shortfall shows up as unmet need. Elderly people go without adequate support, families are under strain, care workers leave a sector that does not pay enough to live on, and quality falls in the institutions that remain.
Will it happen? Some of it is already visible in most developed countries - more than 600,000 people were on US Medicaid home-care waiting lists in 2025, some of whom may not qualify - and it is the path that requires no decision by anyone. That is precisely what makes it the most likely component of whatever mixture emerges.
No country has solved quality, access and cost simultaneously, and the arithmetic suggests none will. The realistic question is which of the three each society chooses to compromise, and whether that choice is made deliberately or by default.
What this means for you
If you have aging parents
The costs and the arrangements are worth understanding before they are needed rather than during a crisis. The decisions made in the first two weeks after a fall or a diagnosis tend to set the pattern for years. What a country's system covers, what it means-tests, and how long the waiting lists run are knowable in advance and rarely looked up in advance.
If you are deciding about childcare and work
When childcare costs approach one parent's take-home pay, the arithmetic appears to favor that parent leaving the workforce. The calculation that matters is not this year's but the lifetime one, including lost pension contributions, forgone promotions and the earnings penalty on re-entry, which together can outweigh the fees being avoided.
If you are doing unpaid care now
It is economically invisible by construction, not by accident: national accounts were built to measure market transactions and this is not one. That has practical consequences worth knowing about - pension credits for caregivers, respite entitlements and caregiver allowances exist in many countries and are consistently under-claimed, largely because nothing in the system announces them.
If you are reading policy arguments about it
Proposals that promise better care, wider access and lower cost at once are not describing a trade-off, which means one of the three is being quietly dropped. The useful question about any care policy is which of the three it sacrifices and who bears that sacrifice, since in this sector the answer is usually women, migrants, or both.
The mechanics behind this
The care economy is where several of this site's underlying mechanisms meet. If the argument above rests on ideas you want built up from the beginning, start here.
