Who works where · 1.7

Care work, paid and unpaid

Care work, paid and unpaid. What is actually the case, and how it compares with what is repeated.

For a practical software perspective on this topic, Monitask's practical guide provides a concise companion guide.

Care is the clearest place to see what the production boundary does, because the same task moves across it depending only on who performs it.

The same work, twice

Washing, dressing, feeding and supervising an elderly person is employment when a care worker does it and is nothing, statistically, when a daughter does it. The task is identical. One appears in the employment figures, generates income, attracts contributions and confers protections; the other appears nowhere.

As populations age, the volume of this task grows, and where it is performed determines whether the growth is visible at all.

The paid side

Social care employment is large, growing faster than employment generally in most high-income countries, and is characterised by low pay, high turnover, part-time and zero-hours contracts, and a workforce that is overwhelmingly female and in many countries disproportionately migrant.

Vacancy rates in the sector run well above the economy-wide average and have done for years, which is the ordinary signal of pay below the market-clearing level in an occupation people otherwise want to do.

The unpaid side

Much larger. Time-use surveys find unpaid care occupying a number of hours across the population comparable to a substantial fraction of total paid employment.

It is performed disproportionately by women, disproportionately by people in their fifties and sixties, and it reduces the carer's own paid hours, earnings and eventual pension. Those effects are well documented.

Why the boundary matters here specifically

Because policy moves work across it. Reducing publicly funded care hours does not reduce the amount of care required; it moves it to households, where it is unmeasured and unpaid.

That shows up in the accounts as a saving and in the employment figures as a fall, and shows up nowhere as an increase in work done. It is the cleanest example available of a measurement boundary shaping what a government can see.

Childcare, and the same structure

The identical pattern with a different demographic. Formal childcare is employment; a parent doing it is not. The cost of the formal option determines which side of the boundary the work falls on, and it does so differently at different income levels.

The entry on part-time work describes the consequence: a person working reduced hours because childcare costs more than the additional earnings is recorded as having chosen it.

What the sector's pay problem actually is

Care is labour, almost entirely, and productivity growth in it is inherently limited: an hour of attention cannot be compressed the way a manufacturing process can. Wages in the rest of the economy rise with general productivity, and a sector that cannot match those gains must either raise prices, raise subsidy, or fall behind on pay.

This is a long-standing and well-described problem in service industries, and care is its clearest instance. It means the sector's low pay is structural rather than a local failure, and that it will not be solved by exhortation.

What would show up if it were measured

Household satellite accounts, where they exist, value unpaid care alongside other household production. Very few countries publish them regularly and almost nobody uses them in the arguments where they would be relevant.

The data problem here is not that the measurement is impossible. It is that it is optional, and optional statistics are the first to be cut.

Why this entry closes this part

Because it contains every theme in it at once: a boundary that decides what counts, a category that bundles unlike arrangements, a gendered distribution, and a very large amount of work that no employment figure describes.

The migration dependency

In several countries a large share of the paid care workforce is foreign-born, and changes to immigration rules therefore act directly on the sector's staffing within months.

That is a policy connection rarely made explicitly when either policy is debated, and it is one of the places where two separate arguments turn out to be the same argument.

What a carer is not counted as

Somebody who has left paid work to care full-time for a relative is recorded as economically inactive, in the same category as somebody who has stopped working for any other reason.

Several statistical agencies now publish the reason for inactivity, which separates them. Where that breakdown exists it shows the caring group to be large and growing, and where it does not exist the group is invisible.

For broader context, consult ILO care-economy resources.