Few economic puzzles are as plain as the pay attached to care work. The labour of feeding small children, washing an elderly parent, sitting through the night beside a hospital bed, or coaxing a frightened patient through a procedure is among the most necessary work any society performs, and among the worst rewarded. A reader who has arrived here from a search about carers’ wages, or about the unpaid hours absorbed by a household, has arrived at the same puzzle from opposite ends, because this article treats the two as one subject. The paid nursery worker and the unpaid daughter are doing continuous work under different arrangements, and the analysis that follows treats them as connected parts of a single system rather than as a market topic and a private topic that happen to share a name.

That connection is the first thing to establish, before any theory. Paid employment in nurseries, care homes, hospitals, and private houses does not replace the unpaid labour of households; the two expand and contract against each other, and work pushed out of one arrangement reappears in the other. When a public service is withdrawn, the feeding and washing and watching do not vanish. They move home. When a household can no longer absorb them, they are bought, if they can be afforded, or they are left undone, at a cost that shows up later in a surgery or a school report. The central question of this article is why the paid version of this indispensable labour is paid so badly, and the answer it develops is structural rather than moral. The framework within which the question is posed, the theory of how households reproduce the labour force that production consumes, is set out in the series article on social reproduction theory explained, and is linked rather than rebuilt here.

Why care work is paid so little: the puzzle in full

Begin with the shape of the fact to be explained. Occupations centred on the direct maintenance of people cluster toward the bottom of the earnings distribution across very different economies, under governments of different complexions, in sectors organised for profit and in sectors organised as public services. The clustering survives changes in who employs the workers and changes in how the service is funded. It survives, too, the evident skill involved: anyone who has watched an experienced dementia carer defuse a violent confusion with nothing but timing and tone knows that the work demands judgement, memory, physical competence, and emotional control of a high order. An explanation of the pay pattern has to explain why a labour market that rewards scarce skill elsewhere fails to reward it here, and why the failure is so stable across institutional settings that otherwise differ enormously.

Two families of explanation compete for the pattern. The first is attitudinal: the work is paid badly because it is held in low regard, and it is held in low regard because it has historically been done by women, unpaid, inside families, so that its performance in a market is read as the exercise of a natural disposition rather than the application of a trainable skill. The second family is structural: the work is paid badly because of its cost structure, because the activity at its centre resists the productivity growth that elsewhere allows wages to rise without prices rising in step, so that every employer of carers, public or private, faces a standing pressure that employers in most other sectors do not. This article gives the structural account the leading role, not because the attitudinal account is false, but because the structural account explains more of the pattern, including parts of it the attitudinal account cannot reach, and because it predicts what happens when the organisation of the sector changes, as a good explanation should.

Why is care work so low-paid?

partly because it resists productivity growth. Providing care faster generally means providing it worse, so care employers face rising relative costs without matching productivity gains. That structural pressure, combined with the association of care with unpaid family labour, sustains low pay across the sector.

Hold the two halves of that answer apart, because they do different work. The association with unpaid family labour explains why the labour is supplied cheaply at the point of hiring: a large reserve of people, overwhelmingly women, have already been trained in the work by doing it without wages, and an employer can recruit that trained competence at a wage that treats it as untrained. The productivity half explains why the cheapness persists and deepens even where the association weakens, even in services staffed by credentialled professionals, even under employers who would pay more if the arithmetic allowed it. A hospital or a care home competes for staff against sectors whose productivity, and therefore whose wage offers, keep climbing. Its own output per hour barely climbs at all, because the hour is the output. The gap has to be closed somewhere: in the wage bill, in the staffing ratio, in the price charged, or in the public budget. Watching where each kind of employer closes it is the empirical thread that runs through everything below.

The pattern also has a distributional wrinkle that the attitudinal account handles badly. Within the sector, pay does not track the intimacy or difficulty of the work in any simple way. Some of the most demanding relational work, the long, slow attendance on people who cannot reward it with gratitude or even recognition, sits near the very bottom of the scale, while better-paid roles in the same building are often those furthest from the bedside. A theory of mere social disdain would expect the gradient to follow visibility and prestige more closely than it does. A cost-structure theory expects exactly the observed mess: pay settles where the funding stream, the staffing arithmetic, and the available labour supply intersect, and none of those three cares much about the moral worth of the task.

What counts as care work: a definition and its borders

Definitions decide what an explanation has to cover, so the borders of the subject matter. In the tradition this article draws on, care work comprises the paid and unpaid labour of maintaining, sustaining, and developing people: childcare, the care of elderly and disabled people, much health work, and the domestic maintenance, cooking, cleaning, laundering, fetching, that keeps a household’s members fed, rested, clothed, and able to appear at their other obligations. The definition is deliberately wider than the occupational category that statistics offices count, because the analysis depends on seeing the unpaid and paid versions as substitutes along a margin. A meal cooked by a parent and a meal cooked by a nursery cook are the same activity under two arrangements; a theory that counted only the second would mistake a change of arrangement for a change in the quantity of work the society performs.

What separates care from other kinds of service work?

Three features mark the border. The recipient is a person whose condition is itself the output, so quality lives in the recipient’s state rather than in a separable product. The work is relational, depending on knowledge of this particular person. And much of it cannot be stored, postponed, or moved without loss.

Each feature carries an analytical consequence. Because the output is a person’s condition, the usual test of efficiency, more output per hour, has no stable meaning: an hour in which a confused patient is kept calm produces nothing countable, yet letting the hour go unstaffed produces countable harm. Because the work is relational, replacing a known carer with a stranger is not a neutral substitution of equivalent labour, as it might be on an assembly line; the replacement destroys part of the product, the accumulated knowledge of how this person eats, sleeps, frightens, and settles. Because the work cannot be stored, demand arrives on its own schedule. A child’s fever, a fall in the night, a confusion at dawn: the labour must be supplied at the moment of need or not at all, which is why the sector’s staffing problem is a problem of coverage across unsocial hours and not merely a problem of totals.

Border cases test any definition, and two deserve stating. The first is emotional labour, the management of one’s own displayed feeling as part of the job, which is a component of most care work but also appears in sales and service roles far from any bedside; its own analysis belongs to the series article on alienation at work today, and it is treated here only as an ingredient. The second is the border between care and education. Teaching small children is plainly both, and the pay pattern this article examines follows teachers of the youngest children with depressing fidelity, which is itself evidence that the pattern attaches to the caring content of a job rather than to its formal classification. Where a role’s caring content rises, its pay tends to sag toward the care pattern, whatever the letterhead says.

The productivity trap: why caring cannot be speeded up

The structural explanation at the centre of this article has a name in this series: the productivity trap. Care work is structurally resistant to the productivity gains that raise wages elsewhere, because doing it faster generally means doing it worse. The claim sounds like a slogan until its mechanism is laid out, and the mechanism comes not from Marxist theory but from the mainstream economics literature on unbalanced growth, where it was developed in the nineteen sixties by the economist William J. Baumol, working first with William G. Bowen on the economics of the performing arts and then in his own analysis of the cost structure of personal services. Their example was a string quartet: a performance of a Haydn quartet requires the same labour time from four musicians now as it did in the eighteenth century, because the performance is the product, and halving the labour produces not a cheaper quartet but a different and worse one, or none.

Transpose the quartet to the ward and the nursery and the point sharpens rather than blurs. A carer who bathes an elderly man in half the time has not produced the same bath more efficiently; she has produced a hurried, frightening, and possibly unsafe bath. A nursery worker responsible for twice as many toddlers is not twice as productive; the attention each child receives, which is the substance of the service, has been halved. A nurse who shortens every conversation with a patient may process more patients, and the metric will record an improvement, but the unrecorded content of the encounters, the noticing of the symptom mentioned in passing, the calming that prevents the crisis at midnight, has been thinned. In manufacturing, productivity growth means the same object made with less labour. In care, the labour is so nearly identical with the object that economising on it degrades the object itself. The input and the output are the same minutes of attended human presence.

What did the cost disease argument actually claim?

In Baumol’s formulation, an economy divides into a progressive sector where output per worker compounds, and a stagnant sector where it barely moves. Wages must roughly equalise across the two, so the stagnant sector’s relative costs rise without limit, and its prices, budgets, or wages absorb the strain.

Follow the arithmetic through and the wage pattern stops being mysterious. Across the economy, pay rises over the long run because productivity rises: an employer whose workers produce more each year can pay more each year and hold unit costs steady. The care employer faces the same labour market. Carers must be paid enough to keep them from leaving for the warehouses, offices, and delivery rounds whose wages are drifting upward on the back of genuine productivity gains. But the care employer’s own revenue per worker cannot drift upward in the same way, because the output per worker is fixed by the nature of the task. Something must give. If the service is sold, its price rises relative to the price of goods, which is why families experience childcare and eldercare as ruinously expensive at the very moment carers experience the wages as miserably low. Both experiences are accurate, and their conjunction is the signature of the trap: the sector is simultaneously too expensive for its purchasers and too cheap for its workers, because the missing productivity growth is the fund out of which both lower prices and higher wages would otherwise be paid.

If the service is publicly funded, the same pressure appears as a budget problem. The finance ministry watches the cost of a fixed standard of provision climb year after year as a share of national income, not because anyone is wasteful but because the standard is denominated in hours of human attention while everything else the state buys gets cheaper per unit. The political system then faces a menu of unpalatable options: let the budget share rise, dilute the standard by raising the number of patients or children per worker, freeze the wages and accept the staffing shortages that follow, or withdraw the service and let households absorb the work unpaid. Every wealthy economy can be found somewhere on that menu, and the argument of this article is that the menu itself, rather than the moral character of the choosers, is what needs explaining.

Notice what the trap does to the usual remedy proposed for low pay. In most sectors, the durable route to higher wages is to raise the value of what a worker produces per hour, by training, by equipment, by reorganisation. Applied to care, that advice mostly misfires. Equipment helps at the margins: hoists spare backs, monitors spare some watching, records systems spare some duplicated writing. Training deepens the quality of the hour without shortening it. But the core hour, the attended presence, does not compress. A policy that waits for productivity to rescue carers’ wages is waiting for the one thing the activity structurally refuses to supply, and a workforce told to wait for it is being told, in effect, to accept the wage pattern as permanent while the explanation of its permanence is mistaken for a law of nature rather than named as a feature of the cost structure that policy could choose to finance openly.

The other half: labour trained for free and treated as unskilled

The productivity trap explains the pressure on wages. It does not by itself explain why the pressure is so readily discharged onto wages rather than resisted, and for that the second half of the snippet answer is needed: the association of caring with unpaid family labour. Employers in this sector recruit from a population that has already acquired much of the competence the job requires, acquired it without wages, inside households, as part of being raised and raising others. Feeding, soothing, cleaning, nursing the sick through the night, managing the moods of a household: these are skills, learned over years, and the labour market receives them as if they were endowments of temperament rather than products of training. A competence that arrives without a certificate and without a training bill is priced as though it cost nothing to produce.

That pricing is the precise point at which the analysis of social reproduction bites on the wage question. The household performs, unpaid, the work of producing and maintaining the labour power that every employer hires, and it also performs, unpaid, the training of the next generation of carers. Capital as a whole depends on this subsidy and does not pay for it. The older question of whether household labour produces value, and how it should be theorised, belongs to the series article on the domestic labour debate, and is deliberately not reopened here. What matters for pay is narrower: because the skill is reproduced outside the wage relation, the wage relation can pretend the skill is not there.

Why does the association with family labour hold wages down?

Because the employer’s comparison is rigged at the source. Labour that a household supplies without charge sets the shadow price for the same labour in the market, and hiring managers read the competence women bring as disposition rather than training, so the wage offered prices the disposition, which is assumed to be free.

The reading of skill as disposition has a long institutional tail. Job evaluation schemes, the bureaucratic instruments that assign pay grades, historically scored the visible, measurable, and mechanical components of jobs more reliably than the relational ones, so that responsibility for money or machinery outweighed responsibility for people, and the strength to lift a box outweighed the judgement to lift a person. Where such schemes have been challenged under equal-value principles, the comparison exercises have repeatedly surfaced the same asymmetry: the female-dominated caring roles had been graded as though their central demands were minor because those demands resembled what the graders assumed women simply do. The point for this article is not the litigation but the mechanism. The undervaluation is not primarily a sneer. It is an accounting convention, inherited from the household’s invisibility, that records the product of years of unpaid training as a natural fact about the worker rather than a cost somebody bore.

Gender enters here as the association’s carrier, not as a free-standing explanation. Men who enter caring occupations encounter the same wage pattern, slightly softened by a tendency to be promoted out of the bedside roles faster, which the literature on the sector describes as a glass escalator carrying male carers toward supervision. The existence of that escalator confirms rather than refutes the structural reading: the penalty attaches to the work and to its feminine coding together, and escaping the penalty by leaving the bedside is precisely what the cost structure predicts for any worker the institution wishes to retain. Meanwhile the coding does its recruiting work in the other direction. Because the work is represented as a vocation, as something done for love, asking for money can be made to look like a failure of devotion, and the workforce’s own sense of obligation to identifiable people in its charge becomes a discipline that no employer has to purchase. A strike in a factory idles machines. A strike in a care home abandons named individuals, and everyone at the bargaining table knows it.

The penalty, the measurement problem, and why the figures must stay directional

Researchers who compare earnings across occupations while holding measured skill requirements constant find a residue they call the care penalty: work involving face-to-face caring for people pays less than work of comparable measured demand that does not. The finding, developed in the sociological literature by Paula England, Michelle Budig, and Nancy Folbre among others, matters here because it separates the penalty from the composition of the workforce. It is not simply that the sector employs people with fewer credentials. The penalty attaches to the caring content of the job itself, persisting across comparisons that strip out education, experience, and hours, and it is one reason the purely attitudinal account and the purely structural account both capture something real: the coding of the work depresses its price over and above what the cost structure alone would predict.

Alongside the occupational penalty sits the motherhood penalty, the earnings reduction associated with bearing and raising children, documented in the research of Budig and England and in a large subsequent literature. Its relevance to an article on paid employment is that the two penalties compound through a single biography. The years in which a worker’s unpaid caring is heaviest are the years in which her paid caring, if she returns to the sector where her experience lies, is discounted for the interruption, so that the household’s subsidy to the economy is collected twice: once as unpaid labour given, and once as paid labour cheapened by the giving. The framework’s name for the whole circuit is social reproduction, and the wage pattern is one of the places the circuit becomes visible as a payslip.

No figures appear in this article for the size of either penalty, for sector wages, for the hours of unpaid work performed in any economy, or for the shares of women and migrants in the workforce. That abstinence is a rule of the series and it is also a matter of intellectual honesty. Unpaid work is measured by time-use surveys, in which respondents keep diaries of their days, and the method has known limitations that anyone citing its outputs should carry in view. Diaries under-record supervisory care, the being-available that structures a parent’s or a daughter’s day without filling it with discrete tasks. They struggle with simultaneity, the meal cooked while a child is helped with reading and a parent’s medication is supervised, which a diary forces into a primary and a secondary activity. They convert the fragmented, interrupt-driven texture of caring into tidy episodes. Estimates built on such foundations move substantially with the conventions chosen, and the honest statement of the pattern is directional: the unpaid economy is very large relative to the paid one, the paid care sector sits low in the earnings order, and the penalties described above recur across methods and countries without depending on any single contested number.

How is unpaid care measured, and why is the measurement contested?

By time diary. Respondents record a day in episodes, and statisticians aggregate the episodes into estimates. The method handles discrete tasks well and availability badly: the parent who sleeps lightly beside a sick child logs sleep, not vigilance, so the surveys systematically understate the caring that consists of being there.

The understatement is not a technical footnote; it bends policy. What the statistics register as the care burden is the visible fraction of it, the baths given and meals served, while the standby fraction, the foregone sleep, the abandoned plan, the job refused because its hours could not bend around a school run, escapes the ledger. Analysts who work with these surveys know the distortion and say so in their methodological appendices, but the numbers escape into debate without their caveats attached, and the sector’s true weight in the economy is then argued about at the level of the visible fraction. A structural account has to insist on the invisible fraction’s existence without pretending to know its size, which is an uncomfortable position in a debate that rewards false precision, and the correct one.

Skilled work that is not allowed to look skilled

Stand in a good nursery at the hour when parents arrive and watch what the senior worker is actually doing. She is tracking a dozen developmental trajectories, noticing which child has gone quiet in a way that means illness and which in a way that means concentration, calibrating a sentence to a parent’s anxiety, settling a dispute between two four-year-olds at exactly the level of intervention that leaves both with their dignity, and holding in memory the medication, the allergy, the custody arrangement, the new baby at home. Presented with this inventory, most observers concede the skill. Presented with the pay grade attached to it, they discover reasons why it does not count: it is not technical, it is not certified, it is natural to her.

The move from skill to nature is the ideological hinge of the whole wage pattern, and it is worth pausing on the word natural. A skill is called natural when its acquisition is invisible to the person pricing it. Nobody calls a surgeon’s hands natural, because the training that produced them passed through institutions the pricer can see, with fees and examinations and a register. The carer’s training passed through a childhood spent minding younger siblings, through years of household apprenticeship, through the accumulated corrections of doing the work under the eye of someone who had done it longer. Because no institution invoiced for it, the market treats it as a property of the person rather than an investment in the person, and pays accordingly. Feminist economists, Nancy Folbre prominent among them, have described the carer’s motivation as a kind of asset the employer draws on without paying for its upkeep: affection and obligation make the worker reluctant to withdraw effort when the wage fails to cover its cost, so the wage can fail, for longer, than it could in an occupation where effort is given strictly in exchange. Folbre’s phrase for the resulting position, the prisoner of love, names the trap from inside the worker’s own commitments.

What separates low pay from low status in this sector?

Pay and status are related but distinct injuries, and a student should keep them apart. Status is the regard in which the work is held, pay is the price its labour commands. The structural argument explains the price; the regard follows the price as much as it causes it, since occupations that pay badly are presumed to deserve to.

Conflating the two injuries produces a characteristic policy error: the campaign that seeks applause for carers as a substitute for money. Public rituals of appreciation, medals, clapping, the rhetoric of heroes, address the status injury while leaving the price untouched, and they can even deepen the trap by replenishing the stock of devotion on which the low wage relies. The workforce notices the substitution quickly. A carer who is told she is priceless is being told, in the same breath, that no price will be computed. The analysis here implies a sterner kindness: the respect that matters is the kind denominated in the staffing ratio and the hourly rate, because those are the variables through which the cost structure actually touches her working day.

The household: the site where the system hides its subsidy

Every analysis of paid care eventually has to walk back into the kitchen, because the household is the sector’s silent partner and its competitor of last resort. It is in households that the labour force itself is produced and repaired: the worker who arrives at the factory gate or the office door fed, rested, clothed, and emotionally reassembled was made that way by work that no wage paid for. Marxist feminism’s founding insistence was that this work is not a private backdrop to the economy but a load-bearing part of it, a condition of production that capital depends upon while systematically declining to count. The point is developed as theory in the cluster’s framework article on social reproduction theory, whose argument this application presupposes.

Consider what the household form does to the wage bargain in the paid sector. A care employer hiring staff is hiring people whose availability is itself rationed by their own unpaid obligations. The applicant who can work nights is the one whose household contains another adult, or grown children, or no dependants; the applicant who cannot is not less skilled, only more encumbered. Shift patterns, part-time contracts, and the sector’s notorious use of fragmented hours all grow out of this collision between two systems of caring that both want the same women’s time. The employer adapts to the encumbrance by designing jobs around it, short shifts at the hours when schools release their charges to other arrangements, and then cites the fragmented job as evidence that the work is marginal and casual. The causality runs from the household’s demands through the job’s design to the wage, and reversing it in the telling is one of the standard misreadings this series exists to correct.

Households also differ, and the analysis must register the differences or become a caricature. The single-adult household cannot redistribute caring internally; it buys what it can, leans on kin and neighbours, or goes without. The household with two earners and one wage high enough to purchase help experiences care as a logistics problem. The affluent household that employs a cleaner, a nanny, and a carer for a parent has not escaped the system; it has moved its own caring up the chain and become an employer at the chain’s upper end, a position examined in the section on care chains below. Class runs through the household form as surely as gender does, and the Marxist feminist tradition at its best has always held the two together, which is part of what distinguishes it from a feminism of the professional stratum alone.

The care pressure table: four sites, one pressure

The argument so far can be gathered into a single findable artifact. The care pressure table sets out the four sites at which care is organised, who performs the work at each, who bears the cost, what happens when public finances tighten, and what remedies are proposed from each site. Read down any column and the pattern of this article appears: the pressure generated by the productivity trap does not disappear when it is moved; it changes hands.

Site	Who performs the work	Who bears the cost	What happens under fiscal pressure	Proposed remedies
The household	Family members, overwhelmingly women, unpaid	The carer’s earnings, health, and time	Absorbs whatever the other sites shed; the unpaid day lengthens	Recognition of unpaid work, shorter paid working time, sharing of the domestic load
Publicly provided services	Salaried staff in nurseries, hospitals, and domiciliary services	The taxpayer, through general revenue	Staffing ratios diluted, eligibility narrowed, wages frozen	Expansion of public provision, universal basic services, professional pay scales
The market for care	Employees of agencies and homes, and directly hired workers	The purchasing household or the insurer	Fees pressed down onto wages; turnover rises; corners cut in staffing	Wage floors, regulation of ratios, limits on extraction by owners
Migrant care chains	Migrant women working in wealthier households and institutions	The worker’s own family, cared for by others further along the chain	Demand rises as other sites shed labour; the chain lengthens	Portable rights, regularisation, support for the families left behind

A table is a claim about structure, and this one claims that the four rows are communicating vessels. Squeeze the second row and the first and fourth swell. Squeeze the third with fee regulation unaccompanied by funding and the work leaks into the informal market, where it is done worse paid and uninspected. The fourth row exists because the first three, in wealthy economies, cannot between them supply the volume of caring the population’s age structure demands at the wages the cost structure permits. Any remedy addressed to a single row while ignoring the vessels’ connections will move the pressure rather than relieve it, and the history of care policy is largely a history of such moves, celebrated as reforms at the row they relieved and invisible at the row they burdened.

What sequence of choices does the table force on an employer?

Read the table as a decision sequence. An employer whose revenue cannot rise with productivity must choose which variable absorbs the gap: the wage, the staffing ratio, the fee, or the public subsidy demanded. The rows show where each choice lands the cost, and the columns show that the cost always lands on somebody’s caring.

Notice, too, what the table leaves out deliberately: any row for technology. Not because devices play no part, but because their part is confined to the edges of each site: the hoist in the care home, the monitor in the nursery, the rota software at the agency, the video call that lets a daughter check on a father from another city. Each device redistributes attention or spares a back. None of them manufactures the attended hour itself, and a table organised by sites rather than gadgets keeps that hierarchy visible. The temptation to solve the care problem by innovation is the temptation to forget which row the innovation would have to appear in, and what, appearing there, it would actually have to do: be present, attentively, for the length of a human need.

Public provision: socialised reproduction and its discontents

Where the state funds nurseries, hospitals, and domiciliary services, reproduction is partly socialised: the cost of maintaining people is collectivised through taxation rather than left to be settled inside each household’s private economy. The Marxist tradition analysed the state that performs this funding long before it analysed the funding itself, and the series treatment of marxist state theory today sets out the fiscal mechanism in its general form: the state must simultaneously sustain accumulation and legitimate itself to the population whose labour accumulation consumes, and social provision sits precisely on that seam. Care is where the seam is most visible, because the service withdrawn is not an abstraction to the family that loses it. It is a day centre closed, a home-help visit shortened, a nursery place that no longer exists, and the work returns through the front door of a specific household by the following week.

Public provision does not abolish the productivity trap; it nationalises the bill for it. The salaried carer in a public nursery faces the same arithmetic as her counterpart in a private chain, with the difference that her employer’s revenue is a budget line rather than a fee, and budgets are set in a political arena where the trap’s pressure presents as an eternally expanding claim on resources. Defenders of provision sometimes speak as though public ownership dissolves the cost problem, and critics sometimes speak as though the cost problem proves public ownership wasteful. Both are wrong in the same way: the cost disease operates under either ownership, because it is a property of the activity, not of the title deeds. What ownership changes is who can be made to absorb the pressure, and how visibly. A public system absorbs it in the open, as a budget argument. A private system absorbs it in the wage packet and the staffing ratio, where it is easier to miss.

The quality of public provision also answers a sceptical question the framework must face: if the work is socially necessary, why is its public version so often shabby? Part of the answer is the trap itself, the standing temptation to dilute ratios as costs climb. Part is political: the users of public care, small children, the frail elderly, the chronically ill, are among the least organised constituencies in any electorate, and their carers, paid and unpaid, are disproportionately women whose political weight has historically been discounted. A service for the unorganised, staffed by the discounted, funded from a budget under permanent pressure: the shabbiness is overdetermined, and the analysis that explains low pay by cost structure alone would be incomplete without this political layer laid over it.

Fiscal consolidation: how cuts travel home

The mechanism connecting a treasury decision to a kitchen table deserves to be traced slowly, because it is the framework’s most concrete prediction and the one most often verified in the literature on austerity. Begin with consolidation: a government resolves to reduce its deficit, and the resolution falls, as it tends to, on the services whose users cannot retaliate. A home-care contract is relet at a lower hourly rate. The provider, facing the productivity trap, cannot find the difference in efficiency, so it finds the difference in the visit: the hour becomes forty minutes, the forty minutes become a quarter of an hour in which a worker is expected to wash, dress, feed, and medicate a person she may barely know. Some needs fall out of the visit altogether. They do not fall out of the world. They are picked up by the daughter who leaves her own shift early, by the spouse whose health then deteriorates under the load, by the neighbour, by nobody, in which case they accumulate until they present at a hospital as an emergency far more expensive than the visit that was saved.

Feminist economists traced this circuit early and named its logic. Diane Elson’s research on macroeconomic policy argued that standard adjustment programmes treated unpaid household labour as an infinitely elastic buffer, a factor that could absorb any quantity of shed work without cost and without limit, because the models in which the programmes were designed contained no household sector at all. The buffer is not elastic. It is a woman’s day, and it tears: in forgone earnings, in deteriorated health, in children minded by children. The three responses Elson proposed for policy, to recognise the unpaid work, to reduce its burden, and to redistribute it between women and men and between households and the state, migrated from her research into the vocabulary of international agencies, which is where research phrases go when they have won the argument and lost the budget.

Why does cutting a public service lengthen the unpaid working day?

Because the need the service met does not shrink with the budget. Needs are set by bodies, not by appropriations. When the funded visit ends, the washing, feeding, lifting, and watching remain, and they are performed by whoever stands nearest, usually an unpaid woman in the same family, whose day expands to contain them.

Trace the circuit once more in the opposite direction and the fiscal saving dissolves into accounting illusion. The treasury saves the price of the visits. The health service pays more for the emergencies. The employer of the daughter pays in absence and resignation. The daughter pays in pension contributions never made, a cost deferred to the state’s own future budget for old-age support. The carer’s own eventual need for care arrives with fewer resources behind it, because the years of absorption were years of low or no contributions. Consolidation, viewed through the household, is less a saving than a transfer of costs across budgets and across decades, from accounts that are audited to lives that are not. The framework’s insistence on following the work rather than the money is what makes the transfer visible; a purely fiscal analysis sees the saving and stops, congratulating itself at exactly the point where the cost changes its name.

None of this requires ill will in the treasury. It requires only that the model omit the household, as models built for other purposes innocently do, and then be used to distribute real cuts among real services. The omission is the politics. What a framework counts determines what a policy can see, and a framework in which reproduction does not appear will reliably produce policies that consume reproduction while recording the consumption as efficiency. That is why the theoretical labour of the tradition this cluster covers, the patient insistence that the household is inside the economy rather than beneath it, has practical consequences out of all proportion to its scholastic reputation.

Marketisation: the same cost structure in private hands

Marketisation is the transfer of provision from public funding and delivery toward purchase: the state or the family buys care from providers who compete for the custom. Its advocates promise the standard benefits of competition, efficiency, responsiveness, choice. The structural analysis makes a colder prediction, and it is the prediction on which the namable claim of this article stands or falls. A market provider faces the same cost structure as a public one, because the trap is in the activity, not the ownership. The provider cannot manufacture productivity growth in attended presence. Its margin must therefore come from the variables that can move: the wage bill, the staffing ratio, the skill mix, the unpaid overtime extracted from workers’ reluctance to leave a resident unwashed, the fee where the purchaser can be made to pay it, and the selection of clients whose needs are cheap to meet.

Watch a marketised service closely and the prediction resolves into a recognisable anatomy. Labour is the overwhelming cost, so labour is where competition does its work. Contracts fragment into visits timed to the minute and travel time squeezed out of the paid day. Training thins to the statutory minimum. Continuity of carer, the relational substance of the product, is sacrificed to scheduling flexibility, because continuity is expensive and its loss does not appear in any metric the purchaser monitors. Ownership structures financialise: the provision itself yields a modest operating return, so returns are amplified by loading the property with debt and extracting value through rent and sale, a pattern that converts a care home into an asset with residents attached. When such a structure fails, it fails suddenly, and the state discovers that it remains the provider of last resort for the residents regardless of what the contracts said, having paid for the lesson that the risk was never really transferred.

None of this implies that every private provider is cynical or that public provision is automatically humane. Small providers, family firms, and cooperatives exist throughout the sector, and some public services are grim. The claim is about the gradient of pressure, not the virtue of the personnel. Set a conscientious owner down inside the cost structure and she will find the same walls closing: her fees capped by what families or the local authority can pay, her wage bill bid up by the labour market, her ratios the only slack variable she truly controls. If she refuses to use it, a competitor who does not refuse undercuts her, and the purchaser, unable to inspect quality in advance and often purchasing in a crisis, takes the cheaper offer. Markets in care select for the dilution of the very thing being bought, not because buyers are foolish but because the thing being bought cannot be verified at the point of purchase and its dilution is invisible until it is too late to matter contractually.

What does a market provider change first when costs rise?

The staffing ratio, because it is the one variable wholly within the provider’s control and the one whose deterioration is hardest for a purchaser to detect in advance. Fees face the purchaser’s budget, wages face the labour market, but the number of residents per carer on a night shift can move quietly and at once.

The deeper difficulty is that care is what economists call a trust good: its quality cannot be fully assessed even after consumption, sometimes especially by the person receiving it, who may be a child, or confused, or frightened of the consequences of complaining. Markets discipline producers where buyers can compare, defect, and report. All three mechanisms are blunted here. The family purchasing in the aftermath of a hospital discharge is comparing brochures under grief and time pressure. Defection means moving a frail person, itself a harm. Reporting depends on a recipient who may lack the capacity or the standing to be believed. The market form is not being applied to a recalcitrant sector by accident; the sector’s product has properties, relational, unobservable, non-storable, that the market form is structurally poor at handling, and the analysis that says so is an analysis of the product, not an insult to the traders.

What, then, does marketisation actually change? It changes who can be blamed. Under public provision, the dilated ratio is a political fact with an address. Under market provision, it is a business decision inside a contract, and the purchaser who set the fee and the provider who set the ratio can each point at the other with complete sincerity. The work of this article is to keep the cost structure in view through that fog of distributed responsibility, because the fog is not a side effect but one of the form’s products. The pressure table’s third row should be read with this in mind: the column for who bears the cost, under marketisation, has a way of filling up with names of people who were never in the room when the fee was agreed.

Care chains: the household that exports its caring

Extend the communicating vessels across a border and the system’s most poignant structure appears. In wealthy economies, the paid care workforce is drawn disproportionately from poorer economies, and the women who migrate to staff nurseries, wards, and private homes in the rich world leave households of their own behind. The sociologist Arlie Hochschild gave the resulting structure its enduring name, the global care chain: a series of links in which care is transferred upward along a gradient of wealth, each woman caring for another’s family while her own children or parents are cared for by someone poorer still, a relative, a neighbour, a woman from a poorer region of her own country. At the chain’s final link, the sequence ends in a household with no one left to transfer to, where the caring is done by the very young, the very old, or not done.

The research that documented the chain’s lived texture is associated above all with Rhacel Parreñas, whose study of migrant domestic workers traced how the women sustain mothering at a distance: remittances as material care, calls and messages as emotional presence, gifts and visits rationed across years. Her work records the workers’ own analysis, which is rarely sentimental. They describe the arrangement as an exchange of one family’s care deficit for another’s income, made under constraints neither family chose, and they are precise about the difference between the care they sell, which is warm and skilled and real, and the presence their own children are owed, which cannot be posted. The concept of the care drain, coined by analogy with the brain drain, names the sending economy’s loss: the departure of experienced carers, including trained nurses, from health systems that can ill afford the loss, toward private households abroad where their clinical training is used to mind children.

Who absorbs the care that migrant carers leave behind?

Usually another woman, positioned one rung lower: a grandmother whose old age becomes a second shift of childcare, an aunt, an older daughter withdrawn from schooling or work, or a paid helper hired from a poorer district. The chain does not abolish the sending household’s care needs; it redistributes them downward, to the link with the least power to refuse.

Stand back from the individual links and the chain reveals the care system of the wealthy economies as partially externalised. The rich household that solves its care problem with a migrant worker, and the rich state whose hospitals run on migrant nurses, are importing reproduction: labour power raised, educated, and emotionally formed at another economy’s expense, then purchased at the point of its usefulness. The sending economy bears the cost of the raising and the grief of the absence; the receiving economy books the benefit and counts the arrangement as a labour market functioning well. Marxist analysis has a long practice of following such transfers, asking who paid for what the purchaser received, and the care chain is that practice applied to the most intimate of commodities. What is transferred is not only labour but love’s logistics, and the accounts do not balance at any link.

The chain also disciplines the receiving end’s wages. A workforce recruited across a steep international wage gradient arrives with reservation wages set by the sending economy’s scale, which employers in the receiving economy experience as a gift: skilled, experienced carers available at rates the domestic labour supply would refuse. Immigration status does the rest. A worker whose residence depends on her employer, or whose papers are incomplete, does not bargain, does not complain to inspectors, does not leave mid-contract when the hours metastasise. The structure does not require a villain twirling a moustache; it requires only a gradient, a border regime, and households at the top with care needs they cannot meet from their own ranks at the going wage. Analysts who follow the labour market side of the sector, a context set out in the series article on precarity and the working class, will recognise the general form: a segmented workforce, graded by security of status, in which the least secure segment sets the effective floor for everyone’s conditions.

How the split was built: the family wage and its long shadow

The division between a paid world of production and an unpaid world of reproduction feels natural only because it was built slowly and then forgotten. In the early factory system, whole families entered the mill, and the wage was a family wage in the literal sense of several wages pooled. The settlement that replaced it, fought over through the nineteenth century, separated the spheres: a male wage imagined as sufficient for a household, a wife withdrawn from the factory into full-time domestic labour, children sent to school rather than to the loom. Historians of the family have shown how uneven, contested, and class-specific that settlement was, how it functioned partly as a genuine protection won by working-class organisation and partly as a discipline that made one worker’s wage carry the reproduction of several lives. Its ideological deposit, however, was uniform: the work inside the home disappeared from the category of work, becoming instead the expression of a wife’s nature and duty.

That deposit still pays out. The breadwinner settlement collapsed as a description of most households decades ago, but its accounting conventions survived it. The unpaid carer is still treated, by tax systems and pension rules and the informal arithmetic of employers, as a person supported by someone else’s wage, an appendage of a real worker rather than a worker whose product is the real worker. When she enters paid care employment, she carries the appendage status with her: the wage is bid as though it were a supplement to a household income principally earned elsewhere, a top-up rather than a living, and part-time and fragmented contracts are designed around the assumption. The assumption is false for the lone parent, false for the migrant whose remittances are a family’s main income, false for the many households where the carer’s wage is the stable one, but the wage structure was built when it was nearer true, and structures outlive the conditions that built them. This is one of the places where the tradition this cluster belongs to earns its keep: it treats the wage form as a historical artefact with a politics, not as a market’s innocent reading of worth, and the founding survey of the tradition, the series article on the marxist feminism complete guide, traces how that treatment was developed.

The family wage bargain also explains a peculiarity of the sector’s labour relations: the persistent framing of carers’ pay claims as a threat to the cared-for. When transport workers strike for wages, the public inconvenience is passengers delayed. When carers press for wages, the framing offered is residents abandoned, children unminded, patients at risk, as though the wage bill and the service were the same fund in a zero-sum household budget. Sometimes, inside a fixed public budget, they partly are, which is exactly why the fiscal section of this article insisted the pressure be followed to its source rather than fought out between the two groups the budget pits against each other. The carer and the cared-for are the two ends of the same underfunded hour, and an analysis that sets them against each other has already accepted the frame that produces both their injuries.

The argument inside the tradition: wages, refusal, and the price of visibility

Marxist feminism did not speak about the household with one voice, and the argument it had with itself in the nineteen seventies remains the sharpest available training in the stakes of the subject. One wing, associated with Mariarosa Dalla Costa and Selma James and the wages for housework campaign, demanded that the unpaid work be waged: not because a wage would liberate the housewife, but because the demand itself exposed the work as work and refused the ideology of love that concealed it. Silvia Federici’s essay for that campaign argued the point with deliberate provocation: to demand wages for housework was to refuse the naturalisation of the role, to say that the smile and the availability were labour extracted, not essence expressed. The wage was a strategy of visibility and refusal before it was a household budget proposal.

The counter-arguments came from inside the movement and were serious. A wage, critics objected, might entrench the very division of labour it named, paying women to remain in the kitchen with the state’s blessing and a price on their confinement. Others doubted that the demand could be won without a transformation of the whole wage system, and preferred to aim at the socialisation of housework, collective services that would lift the work out of the isolated household altogether. Still others, and this objection matters for the application in this article, warned against collapsing all of women’s activity in the home into the category of labour for capital, as though a grandmother’s afternoon with a child were exhausted by its function for the labour market. The tradition’s mature position, to the extent it has one, absorbed the corrections without surrendering the founding insight: the work is work, its unpaid status is a political arrangement rather than a fact of nature, and the strategies for changing the arrangement, waging, socialising, sharing, shortening the paid working day, remain live options argued among identifiable schools. The theoretical core of the quarrel, whether domestic labour produces value, is treated where it belongs, in the series article on the domestic labour debate, and is flagged here only so the reader can see which arguments in the present article do and do not depend on its outcome.

For the wage question in the paid sector, the seventies argument bequeathed a diagnostic habit: whenever caring is described as its own reward, ask who is being paid in money instead. The hospital that calls nursing a calling while negotiating hard over agency fees, the family that calls a daughter’s care natural while a son’s career is treated as sacrosanct, the state that praises informal carers in speeches while assessing their benefits as though devotion were income: each is spending the ideology the campaign named. Ideology here is not a decorative prejudice. It is a component of the wage-setting machinery, as material in its effects as the staffing ratio, and the structural account of this article keeps it in that place: a real mechanism, operating through the pricing of skill as nature, and still secondary to the cost structure that would keep the sector’s wages low even if every speech about angels were struck from the record.

Who does the paid work: a workforce graded by security

Ask who staffs the sector and the answer arrives in layers. At the credentialled end, nurses and therapists carry professional registration, scarce training, and some bargaining power, and their pay, while it lags comparable professions, reflects that shelter. Below them in the formal hierarchy, and vastly more numerous, stand the care assistants, home carers, nursery workers, cleaners of wards and homes, kitchen staff: the roles where entry requires no certificate because the competence is presumed to walk in off the street with the woman who holds it. The presumption is the wage policy. Where a role is defined as requiring only a warm heart and a clean record, it will be paid for a warm heart and a clean record, and the accumulated craft of the experienced worker, the one who can settle the ward at night with a skeleton staff, is remunerated at the rate of the novice who cannot, which is why the sector haemorrhages experience and then complains of a skills shortage of its own manufacture.

Layered over the hierarchy of credentials is a hierarchy of security. Directly employed staff with permanent contracts occupy the sheltered core. Agency workers, hired by the shift to plug gaps, often earn a higher hourly rate and a far more precarious living, no continuity of workplace, no claim on training, no sick pay worth the name, and a phone that must stay answered. At the exposed edge stand workers in the informal economy and those whose immigration status binds them to an employer or excludes them from legal employment altogether, for whom the wage question is inseparable from the question of whether they can afford to be visible to any authority at all. The three layers are not separate labour markets; they are one market, graded, and the existence of the exposed edge disciplines the core. Every negotiator for the core knows that a service can be re-tendered, agency-staffed, or informally supplied, and the knowledge sits in the room even when nobody mentions it.

Gender composition, the fact every reader already knows, is better stated as a mechanism than as a statistic. The sector recruits women because women arrive pre-trained by the household, because the wage is set at a level that presumes a second income or a willingness to subsidise the job from devotion, and because the pipeline of girls toward caring roles is laid down early in schooling and family expectation. The composition then confirms the wage: a job done by women is suspected of being easy in the way that a job done by men is suspected of being dangerous and due compensation for the danger. Break the circle at any point and the others weaken, which is why serious remedies in the literature are always compound: revaluation of the grading schemes, wage floors that remove the bottom of the market from competition, and recruitment that refuses the presumption of feminine nature, all three together, because the structure is a circle and circles are not broken by tugging one arc.

The demand side: more need, arriving on a schedule nobody chose

Everything said so far concerns the supply and pricing of care. The demand side tightens the trap from the other direction. The populations of wealthy economies are ageing, which in this context means a rising share of people in the years when frailty, dementia, and multiple chronic conditions make daily assistance a necessity rather than a comfort. Simultaneously, the pattern of family life supplies fewer unpaid hands per dependent person than the classic household form did: smaller families mean fewer adult children per elderly parent, geographical mobility scatters those children across distances that make daily attendance impossible, and the women whose unpaid labour once absorbed the need are in paid employment for more of their lives. The result is a structural expansion of the demand for paid care colliding with the cost structure this article has described, and the collision, not any minister’s decision, is what the phrase crisis of care gestures at when it is used carefully.

The political scientist Nancy Fraser gave the crisis its most cited theoretical statement, arguing that social reproduction is a standing site of crisis tendency in capitalist societies: the system free-rides on caring capacities it simultaneously depletes, pressing households to supply more paid labour while withdrawing the supports that made unpaid caring possible, until the capacities themselves, families’ time, energy, and stability, are exhausted in ways that recoil on the economy that consumed them. However a reader judges the wider theory, the formulation captures the direction of the strain this article has been tracking: the vessels of the pressure table are being filled from a shrinking reservoir, and each transfer that relieves one site deepens the depletion at the source. A nurse recruited from abroad is a gain to the ward and a subtraction from her own family’s stock of care. A daughter who leaves her job to mind a parent is a gain to the parent and a subtraction from the tax base that funds the parent’s pension. The system of transfers can continue, and does, but it continues by consuming stocks it does not replenish, and the language of crisis is the language for a process with that shape.

It is worth insisting, against a lazy reading, that none of this is an argument that people live too long or that families have failed. Longevity is a triumph; a society in which most people survive into their eighties is one of the great achievements of the very economies under discussion. The crisis lies in the meeting between that achievement and an organisation of care that was designed, insofar as it was designed at all, for a world of shorter lives, larger families, and a reserve army of unwaged wives. The need is honourable. The organisation is an anachronism. Holding those two judgements apart is the difference between an analysis and a complaint, and it is the difference this article has tried to maintain throughout.

The honest counter-case: the household the theory assumed has thinned out

An analysis that cannot state what has changed against it is a catechism, not an analysis, and the changes here are large. The classic accounts of domestic labour were built around a household containing a male earner in full-time employment and a female carer outside it. That household is now a minority form in many economies. Women’s participation in paid employment has risen across the long run, a transformation documented in the economic historian Claudia Goldin’s research on the evolution of female labour supply, which traces its S-shaped advance across the development process and its acceleration as service employment, education, and contraception reshaped the choices households faced. Fertility has fallen, so the years consumed by intensive childcare are fewer in a lifetime. Households themselves have changed shape: more people live alone, partnerships form later and dissolve more readily, and the multi-generational household under one roof, once a normal container for both childcare and eldercare, has become the exception in the wealthy economies.

Household technology is the counter-case’s favourite exhibit, and it deserves a precise statement rather than a wave. Washing machines, piped water, refrigeration, prepared foods, and the vacuum cleaner abolished specific forms of drudgery that consumed the nineteenth-century housewife’s week: the hauling of water, the day-long wash, the daily marketing and cooking from raw ingredients in economies without cold storage. The time-use research of Jonathan Gershuny, assembled from diary studies across several countries and decades, gives the balanced verdict the slogan-mongers omit. Gershuny found that the total time devoted to domestic work fell less than the technology promised, because standards rose to meet the machinery: clothes washed more often, meals more varied, floors expected cleaner, children supervised more intensively. He also documented a slow convergence, men’s unpaid work rising from a low base, women’s falling, without the two meeting, a pattern he characterised as lagged adaptation. The machine changed the content of the housewife’s day more than its length, and abolished its heaviest labour without abolishing its structure of responsibility.

What has household technology actually changed?

It changed the composition of domestic labour more than its volume. Drudgery with water, fuel, and food storage largely vanished in wealthy economies, while the standard of cleanliness, nutrition, and child supervision expected of a household rose to absorb much of the time released, leaving the managerial and emotional core of the work intact.

A further change cuts the other way, and honesty requires logging it alongside the rest. The same decades that thinned the breadwinner household intensified the standards of care at both ends of life. Childhood is more supervised, more scheduled, more anxious than the childhood of the classic accounts; developmental expectations arrive with expert literatures attached, and the parent who meets them is performing more deliberate, more informed, more time-hungry caring than the grandmother who sent children out to play until dark. Eldercare, similarly, is now the management of complex medication regimes, appointments, and technologies at home, a quasi-clinical workload unimaginable when the old simply faded in a back room. The household form changed. The work’s volume per dependant, at the standards now expected, rose. An account that registered only the first change would be describing a different society from the one its readers live in, which is the brief’s warning and the reason this section exists.

The strongest objection, stated at full strength

The objection can now be put as its proponents in the research on household composition and participation would put it, without softening. The framework, they say, is a theory of a vanished world. Its founding picture, a waged husband, an unwaged wife, a flow of value between factory and kitchen, describes a household form that has been dissolving for half a century and is now statistically marginal in the economies the theory claims to explain. Participation has risen, fertility has fallen, technology has mechanised the house, men do measurably more than their fathers did, and a vast apparatus of markets and services has grown up to sell what the housewife once made. To continue analysing care through the breadwinner lens is, on this objection, like analysing agriculture through the ox-plough: the categories may retain a museum charm, but the productive system they described has been replaced, and a theory that cannot see the replacement forfeits its claim to explain the wages, services, and households that actually exist.

There is force in the objection beyond its rhetoric, and the honest answer begins by conceding its target. Specific claims in the classic literature about the domestic division of labour, who does what in the home and on what terms, do require updating rather than restatement. The housewife of the founding texts, full-time, lifelong, economically dependent by design, is a historical figure, and any writer who deploys her as the typical contemporary carer is writing fiction in the indicative mood. The tradition’s textbook summaries are guilty of the deployment often enough that the objection lands on real targets. Conceded, then: the ethnography must be rewritten for dual-earner households, lone parents, same-sex couples, commuters, and the scattered kin networks that actually hold dependants up.

What survives the concession is the framework’s actual object, which was never the ethnography. The theory of social reproduction tracks a transfer: the movement of caring labour between households, markets, and states as each site expands or sheds it, and the systematic refusal of the economy’s accounting to register the transfer when it lands on the unpaid end. That object has not dissolved with the breadwinner household. It has become more visible. The dual-earner household does not abolish the transfer; it performs the transfer at higher speed, buying in the market, leaning on grandparents, importing a migrant worker, absorbing the residue in exhausted evenings. The framework predicts exactly the phenomena the objection cites as its refutations: the growth of the paid care sector as households shed labour they once performed, the emergence of care chains as the shedding crosses borders, the fiscal fights over provision as the state is asked to catch what families drop. A theory whose supposed refutations are among its own predictions has not been overtaken by the facts. It has been confirmed in the specific way theories are confirmed, by the world’s obliging production of its consequences. The answer to the objection, in short, is that the framework was always about the transfers, and the transfers have intensified; what must be discarded is the period costume in which the founding texts dressed them, and discarding costumes is maintenance, not surrender.

Can care be sold whole? The limits of commodification

If the household sheds labour and the state rations it, the market offers to absorb the difference, and an obvious question follows: why should the transfer not run to completion, care becoming a fully commodified service like any other, bought and sold without residue? The question is empirical as well as theoretical, because the experiment is running at scale, and the structural analysis identifies three points where the commodity form strains. The first is the productivity trap itself, which ensures that fully marketised care is chronically expensive relative to incomes, so that the market clears only by diluting quality or by drawing on the subsidised labour supplies, migrant, informal, devotion-driven, that let the price appear viable. The second is the trust problem examined earlier: quality in care is difficult to verify, so the market’s selection pressure works against the attribute the purchaser most wants. The third is the strangest and the most interesting. Part of what is being bought is a relationship, and the buyer wants the relationship to be genuine, which places the seller in the position of being paid to feel what the contract cannot specify.

The sociologist Viviana Zelizer’s research on the intersection of intimacy and economic transaction is the essential corrective to crude versions of both sides of this argument. Against the view that money corrupts intimacy on contact, Zelizer showed that people constantly and competently mix the two, marking different relationships with different kinds of payment, distinguishing the wage, the gift, the allowance, and the fee with great precision, and sustaining real affection inside paid arrangements without confusion. Paid care is not thereby exposed as counterfeit. A nanny may love the child, a carer may grieve for the resident, and the wage no more falsifies the feeling than a salary falsifies a teacher’s interest in her students. What the commodity form changes is not the feeling’s authenticity but its security: the relationship exists at the pleasure of a contract, terminable by either side’s economics, and both parties know it. The child does not know it, which is why the turnover figures in marketised childcare are not a staffing statistic but a developmental fact.

Emotional labour, the managed display of feeling as part of the job, is the point where commodification reaches into the worker’s own inner economy, and its analysis, developed in Hochschild’s research on service work, is carried in this series by the article on alienation at work today, which treats the flight attendant’s smile and the carer’s patience as members of one family. Here it is enough to mark the boundary of the present argument: the productivity trap and the pressure table explain the sector’s economics without needing the full theory of emotional labour, but any account of what the job costs the worker that stopped at hours and wages would be missing the expenditure she makes from her own stock of feeling, unrecorded in every ledger the sector keeps.

Which parts of care resist being sold at all?

Judgement exercised over years about one particular person, availability at unsocial moments, and the willingness to notice what nobody has asked you to notice. A contract can purchase attendance and tasks. The accumulated attentiveness that makes the tasks worth doing adheres to the relationship, and it leaves the room when the contract churns the staff.

Families know the limits of the purchasable from the other side. The wealthy household can buy excellent care and frequently does, and the transaction can be humane on both sides of it. What it cannot buy is the abolition of the family’s own caring obligations, which reappear as management: the choosing, supervising, supplementing, and worrying that sociologists of the household file under the mental load, the planning and monitoring work that persists when every manual task has been delegated. Studies of households that outsource heavily find the managerial layer stubbornly resident, and resident in the same person as before, which is why the image of care as a burden that money simply lifts is false even at the top of the income scale. Money buys hands. It does not so easily buy the standing responsibility for a life, and the residual responsibility is a large part of what the word care means to those who carry it.

Time: the currency in which care is actually paid

Strip the sector to its material basis and care is a claim on time, specifically on time that cannot be banked, overlapped, or recovered. The paid working day and the unpaid one draw on the same twenty-four hours, and the history of the labour movement’s central demand, the shortening of the working day, is also a history of care policy conducted under another name. Every hour struck off the standard week since the nineteenth century has been, among other things, an hour returned to the work of maintaining people, and the feminist wing of the tradition has pressed the point explicitly: proposals for a shorter paid week recur in the remedies literature not as a lifestyle preference but as the one reform that directly enlarges the resource care consumes. Time, unlike money, cannot be in two places, and a society that wants more caring done, to higher standards, by less exhausted people, is asking for time, and will have to say whose paid hours it proposes to convert.

The temporal structure of care also sabotages the usual flexibilities of the labour market. Demand for care peaks at the day’s edges, early mornings, evenings, weekends, nights, precisely the hours the rest of the economy calls unsocial and prices at a premium. Care employers, trapped on costs, price them at a discount instead, staffing the edges with the workers least able to refuse. The result is a workforce whose paid hours are scattered across the clock’s worst territory while its own unpaid caring, which obeys the same clock, collides with the paid shifts at both ends. The home carer who leaves a client’s breakfast to rush her own child to school, then returns to a stranger’s lunchtime, lives the collision twice daily. An analysis that counted only her wage would miss the shape of her exploitation, which is as much a confiscation of sequence and simultaneity as of money: she is required to be in two kitchens at the same hour, and the impossibility is managed by her running.

There is a research tradition on exactly this collision, the study of time famine and schedule control, and its findings run in the framework’s direction: perceived time pressure concentrates among those carrying both paid work and unpaid dependants, and control over one’s schedule predicts wellbeing more reliably than income once basic needs are met. The policy translations differ, predictable hours legislation, rights to part-time work without penalty, care leave that pays enough to be taken by the higher earner, but the analysis underneath them is one: the scarcity in the care economy is not, at root, a scarcity of compassion or even of money. It is a scarcity of hours in the lives of the people expected to do the caring, and every remedy that does not add hours, by shortening paid work, socialising tasks, or staffing generously, is a redistribution of exhaustion wearing a reform’s clothes.

Remedies, stated as positions: who proposes what, and what each proposal sees

The remedies proposed for the care problem sort into families, and the families are best understood as wagers about which mechanism in this article is the operative one. They are presented here as positions held by identifiable schools, not as recommendations of this article, and the set deliberately includes remedies from outside the Marxist tradition, because the debate is not the tradition’s private property and its other participants often see a part of the structure more clearly than the whole.

The first family wages the household. Descending from the wages for housework campaign, its contemporary descendants propose carer’s incomes, credits toward pensions for years spent caring, and payments that recognise the dependent relative’s household as a workplace. The wager is that visibility through price is the fastest route to status and security, and that a carer with an income of her own bargains differently at every table, including the kitchen one. Its critics, including many feminists, wager back that payment may seal the carer into the role and relieve everyone else of the obligation to change.

The second family socialises. Its adherents, social democrats and the Marxist tradition’s service-oriented wing together, propose expanding public provision until care resembles schooling: a universal service, free at the point of use, staffed by professionals on public pay scales. The phrase universal basic services has been advanced in the policy literature for the bundled version of this proposal. The wager is that only collectivised funding escapes the trap’s arithmetic at the point of delivery, since the taxpayer, unlike the family, can be made to pay the rising relative cost of a stagnant-productivity service without diluting it. Its critics point to the fiscal exposure the trap guarantees, and to the record of public services diluted from within when the exposure bites.

The third family regulates the market it keeps. Wage floors for the sector, mandated staffing ratios, licensing that bites, limits on the financial extraction available to owners, portable entitlements for migrant workers: the wager is that the market’s known failure modes in care are specific enough to be fenced. This family includes liberal economists whose framework owes nothing to Marxism and who arrive at regulation through the trust-good problem alone, a convergence the reader should register as evidence that the structural diagnosis is not a sectarian artefact. Its critics, from the right, predict exit and informality; from the left, that fencing a structure built on cheap labour leaves the cheapness enthroned.

The fourth family redistributes time. Shorter standard working weeks, care leave designed to be taken by men, rights to flexible scheduling controlled by the worker rather than the employer: the wager is that the binding constraint is hours, and that no rearrangement of money compensates a population with no time to care in. Its ancestry in the labour movement is the oldest of any family’s, and its feminist variant adds that leave not taken by fathers is leave that functions as a subsidy to the existing division. Its critics ask who staffs the services while everyone works less, a question with a real answer, more staff at higher cost, which returns the argument to the fiscal exposure the second family accepts openly.

A fifth position should be stated because it is widely held and the analysis has implications for it: the market-optimist position that rising demand for care will itself raise care wages, as employers bid against scarcity, and that technology, monitoring, robotics, remote presence, will relieve the labour constraint as it has in other sectors. The structural account answers on both points without needing to deny either outright. Scarcity has coexisted with low pay in the sector for decades, because the purchaser’s ability to pay, not the worker’s scarcity, is the binding constraint where the buyer is a fixed public budget or a pensioner’s savings. And the technology that assists caring does not substitute for its core hour, as the productivity section argued; the robot that lifts is welcome, and it does not sit with the frightened. The optimist position is the one the evidence of the wage pattern most embarrasses, which is why it is usually defended with the future tense.

Why do remedies cluster into families rather than single policies?

Because each remedy answers a different diagnosis of where the pressure originates: in the household’s invisibility, the state’s fiscal exposure, the market’s selection effects, or the sheer scarcity of hours. Serious proposals in the literature are compounds, and the families persist because the diagnoses are each partly true of a system with four communicating sites.

Notice what an honest map of remedies reveals about the problem’s shape. Every family, followed far enough, arrives at the same wall: somebody must pay the rising relative cost of an activity that cannot be made cheaper without being made worse, or somebody must do more of it unpaid, or the standard of what counts as adequate care must fall. The families differ in where they send the bill: to the treasury, to employers, to families, to the future. There is no fifth destination, and a debate that pretends otherwise, that promises better care, higher wages, lower taxes, and lower fees in one programme, is not a debate about care but about the location of a cost its participants have agreed not to name. The contribution of the structural analysis to policy is mostly this act of naming, which is why the article’s discipline throughout has been to follow the cost to wherever it lands rather than to where it is announced.

In the national accounts: the boundary that makes the subsidy invisible

Statistical systems draw a production boundary, and what falls outside it does not exist for the purposes of economic policy. Unpaid household production falls outside. The meal cooked for wages in a canteen enters the national product; the identical meal cooked at home does not, and when the same family shifts from the home meal to the canteen meal, measured output rises although the society has eaten exactly as before. Economists of the feminist school, Marilyn Waring’s critique of national accounting being the landmark, have pressed the consequences for decades: an economy can register growth by commodifying its own reproduction while the underlying volume of caring stands still or falls, and can register a fiscal saving by pushing services back into households while the work, and its true cost, merely change ledger. Satellite accounts that value household production exist in several statistical systems, compiled at intervals and filed where policy cannot trip over them.

The boundary matters for this article because it rigs the apparent economics of every remedy. Expand public childcare and measured output rises twice, once for the service and once for the parental earnings it releases. Cut the service and measured output falls by less than the real loss, since the returning household labour is invisible to the accounts, so the cut presents as a smaller economic event than it is. A finance ministry steering by the official gauge will systematically undervalue the unpaid sector and overvalue shifts across the boundary in either direction of convenience. The tradition’s insistence that reproduction is production is, at this unglamorous statistical level, simply correct, and its opponents in practice do not defend the boundary as truth; they defend it as convention, which concedes the argument while declining the conclusion.

It is worth pausing on why the convention survives its critics. Valuing household production requires imputing prices to unpriced activity, and every imputation embeds a decision: value the hour at the carer’s forgone wage and the wealthy household’s cooking becomes fabulously productive; value it at a hired replacement’s wage and the numbers shrink to something a treasury can contemplate. The range between methods is wide enough to be politically unusable, a figure for every prior. But the inference drawn from the dispute, that the unpaid sector should therefore be counted at zero, does not follow. Zero is also an imputation, and the least defensible one, since it asserts with false precision that the activity contributes nothing. The careful position, and the one this article has taken with penalties and hours alike, is to work with the direction and the structure and to refuse the spurious figure: the unpaid economy is large, its exact size is convention-dependent, and policy that treats it as nil will manufacture surprises at regular intervals, usually in the lives of women.

The recruitment crisis that is not a mystery

Employers in the sector periodically announce a staffing emergency: vacancies unfilled, agencies unable to supply, services closing for want of hands. The announcements are usually framed as a puzzle, a failure of the workforce to materialise, sometimes as a failure of the young to want the work. The structural analysis dissolves the puzzle without remainder. A job’s vacancy rate is a price signal wearing a disguise. At the wage offered, with the hours fragmented as offered, with the physical and emotional costs as they are, the supply of willing labour falls short of the demand, and it will continue to fall short for exactly as long as the offer is unchanged. No other industry is permitted this confusion: a restaurant that offered a cook’s job at half the going rate would not announce a mysterious shortage of cooks, and journalists would not report the shortage as a social trend. In care, the convention that the wage is fixed by nature, by budgets, by the inherent cheapness of the work, converts a pricing decision into a demographic fate.

The sector’s own responses to its shortage confirm the diagnosis by their selection. Recruitment campaigns appeal to vocation, to the rewards of the heart, to flexible hours that fit around the worker’s own caring, which is to say they compete on every margin except the wage. International recruitment expands, drawing the chain longer. Retention is pursued through resilience training and wellbeing programmes addressed to the individual worker’s capacity to endure, rather than to the ratio that determines how much endurance the shift will consume. Each response is rational for an employer locked inside the cost structure, and each leaves the structure untouched, and the shortage duly returns on schedule, announced with fresh surprise. The shortage is the productivity trap expressing itself in the labour market: a sector that cannot raise its price and will not raise its wage must run permanently short of the labour it needs, and call the shortfall by any name except its own.

What the structural account stakes: predictions and the evidence that would break it

An explanation earns its place by risking itself. The attitudinal account of low care pay predicts that as social regard for carers rises, pay follows; the structural account predicts that regard can rise indefinitely, through campaigns, crises of public gratitude, and official rhetoric, while pay stays pinned until somebody funds the rising relative cost of the service. The two accounts part company observably, and the decades in which carers have been publicly celebrated and privately underpaid count, on the structural reading, as evidence rather than as irony. Similarly, the structural account predicts that marketising a service will express itself first in staffing ratios and turnover rather than in gleaming efficiency, that fiscal consolidation will lengthen the unpaid working day of women in the affected households whether or not any statistician records it, and that sectors of care where genuine productivity growth is possible, the logistics of laundry, catering, scheduling, will see wages behave more like ordinary wages, while the attended hour itself refuses to follow. Each prediction is stated in the literature of the field, and each is the kind of claim the pressure table was built to organise.

What would count as evidence against the productivity trap?

A sustained period in which a care service raised output per worker substantially while independently assessed quality held steady, and wages rose with the productivity rather than with general earnings. Isolated gadgets would not do it; the trap is a claim about the core hour, so the refutation must show the core hour compressing without loss.

Honesty about the trap’s limits matters as much as its defence. The argument explains the pressure on pay in the attended core of care. It does not explain everything about every care job. Ancillary work in the sector, cooking at scale, laundry, transport, records, has productivities like any other industry’s, and its wages are set by the wider labour market for those trades, which is why the hospital porter and the hospital cook do not exhibit the trap in its pure form. Nor does the trap explain the grading insult documented earlier, the evaluation schemes that scored responsibility for people below responsibility for things; that is the association mechanism’s work, and the two mechanisms are complements, not rivals. A reader who finds a passage in the literature attributing the entire wage pattern to Baumol, or the entire pattern to prejudice, is reading a pamphleteer. The position of this article is the compound one, and the compound is stronger for admitting the seam.

Misreadings this analysis exists to correct

It may help to gather the errors the preceding sections have been correcting one by one, since they recur in coverage of the sector with metronomic regularity. The first is that care is undervalued only because it is associated with women. The association is real and operative, but the word only smuggles in the conclusion that changing attitudes would change the wage. Strip the association away in imagination, people the sector with men, and the cost structure remains, the fees still capped, the budgets still exposed; the wages would rise somewhat with the coding and then stop at the wall the trap describes. The association explains the wage’s level partly and its social tolerability greatly. It does not explain its persistence against every change of government and ownership.

The second misreading is that marketisation raises efficiency in care, where efficiency must be defined before the sentence means anything. If efficiency means output per pound spent, dilution achieves it on paper by redefining the output downward. If it means the same human result per hour of labour, the trap denies the possibility in the core activity. Most policy uses of the word slide between the two meanings mid-paragraph, banking the moral credit of the second for the arithmetic of the first.

The third is that unpaid care has fallen to negligible levels, a claim that survives by confusing the decline of one household form with the decline of the work. The time diaries say otherwise, and so does the observable exhaustion of the dual-earner generation, and so, most simply, does the arithmetic of ageing: more dependants, living longer, with more complex needs, are not cared for by nobody. The fourth is that care work is unskilled, refuted by any honest inventory of a senior carer’s hour, and sustained only by the accounting convention that prices visible training and ignores the household’s. The fifth is that the classic analysis still describes the typical household, the mirror-image error of the third: the breadwinner home is gone as the statistical norm, and the framework survives not by denying its departure but by tracking where its work went. Between the third misreading and the fifth lies the whole discipline of the subject: the work persists, the forms change, and an analysis is judged by whether it follows the work through the forms.

The other end of the relationship: dependence, dignity, and the assisted life

An article of this length about care risks a distortion its subject would be the first to name: it has spoken of the cared-for mostly as the occasion of labour, a pressure in a table, a term in a cost structure. The people at the receiving end have their own analysis, and it corrects the carer’s literature at a fundamental point. The independent living movement, built by disabled people themselves, objected to the very word care for what it implied: a passive recipient, a grateful burden, a relationship defined by the giver’s virtue. Its counter-vocabulary, personal assistance, redescribes the same hours as the assistant’s execution of the disabled person’s own direction, labour in the service of another’s autonomy rather than custody of another’s helplessness. The redescription is not cosmetic. It relocates the skill: the assistant’s competence lies partly in taking instruction from the person who knows the body’s routines best, and the power in the relationship is meant, on this model, to run from the assisted person outward.

Feminist scholarship absorbed the correction and complicated it in return. Thinkers in the ethics of care tradition, and Marxist feminists in dialogue with them, pointed out that dependence is not a deviant condition of some lives but the universal condition of every life at its beginning, in illness, and commonly at its end, so that a political philosophy built on the fiction of the independent adult contractor starts from a false anthropology. The point cuts both ways, and this article holds both edges. Against the market’s fiction, it says the need for care is not a market failure or a private misfortune but the human baseline, and a society’s organisation of dependence is a central political fact about it. Against any romanticism of the caring relation, the independent living critique says the recipient is a citizen with authority over the terms of assistance, not the raw material of somebody else’s vocation. A labour analysis that forgot the second edge would reproduce, inside its radicalism, the same erasure it charges the wage system with: the person disappears, this time into the carer’s heroism rather than into the employer’s ratio.

The two standpoints meet in an unexpected place: both want the work counted, funded, and staffed as the skilled public necessity it is. The disabled person’s movement and the carers’ unions have historically eyed each other with suspicion, the one fearing that professionalisation means custody, the other that personal budgets mean casualisation, and the suspicion is itself a product of the scarcity this article anatomises: under a fixed and inadequate budget, the recipient’s control and the worker’s security genuinely trade against each other, and the fight between them is another of the budget’s arranged fights, like the one between carer and cared-for over the wage. The structural reading insists on the usual move: the trade-off is real inside the constraint and manufactured by it, and politics consists in refusing to mistake the constraint for a law.

The double day of the paid carer

Follow one worker through the argument and the abstractions acquire a face. The home carer rises before her own household, prepares her children for a school she will not see, and spends the day performing for strangers’ parents the intimate labour her own are beginning to need from her. Her paid work is the unpaid work of her evenings, professionalised and timed; her unpaid work is her paid work, unprofessionalised and endless. Sociologists call the pattern the double day, and in this sector it is not a metaphor but a rota. The woman the system relies on at both ends is frequently the same woman, and the system’s accounts record her twice at a discount: her wage depressed by the presumption that her skill is nature, her home labour unrecorded on the presumption that it is love.

Her position in the chains of this article is worth a slow look, because she is the pressure table’s human form. At work she is row three or row two, the market’s or the state’s employee. At home she is row one. If she is a migrant, she is row four at work and row one at a distance, her own row-one obligations discharged by a grandmother she supports with the wage that row four pays. Cuts to row two land on her twice, once as a worker whose hours are thinned and once as a daughter whose mother’s visits are shortened. A rise in her wage is a rise in the cost of the care her own family must buy. There is no standpoint outside the system from which she can simply win, which is not a counsel of despair but a precise description of why the sector’s politics are so difficult and why single-row remedies fail the very people they celebrate.

What she knows, and what the analysis should record as knowledge rather than colour, is the difference the staffing ratio makes to a life at each end of it. She knows which resident stops eating when the familiar face is rotated away, which client’s confusion is a urinary infection announcing itself and which is fear, how long a morning routine can be compressed before it becomes a handling rather than a washing. This is the inventory the grading schemes could not see. It is held in memory, it is destroyed by churn, and it walks out of the door every time an experienced worker leaves for a warehouse job that pays the same for less of her soul. The sector’s true capital stock is this accumulated knowledge of particular people, and it depreciates exactly as fast as the labour market is allowed to churn it, unrecorded in any balance sheet the owners keep.

Two ends of life: childcare and eldercare under one trap

The sector’s two great branches face the same cost structure with opposite political economies, and comparing them tests the framework. Childcare is purchased by young households at the poorest point of their earnings lives, for a need that is intense, temporary, and developmentally front-loaded: the years when the ratio of adults to children must be lowest are exactly the years when the service costs most per hour and the purchasers can pay least. The predictable result, visible across marketised systems, is a three-way split: affluent families buy quality, poor families patch together kin and informality, and the workforce’s wages settle low enough to keep the middle tier’s fees within reach of a second earner’s salary, the whole arrangement balanced on the nursery worker’s pay packet like a table on a single leg. Eldercare inverts the emotional geometry. The need is open-ended, often escalating, and purchased from assets or public funds rather than current earnings; its workforce is older itself, its settings less visible, its recipients less able to advocate. Where childcare’s politics are driven by parents who will someday stop needing it, eldercare’s are muted by the fact that its users rarely recover the voice that frailty takes.

Yet the trap is visibly the same trap. Raise the nursery ratio and the toddlers get less of the attention that is the service. Raise the care home ratio and the residents get less of the surveillance and society that keep them safe and oriented. Pay rises in either branch, unfunded, translate into fees that exclude the marginal family or into local authority contracts that buy fewer hours. The difference between the branches lies in where the shed labour goes: out of the nursery it goes to grandparents and to mothers’ reduced hours, out of the care home it goes nowhere, because no household can replicate a night ward, so it comes back as hospital admissions and the statistic euphemised as delayed discharge. A framework that treats the branches separately will misread both; the eldercare crisis and the childcare crisis are one crisis of social reproduction wearing two of life’s costumes, and the policy debate that funds nurseries by raiding domiciliary budgets is moving pressure within the table in the manner this article has described too often to need repeating.

Grandparents and the kinship infrastructure

Between the household and the market stands an institution the accounts ignore almost completely: the grandparent. In economies where formal childcare is expensive and parental leave short, grandparents constitute a vast unpaid childcare infrastructure, and their labour is a hidden subsidy to employers who never see it on a cost line and to states that never budget for it. The grandparent’s position in the pressure table is row one extended across a generation, and it has its own political economy. Her labour is given inside the moral economy of kinship, where refusal is expensive in love’s currency and payment is nearly unspeakable. It is also, frequently, the arrangement everyone prefers: the child is with family, the parents trust the carer absolutely, the grandparent gains a centrality in the family’s life that modern retirement otherwise rarely offers. The analysis must be able to say two things at once: that this is often the best care available, freely and gladly given, and that an economy running on it is running on the unpaid labour of older women, whose pension entitlements and health bear the costs, and whose withdrawal, through illness or distance or their own employment extending later into life, would expose how much of the formal economy’s daily functioning is mortgaged to their mornings.

The gladness is not a refutation, and treating it as one is the sentimental error symmetric to the crude error of treating all family care as exploitation. People care for their kin out of love, obligation, identity, and delight, in proportions no survey will separate, and a theory that described a grandmother’s afternoon as nothing but a subsidy to capital would deserve the mockery it received. What the framework claims is narrower and survives the gladness intact: that the total quantity of this labour is a structural variable in the economy, that its distribution is patterned by gender and class, that its invisibility in the accounts distorts policy, and that its providers bear costs which fall due in their own old age. Love explains why the work is given. It does not explain why the giving is so reliably extracted from the same people, nor why the economy is arranged as though the supply were infinite and free. Between the crude error and the sentimental one lies the position this whole cluster of articles occupies: the work is work, the love is real, and the arrangement that farms the love is a fit subject for political economy.

How does the framework handle care done gladly within families?

By separating the motive from the structure. The gladness is the giver’s and is real; the pattern of who gives, on what terms, and at what cost to their earnings and old age is the economy’s and is analysable. Affection explains the supply of any single act of care. It does not explain the distribution of the total burden.

That separation also disciplines the tradition’s own rhetoric, which has sometimes reached for the language of theft where the language of structure would serve better. A daughter caring for her mother is not the scene of a crime, and the mother is not an extractor. The extraction, where the word belongs, happens at the level of the arrangement: a fiscal system that budgets on the assumption of her unpaid availability, an employer who schedules on the assumption of her flexibility, a service that discharges patients into her spare room and calls the arrangement community care. Naming the level correctly matters practically, because remedies aimed at the wrong level, the exhortation of families to care less or of individuals to bargain harder at the kitchen table, ask the people with the least room to solve a problem whose levers are elsewhere.

Insurance and the funding of old age: the market’s quiet retreat

One funding model deserves its own examination because its failure is so instructive: private insurance for long-term care. The product seems natural. Old age’s dependency is a risk, risks can be pooled, premiums paid across a working life can purchase care at the end of it. Yet in economy after economy the market for such insurance has stayed thin, and providers who entered it have retreated, and the reasons given by the industry itself read like a checklist of the structural problems this article has catalogued. The risk is not independent across policyholders in the way actuaries prefer: longevity rises for whole cohorts at once, and the cost of the promised care rises with the cost disease, so the insurer is short exactly the variable, future care prices, that the productivity trap sends climbing without limit. A policy priced honestly against that trajectory is unaffordable to most; a policy priced affordably is a promise the seller quietly hopes to renegotiate. Add adverse selection, the tendency of those who expect to need care to buy and those who do not to abstain, and the pool worsens further.

The lesson generalises beyond insurance. Any funding mechanism that relies on setting aside a sum, rather than on taxing current production at the moment of need, runs into the same wall: the care of the future will be performed by the workers of the future, at the wages of the future, and no financial instrument manufactures their hours. Pay-as-you-go social provision is not an ideological preference in this light; it is the only structure that matches the liability to the resource, since the liability is denominated in hours of attention and the resource is the working population’s time. Financialised pre-funding converts the claim into assets whose value depends on the same economy that must supply the carers, adding a layer of promise without adding an hour of care. The point is rarely made in the funding debate, which is conducted entirely in money, and it is the kind of point a labour-centred analysis exists to make: at the end of every funding chain stands a person in a room with another person, for an hour that somebody’s working day must contain.

Reading a care statistic without being read by it

Since this article has refused the reader every figure, it owes the reader a method for the figures encountered elsewhere, because the debate is full of them and most arrive pre-interpreted. Begin with the boundary question: does the number count unpaid work, and if it does, by which imputation, replacement cost or forgone earnings, two methods that produce totals differing by multiples? Continue with the denominator: a share of the workforce, a share of spending, a share of hours, each supports a different headline from the same underlying reality. Ask who is missing from the count. Informal and migrant workers vanish from employment statistics by construction; supervisory care vanishes from time diaries by design; the residents of unregistered settings vanish from inspection data entirely. Ask what the trend is being compared against, since a rising share of spending on care is exactly what the cost disease predicts for a society holding its standards constant, and reporting it as profligacy without the counterfactual is an editorial act disguised as arithmetic.

The discipline extends to the heart-warming statistic as much as to the alarming one. Numbers celebrating the growth of the care economy as an jobs engine usually omit that the growth is the trap’s arithmetic working itself out: a sector whose productivity is stagnant must employ a rising share of the workforce to deliver a constant standard to a growing dependent population, which is simultaneously a triumph of employment and a symptom of the cost structure, depending on the column it appears in. The framework reader learns to hold both descriptions at once and to ask, of any care statistic, the question this article has asked of every phenomenon it touched: where did the cost land, and on whose day? A number that cannot answer that question is not false, exactly. It is unfinished, and unfinished numbers are the ones that do the most work in public argument, because each side completes them in its own interest before the reader has noticed the gap.

The professional exit and the argument with liberal feminism

There is a feminism for which the care problem has a ready solution: the professional woman hires help, and equality at the top of the labour market is achieved by transferring her household’s caring to another woman lower down the market. The Marxist feminist objection to this settlement is not envy and not purism; it is that the settlement does not abolish the structure, it staffs it, and that a feminism whose victory condition is the right to be on the purchasing side of the care chain has quietly redefined emancipation as a change of position within the arrangement rather than a change of the arrangement. The objection has been pressed within feminist debate by critics of the corporate-leaning variant of the movement, and it draws blood because the transfer is visible: the liberation of the professional household and the exhaustion of the migrant household are the same transaction read from two ends, and a framework that cannot hold both ends in one view is not a theory of women’s emancipation but a class memoir.

Fairness requires the reply, and the reply has weight. The professional woman did not design the chain; she faces the same time famine as everyone else, in a household form that no longer contains an unwaged adult, and her purchase of care is often what allows a migrant worker an income that worker actively prefers to the alternatives at home, a preference the worker’s own testimony, in Parreñas’s research and others, states clearly enough. Condemning the purchaser lets the structure walk free, again. The framework’s actual target is the set of arrangements that make her choice and the worker’s choice the only ones on the table: the absence of public provision, the length of the working day, the wage structure that makes the chain’s gradient so steep. Between the two feminisms the difference is finally about where the analysis stops. One stops at the composition of the boardroom. The other asks who is minding the boardroom’s children and the cleaner’s children, and declines to call the answer a private matter. On the evidence of the wage pattern with which this article began, the second question is the one that leads somewhere.

Men at the bedside: the escalator, the stigma, and the test they provide

The minority of men in the sector are analytically precious, because they let the reader separate the penalty’s components. If low pay followed the worker’s sex alone, men entering care would be paid as men are paid elsewhere. In the event, men in caring occupations earn less than men in comparably skilled non-caring work, confirming that the penalty attaches to the work; and within the sector men are promoted toward coordination, management, and the technical specialisms at rates their share of the workforce does not explain, confirming that the gender hierarchy operates inside the penalty as well. The two findings together vindicate the compound account this article has defended throughout: a cost structure that depresses the whole sector’s wages, and a coding of the work as feminine that depresses its grading further and arranges its internal promotions.

Men’s own accounts of the work add a texture the statistics miss. Male carers report the suspicion tax: the assumption, applied to them in childcare settings especially, that a man’s wish to work with children requires explanation in a way a woman’s does not. The suspicion is the femininity coding read in reverse, the same assumption that prices the work as nature making the man who chooses it legible only as an anomaly. It falls hardest exactly where the work is most intimate and least paid, and it helps keep the sector’s recruitment pool halved at the source, an outcome the sector then laments as a workforce shortage. A labour market that first declares a job to be women’s nature, then treats the women as unskilled for possessing the nature, then treats men who apply as suspect for lacking it, has achieved a perfect closure, and the closure’s perfection is the best evidence that what is being policed is not competence but the boundary of a low-wage enclave whose cheapness several other arrangements have come to depend on.

Professionalisation: the strategy of the certificate and its limits

Faced with the pricing of skill as nature, the sector’s credentialled wing pursued the classic remedy: make the training visible. Nursing’s long campaign for registration, degree-level entry, and protected title was a deliberate strategy to move its labour from the nature column to the skill column of the ledger, and it worked in the way the framework predicts, partially and unevenly. Registered nurses won a pay structure and a public standing that the unregistered carer still lacks, and the line between the two became one of the sector’s internal frontiers, patrolled by qualifications. But professionalisation has a ceiling set by the same cost structure: the more of the workforce is credentialled and priced as skilled, the faster the service’s costs climb against its capped revenue, so the system responds by redividing the labour, pushing tasks down to cheaper, unregistered grades, a process the workforce calls dilution and management calls skill mix. The certificate wins ground for its holders while the work escapes downward to new pairs of hands priced as nature again.

The pattern generalises into a melancholy law of the sector: every successful revaluation creates a new unvalued tier beneath it. Childcare repeats it, as graduate early-years teachers are layered over nursery assistants whose ratios carry the day. Social work repeats it, as qualified practitioners manage cases whose visits are made by others. The strategy is not futile; the tiers are real improvements in the lives of those who climb them, and the professions’ defence of standards protects recipients as well as members. But as a solution to the wage pattern it is structurally incomplete, because it operates on the coding of particular roles while leaving untouched the revenue constraint that requires most of the sector’s hours to be cheap. Somebody’s hour must be priced low for the present funding of care to add up, and professionalisation decides whose, which is a distributional achievement and not an escape. The escape, if there is one, runs through the funding question the remedies section mapped, and the professions’ own campaigns arrive there in the end, arguing for budgets in the language of safety after the language of status has won all it can.

Three ways to organise a necessity: what comparison shows

Comparison across welfare systems is the nearest thing this field has to a laboratory, and the comparative literature, Gøsta Esping-Andersen’s typology of welfare regimes being its most cited map, finds the wealthy economies sorting into recognisable arrangements. Some systems collectivise care through public services, defamilialising it in the literature’s term: the individual’s access to care and the carer’s employment are both detached from family position, and the state appears as employer and funder on a large scale. Some support the family’s own caring with payments, leave entitlements, and services that supplement rather than replace kin, keeping the household as the primary site while cushioning it. Some leave the matter largely to market purchase and unpaid family labour, with public provision residual and means-tested for those who can command neither. The typology is a map, not a league table, and every real system mixes the logics, but the map earns its place here because of what does not vary across it.

What does not vary is the wage pattern at the article’s centre, and the productivity pressure underneath it. The collectivised systems employ their carers on public scales and still find care consuming a rising share of budgets, still fight the ratio battles, still recruit internationally to fill the rotas. The market-led systems exhibit the pattern in its rawest form. The familial systems conceal it most successfully, inside households, at the price of the carers’ employment and independence. Comparison therefore performs the same service for the reader as it did in the historical sections: it separates the variable from the constant. Ownership, funding, and family policy are the variables, and they decide where the cost lands and how visible it is. The rising relative cost of attended human time is the constant, and it lands somewhere under every flag. A comparative politics of care that attends only to the variables will keep being surprised by the constant; the regimes differ in their answers to the question this article poses, and none has abolished the question.

The concept of defamilialisation deserves its own pause, because it names a dimension along which women’s independence can be measured without reference to wages at all: the extent to which a person can maintain a socially acceptable life without relying on family relationships, either as carer or as cared-for. A lone mother in a service-rich system, an elderly widow in a system with real domiciliary provision, possess an exit option from family dependence that their counterparts elsewhere do not, and the option changes the bargaining inside families even when it is never exercised. Feminist critics amended the typology here, pointing out that the original map had been drawn from the standpoint of the male worker’s relation to the market, and that regimes identical on that axis could differ wholly on the axis of who cares and at what cost to her. The amendment is now standard in the literature, and it is the comparative version of this article’s founding insistence: follow the caring, and the map of any welfare state redraws itself around the households and wards where its real transfers occur.

Inside the institution: skill mix, churn, and the depreciation of knowledge

Zoom in from regimes to a single ward or nursery and the cost structure becomes a management practice with a daily texture. The budget arrives as a staffing establishment, a number of posts at grades, and the manager’s craft is the continuous re-solving of an equation whose variables will not sit still: sickness, turnover, a resident whose needs double overnight, an inspector’s new requirement. The solutions have names in the workforce’s vernacular. Skill mix is diluted, the registered nurse supervising more assistants across more beds. Bank and agency staff fill the gaps at premium hourly cost and zero continuity. The experienced worker’s informal knowledge, which resident must be watched after medication changes, which child’s silence means trouble, is treated as a free good that the rota can spend without replenishing, until the day two experienced staff leave in the same month and the ward discovers that its real operating system walked out with them, unrecorded in the handover notes because the handover notes were designed for facts, and this knowledge is judgement.

Churn is usually discussed as a recruitment cost, a matter of advertising and induction. The deeper cost is epistemic. Care of particular people accumulates a knowledge stock that is the service’s actual quality, and the stock is held in persons, not files. A sector that churns its workforce at the rates low wages guarantee is continuously liquidating its quality capital and reporting the liquidation as flexibility. The recipients experience it as the stranger who does not know how the leg ulcer is dressed, the new nursery worker who has not learned that the quiet child is the one to watch. Managers experience it as the impossibility of planning. Owners experience it, if the contract pays by the bed or the place regardless, not at all, which is why the pressure table insists on asking where costs land: the liquidation of knowledge is a cost, and it lands on the recipient and the remaining staff, the two parties with no seat at the contract’s signing.

This is also where the analysis parts company with a purely monetary radicalism. It would be possible to fund the sector lavishly and still organise it badly: churn the staff, fragment the visits, grade the work by its visible tasks, and the human result would disappoint the budget’s generosity. The framework’s categories, reproduction, the labour process, the pricing of skill, direct attention to the organisation of the hour and not only to its price, which is why the tradition’s best writing on care reads like an ethnography of shifts and rotas rather than a treatise on value. The money is necessary. The money is not the whole of it, and a reader who finishes this article believing only that carers should be paid more has grasped the conclusion and missed the subject.

Technology at the bedside: what the machines can and cannot carry

Every generation of the sector’s reformers arrives carrying a device, and the devices deserve a fairer hearing than either the enthusiasts or the romantics give them. Monitoring technology genuinely substitutes for one kind of caring labour: the watching. A sensor that alerts a distant centre when a frail person falls or fails to rise can replace the night attendant’s rounds for the specific function of detection, and for some recipients the replacement is a liberation, privacy restored, the home their own again. Records systems genuinely spare duplicated writing and lost information. Lifting equipment genuinely saves the backs whose destruction was once an occupational certainty. A serious analysis concedes all of this without flinching, because the trap was never a claim that nothing in care can be mechanised. It is a claim about the core: the response to the fall, the conversation that detects the depression behind the silence, the hand that persuades the frightened patient to eat. Detection scales. Response does not, or scales only at the price of becoming the thin, scripted contact that recipients describe, in the literature’s saddest recurring phrase, as being processed.

The newer wave of managerial technology cuts both ways in a pattern the workforce recognises instantly. The same digital systems that can coordinate a domiciliary round humanely, matching workers to nearby clients and protecting travel time, are used in the marketised segment to compress the round to its theoretical minimum, paying by the minute of contact and converting the worker’s journey, her thinking time, her buffer against the unexpected, into unpaid interstices. The technology is neutral in the brochure and directional in the contract: it optimises whatever the purchaser has specified, and where the specification is minutes per task, it optimises the humanity out with great efficiency and complete deniability. Families meet the same systems from the other side, as the administration of care migrates onto apps and portals whose operation becomes a new layer of unpaid household labour, the booking, uploading, chasing, and authorising that digital-by-default services transfer to the user. The work does not vanish. It changes rows in the table, from the salaried administrator to the daughter with a phone, and the transfer is recorded, if at all, as a saving.

Robotics, the horizon exhibit, should be discussed with the same discipline. Machines that lift, carry, remind, and entertain exist and will improve, and in societies whose shortage of young workers is acute they will be adopted at scale for the functions they can perform. What the research on their early deployment finds is instructive: facilities adopting robots do not reliably reduce their human staffing, because the machines generate their own work of supervision, maintenance, and rescue, and because recipients treat the robot as an addition to the human relationship rather than a substitute for it, asking, reasonably, where the person is. The likely future is the past’s pattern extended: technology colonises the edges of care, the lifting, watching, reminding, recording, and the attended hour at the centre remains human, relatively more expensive, and more visibly the point of the whole enterprise. The trap does not fear the robot. The trap predicts the robot’s job description.

Care across the cycle: the sector that grows in slump and boom alike

The care economy has a macroeconomic peculiarity that sets it apart from the industries political economy usually studies: its demand does not fall in recessions. Babies are born, the old grow frail, illness keeps its own calendar, and the state’s provision, where it exists, continues; indeed the sector is routinely listed among the recession-proof employers, and its employment grows through downturns that empty factories and offices. Growth through slump sounds like security, and for the individual worker it partly is, but the aggregate pattern has a darker implication the framework brings out. A sector that expands in all weathers, whose costs rise structurally, and whose funding depends on the tax base generated by the cyclical sectors, becomes a rising claim on a fluctuating revenue, which guarantees that the fiscal fights described earlier intensify precisely when the economy is weakest. The care budget is then cut in the slump, its work shed to households at exactly the moment households are least able to absorb more, and restored in the boom with the lags of politics, a ratchet that the austerity section’s mechanism converts into lengthening unpaid days for the same population, cycle after cycle.

Demography gives the sector its second peculiarity: the predictability of its demand. Most industries guess at their future markets. The care sector’s future clients are already born, already ageing along trajectories actuarial science handles confidently, and the broad shape of need twenty years out is among the most knowable quantities in social policy. That the provision for a foreseeable need is nevertheless chronically improvised is therefore a political fact of a specific kind: not uncertainty mismanaged, but a known bill perpetually deferred because the deferral’s costs land on the parties this article has kept naming, households, women, migrants, the old themselves, whose protests arrive slowly and unorganised. The framework’s contribution to the demographic debate is to refuse the framing of ageing as a surprise visited upon innocent policymakers, and to ask instead who profits, in quiet budgetary relief, from treating the most predictable expenditure in the state’s portfolio as an emergency each time it presents itself.

Burnout and moral injury: the worker’s body as the final buffer

Every system of buffers ends somewhere, and in the care system the final buffer is the worker’s own body and conscience. When the ratio is diluted past the point where adequate care is possible, the shortfall does not present as an accounting entry. It presents as a nurse deciding which call bell to answer slowly, a home carer skipping the conversation that is the only human contact in a client’s day, a nursery worker too depleted to notice the quiet child. The workforce has a vocabulary for the sequelae: burnout for the exhaustion, and, borrowed from military medicine, moral injury for the specific wound of being prevented by the organisation of one’s work from giving the care one knows is owed. The borrowing is apt. The injury is not fatigue alone; it is complicity enforced by rota, the daily experience of delivering a service one would not accept for one’s own parent, and knowing exactly what has been traded away and by whom.

The management literature processes these injuries as individual resilience failures and prescribes mindfulness, and the structural analysis should be precise about why the prescription is an insult in this specific workplace. Resilience training addressed to a worker whose injury is a staffing ratio is a technology for adapting people to the intolerable while leaving its arithmetic untouched, and it completes the circuit the ideology of vocation began: first the work is priced as devotion, then the damage devotion does is treated as a private weakness of the devotee. None of this denies that individuals differ in hardiness or that support for staff is worth having. It locates the variable. Put the same workforce on humane ratios and the injuries recede in the pattern of an occupational exposure being removed; the studies of magnet hospitals and well-staffed services in the nursing literature run consistently in that direction. The worker’s psyche is the last place the cost lands, and the surest sign that every other buffer, budget, household, chain, and ratio, has already been spent.

Collective action in a sector designed to prevent it

The obstacles to organising this workforce read like a design specification. The workers are scattered across thousands of small workplaces and private homes, where no gate exists at which to gather. Their hours are fragmented, their turnover high, their employers multiple and changing with each contract round. Above all, their leverage is booby-trapped by the nature of the service: the strike weapon works by withdrawing labour, and withdrawing care harms the cared-for immediately and visibly, which is why carers’ disputes are fought under a moral scrutiny no other workforce faces and why the workforce itself is divided about using the weapon at all. Employers and governments have not been slow to convert the workers’ own ethics into a restraint more reliable than law.

And yet the sector’s history of organisation is not a blank. Where care workers have won, the victories share a shape the analysis can read. They have organised across the fragmentation by organising the funding stream instead of the workplace, taking the demand to the public purchaser who sets the fee that sets the wage. They have framed the claim as the recipients’ claim as well as their own, staffing and pay argued as a single question of what a visit or a ratio shall be, recruiting families and disabled people’s organisations as allies rather than treating them as the opposition across a bargaining table. They have built, in the domestic work sector, international instruments recognising household employment as work like any other, a juridical beachhead won by a workforce once considered unorganisable in principle. Each victory confirms the structural analysis that predicted the obstacles: the fight is never only with the employer of record, because the employer of record is the prisoner of the same cost structure, and the real counterparty is whoever holds the budget from which the sector’s revenue flows. A unionism adequate to care has to bargain, in the end, with the treasury and the fee schedule, through whatever intermediaries stand in the way.

The lesson for the reader of this series is the general one this cluster keeps teaching in specific forms: the structure of a workforce’s exploitation dictates the structure of its possible resistance, and care’s resistance must be as relational as its labour. The strike fund matters less here than the alliance, the public argument about ratios, the insistence that the worker’s wage and the recipient’s dignity are one line item, not two.

The child as investment: what human capital language sees and misses

Mainstream economics does possess a vocabulary for the household’s product, and its arrival in policy debate is one of the interesting convergences this series watches for. Economists of education and development speak of human capital, of the early years as the period of highest return on investment, of childcare as infrastructure for the economy’s future productivity, and some of the most cited research in the field argues for early intervention in frankly pecuniary terms: spend on the youngest children and the returns arrive as earnings, taxes, and avoided costs of remediation. Marxist feminism reads this literature with a double response. The convergence is real and welcome where it funds nurseries. An argument that persuades treasuries is an argument with effects, whatever its metaphysics. But the divergence is instructive, because the human capital frame values the child as a future worker and the carer as an input into the child’s production function, and it therefore sees exactly what the national accounts see, the portion of reproduction that returns as taxable capacity, and misses the rest: the care of those who will never be workers, the dying, the profoundly disabled, the old whose accounts are closed. In the investment frame, eldercare is consumption and childcare is investment, and a policy logic built on the frame will fund the nursery and starve the hospice with a clear conscience. The reproduction frame, by contrast, starts from the maintenance of people as the economy’s purpose and precondition alike, and can say why the hospice hour and the nursery hour are the same kind of fact. The two languages sometimes vote together. They are not the same language, and the reader should know which one is speaking in any document that arrives claiming the authority of economics.

Deserts and distances: the geography of provision

Care has a geography, and it is not the geography of need. Providers cluster where fees are high and staff can be recruited, which is to say in prosperous districts whose residents could most easily purchase alternatives, and thin out where incomes are low, housing scattered, and travel between clients eats the working day. The rural home-care round is the limiting case: distances turn the visit model inside out, the carer’s paid time consumed by roads between houses, her effective ratio set by the map rather than by any manager’s decision, and the service’s viability hanging on cross-subsidies that marketised contracting, priced per visit, systematically removes. The result, described in the literature of several countries’ systems as care deserts, is the spatial form of the general rule: a market in an unprofitable necessity distributes itself by profitability, and the necessity remains where the market leaves.

Households fill the deserts as they fill every gap, with kin labour stretched over distance, the adult child managing a parent’s care by phone and weekend driving, a form of long-distance care work the research literature has begun to catalogue: the arranging, monitoring, crisis-flying, and guilt that constitute caring at a range. Its costs are real and almost entirely invisible to every ledger, since no visit is purchased and no diary category captures the management of a care package from another city. The geography point completes the pressure table’s implicit argument. The table’s rows are sites, and sites have locations, and the transfers between them have distances attached, so that a full accounting of the care system would be a map as much as a ledger: of where the old are, where their children went for work, where the migrant workers crossed, and where the services quietly withdrew to. The withdrawal is rarely announced as a decision. A contract not renewed, a home closed as unviable, a round discontinued: the map redraws itself by accumulation, and the households discover their new positions in it one crisis at a time.

The gatekeepers: assessment, eligibility, and the labour of rationing

Between need and provision stands a workforce this article has not yet honoured with attention: the assessors. Where care is publicly funded and budgets bind, somebody must convert a person’s frailty into a category, substantial, critical, eligible, and the conversion is performed by social workers and nurses whose professional labour consists substantially of rationing. Their position is the system’s contradiction in a single job description. Trained to identify need, employed to contain its cost, they operate the threshold at which the state’s promise meets the budget’s refusal, and they absorb, in person, the anger of families who experience the threshold as a betrayal by the human being across the desk. The assessment interview is where the abstract fiscal pressure of the earlier sections acquires a face and a clipboard.

Gatekeeping labour has its own pathology, visible in every rationed system: the threshold creeps. Eligibility defined at one level of need is administered, under budget pressure, at a higher one, without any announced change of policy, until families learn that the form must describe the worst day, that honesty about coping is penalised, that the system’s real criterion is not need but the demonstration of catastrophe narrowly averted. An informal curriculum in catastrophic self-presentation spreads through carers’ networks, and the assessors, who know the curriculum exists and know why, must pretend to evaluate the descriptions at face value. The charade is nobody’s fault and everybody’s working day. It is what a society’s refusal to fund its own standard of care looks like at the point of delivery: not a closed door, which would at least be a decision, but a door that opens only to those who can perform desperation fluently, staffed by professionals hired for their judgement and employed to suspend it. The structural analysis asks the reader to see the assessor neither as villain nor as victim but as the human location of a contradiction, the place where the productivity trap’s bill, deferred through every other buffer, is finally presented to a named individual in a small room.

The socialisation experiments: communal kitchens and their long echo

The Marxist tradition once tried to abolish the private household’s labour rather than reprice it, and the attempt belongs in this article because its failure instructs. In the revolutionary years after 1917, Alexandra Kollontai and the women’s department of the Russian party argued that the individual household was an economic anachronism, a million tiny workshops performing badly what collective provision could perform well, and that women’s emancipation required transferring cooking, laundry, and childcare to social institutions as a matter of design. Experimental communal houses were built with shared kitchens and nurseries; public laundries and canteens were opened as the infrastructure of a new daily life. The experiments mostly collapsed within years, under the weight of civil war, scarcity, and a population that often experienced the canteen’s soup as an insult added to hunger, and the programme was quietly shelved as the regime consolidated. The textile worker’s aphorism attributed to the period, that the communal pot had liberated nobody who still had to queue for it, captures the verdict of experience.

The easy lesson, that socialisation fails, is too easy, and the harder one is more useful. The experiments attempted to socialise the tasks before the society could fund the standard to which households aspired, so the collective provision was experienced as degradation rather than liberation: the laundry that returned shirts grey, the nursery too thinly staffed to be trusted. Socialisation is a funding commitment before it is an architectural one, a lesson the modern version of the proposal, universal services, has absorbed by arguing in terms of budgets and staffing rather than blueprints. What survives from Kollontai is the analytical point, still startling in its clarity: the private household is a unit of production, its inefficiency and its confinement are economic facts rather than domestic atmosphere, and any society that wishes to change women’s position must decide, consciously, which of the household’s functions it will collectivise, which it will marketise, and which it will leave in the kitchen with its eyes open about the cost of the leaving. The cluster’s later articles on the socialist states’ record take up what happened when such decisions were made at scale; here the point is that the menu was first written by the tradition’s own feminists, in the middle of a revolution, and has not been improved in its essentials since.

Paid in lodging and love: live-in care and the au pair margin

At the sector’s edge stand arrangements that evade the wage form almost entirely, and they repay attention because they show what the wage form conceals elsewhere. The live-in carer receives board, lodging, and a payment whose hourly meaning dissolves when the working day has no edge: she is at work in the sense that matters, available, all night and all day, and off duty only in the sense that she may sometimes close a door. Labour law in many systems has struggled with the category, carving exemptions for domestic employment that would be unthinkable in a factory, and the exemptions fossilise the old assumption that the household is not a workplace and its labour not quite labour. The au pair arrangement dresses a similar exchange in the language of cultural exchange and family membership, the young woman becoming one of the family for stipend purposes while remaining, for the distribution of the household’s unpleasant hours, remarkably like staff.

These margins matter beyond their numbers because they are the pure form of the sector’s founding exchange: caring traded against maintenance and belonging rather than against a wage that prices the hours. They reveal what the ordinary care wage still contains in diluted form, the presumption that part of the worker’s reward is the relationship itself, the home she lives in, the gratitude of the family, the sense of being needed. Employers in the formal sector make the same presumption in politer language when they describe the job’s satisfactions as a component of its compensation, and the presumption is not wholly false, which is what makes it effective. The satisfactions are real. They are also, in every other industry, considered the worker’s private business rather than a line in the employer’s accounts. Nobody proposes that a bricklayer’s pride in a well-built wall be netted against his wage. The netting is reserved for those whose work is love, and the reservation is the whole ideology in miniature.

Quality: the word everyone uses and the thing nobody can count

All parties to the care debate claim quality as their banner, and the term’s slipperiness is functional: it allows the funder, the provider, the inspector, and the family to sign the same document while meaning different things. Disassemble it and quality in care has at least four components that do not move together. There is safety, the absence of harm, falls, errors, infections, which is countable and therefore beloved of regulators. There is process quality, the following of correct procedure, documented and auditable. There is relational quality, the recipient’s experience of being known, respected, and attended to as a person, which is the substance of the service and the component least legible to any instrument. And there is continuity, the same faces across time, which underwrites the relational component and is destroyed by churn without registering in any safety statistic. A service can improve on the first two measures while degrading on the last two, and the history of inspection regimes contains melancholy stretches in which exactly that improvement was certified, the paperwork immaculate, the residents unvisited by anyone who knew their names.

Inspection itself becomes, under these conditions, another theatre of the cost structure. The inspectorate samples, the provider performs for the sample, the family’s testimony is sought through channels the frightened or loyal may not use, and the resulting grade enters the market as information for purchasers who have been shown, in the marketisation section, to be in the worst possible position to act on information. None of this argues against inspection, which catches horrors and disciplines the worst. It argues against the belief that inspection solves the trust problem that defines the commodity. The trust problem is not a deficit of information that better instruments will cure; it is a property of a service whose essence is a relationship between particular people over time, in private, where the vulnerable party may be unable to report. Societies handle such services, where they handle them well, by funding them generously enough that the pressures to degrade stay weak, staffing them stably enough that knowledge accumulates, and treating the workforce’s own testimony as the early-warning system it demonstrably is. Whistleblowing carers are the sector’s true inspectorate, and a system that prices their labour at the bottom of the economy while relying on their conscience for its safety has made a choice about which of its values is decorative.

The night: care’s claim on the whole clock

The argument about time has a frontier that deserves naming in its own right, because it is where the system’s demand becomes literally unlimited. Human need does not observe office hours, and a fraction of it, the confused wandering, the pain that peaks at three in the morning, the infant’s hunger, is specifically nocturnal. Somebody’s night is therefore the raw material of somebody else’s safety, and the disposition of nights is among the most honest measures of a care system’s justice. In the household, the night falls to the same person who carried the day, and sleep deprivation is the private tax of the family carer, levied without record. In the institution, the night is staffed thin on the theory that sleeping residents need less, a theory that holds until the night it does not, and the thinness is chosen by the same arithmetic that dilutes every other ratio. In the market for live-in care, the night is the grey zone where the wage form dissolves altogether, the worker sleeping in the house, half available, paid in the fiction that rest interrupted is rest.

Sleep is also where the care system’s costs return to it disguised as other budgets’ problems. The exhausted family carer’s own health deteriorates along pathways epidemiology can trace; the night worker’s body pays the premium that shift research has documented for decades. An economy that will not fund the night funds the consequences, in clinics and in claims, and files them under different headings so that the saving can stand. The point is worth making in this article’s own idiom: the night is the pressure table after dark, every row’s costs still accruing, every vessel still connected, and the person awake in it, paid or unpaid, known or stranger, is performing the oldest shift in the world for the people the day belongs to.

A note on method: why this article brackets the value question

Readers of the series will notice that an article on care and capitalism has proceeded without settling the question a certain kind of Marxist reaches for first: whether care labour produces surplus value. The bracketing is deliberate, and its rationale is worth stating because it is the rationale of the application genre itself. The wage pattern under examination exists in public services where no profit is extracted, in non-profit provision, in direct household employment, and in the marketised chains, and an explanation that required the surplus-value category to operate would fail in the settings where the pattern is most purely displayed. The structural account given here, cost disease, association with unpaid labour, the communicating vessels of the four sites, explains the pattern across all the ownership forms, including the ones where nobody’s capital is being valorised by the carer’s hour. Where the value question genuinely governs the analysis, in the taxonomy of productive and unproductive labour, the series treats it in the article on surplus value in modern firms and in the domestic labour debate for the household’s specific case, and the FAQ below answers the question in the form readers most often pose it. A framework is not weakened by declining to use every tool for every job. It is weakened by the opposite habit, and the habit’s name is scholasticism.

There is a corresponding bracket on the other side. The article has treated households as units in a system of transfers while knowing that households are also where people love, quarrel, forgive, and grow old together, and that no political economy of care worth reading can afford to lose the second knowledge inside the first. The tradition at its best held both, which is why its classics on the family are read as literature as well as theory. The bracket is methodological, not a judgement about what ultimately matters.

What a serious pay rise would do: following the money round

Because the remedies section confined itself to mapping positions, the reader is owed one piece of plain analysis: the mechanics of the remedy most often demanded, a substantial rise in carers’ pay, followed through the system with the article’s own tools. Begin with the objection as it is usually minuted: the cost. A sector employing a large share of the workforce, at low wages, is expensive to reprice, and the gross figure for a meaningful rise is the kind of number that ends meetings. Follow the money past the meeting, however, and the net figure shrinks by routes the framework predicts. A large fraction of the rise returns to the treasury directly, in tax on the higher wages and in reduced in-work benefit payments to a workforce whose poverty the state subsidises through the welfare system, which is to say that the taxpayer already pays part of the care wage bill and pays it in the degrading form of top-ups to full-time workers. Another fraction returns through retention: churn is expensive in recruitment, induction, agency premiums, and the quality losses this article has priced in harm, and a workforce that stays is cheaper per unit of adequate care than the gross wage comparison suggests. A third fraction circulates locally, since low-paid workers spend their incomes near home with a completeness that finance ministers usually celebrate in other contexts.

Against these returns stand the real risks, and the analysis is not a pamphlet, so they stand in the open. Where care is purchased from fees, a mandated rise passes into prices unless the funding follows it, and the marginal family is priced out into the informal market or back onto its own women, the vessels communicating as ever. Where provision is public, the rise is a permanent addition to the budget line most exposed to the cost disease, and a government that grants it without a funding settlement has merely scheduled the next ratio dilution. And a rise captured entirely by the credentialled tiers, leaving the unregistered majority behind, would reproduce the professionalisation trap at a higher level. The pay rise, in short, is a fiscal decision before it is a labour-market one, and the framework’s service is to show why: in a sector where the purchaser is substantially the state and the product cannot be cheapened, the wage question and the tax question are the same question wearing two hats, and politics that insists on answering them separately will answer both badly.

Why does public applause leave the wage untouched?

Because applause addresses the status injury, which is real but cheap to treat. Honours, ceremonies, and official gratitude cost a treasury nothing, while pay responds only to funding decisions. A workforce that accepts celebration in place of a settlement is paid in the currency the system prints for free, and the wage stays where the cost structure put it.

Fathers, leave, and the slow convergence

The counter-case section registered the direction of travel in men’s unpaid work: rising, from a low base, without converging. The research on the gap’s persistence is worth its own passage, because it identifies the mechanisms that a remedy would have to move. Gershuny’s lagged adaptation named the first: men’s domestic contribution responds to women’s employment with a delay measured in decades, as though the household’s division were a flywheel rather than a negotiation. Household bargaining research adds the second: the partner with the lower earnings rationally specialises in the home when a child arrives, and since the earnings gap precedes the child, the specialisation follows the market’s prior inequality into the household and then widens it, a ratchet that individual goodwill operates only against the gradient. The third mechanism is the design of leave itself. Leave offered to a household as a shared pool is taken overwhelmingly by the mother, because the ratchet says so; leave reserved for the father on a use-it-or-lose-it basis changes the household’s default and, in the systems that adopted the design, measurably changed the taking, which is the strongest evidence available that the division of care is an artefact of institutions rather than of nature. The position is presented here as the literature presents it, a design finding, not as this article’s prescription.

The professional culture of involved fatherhood supplies the counter-case’s most double-edged exhibit. The new father of the discourse, present at the birth, fluent in the child’s routines, is real, and any account that denied the generational change in men’s aspirations would be repeating the fifth misreading. Yet time diaries continue to show the asymmetry surviving inside the aspiration, concentrated in the categories this article has taught the reader to watch: fathers’ caring skews toward the schedulable, the playful, and the public, mothers’ toward the routine, the nocturnal, and the managerial, the remembering of vaccinations and the noticing of shoes outgrown. The division is no longer between a carer and a stranger. It is between two kinds of caring, one of which can be planned around a career and one of which is the planning. The framework’s point is not to indict the fathers, who are responding to the same wage structure and workplace expectations as everyone else, but to prevent the discourse of convergence from being cashed before the diaries confirm it. When the convergence arrives in the measurement, the framework will register a genuine transformation of the household form, and its categories, built to track transfers, will track that one too.

Priced out of the neighbourhood: housing, distance, and the key worker

Care is delivered in places, and the places where the need is richest are increasingly the places where the carer cannot afford to live. The affluent district generates the demand: ageing owners in valuable houses, dual-professional families buying childcare, private provision clustering where the fees are. Its housing market, meanwhile, excludes the workforce that the demand requires, so the carer commutes from the cheaper periphery to the expensive core, her working day lengthened by the journey at both ends, her wage silently taxed by the fare. The live-in arrangement returns under this pressure wearing modern dress: the carer housed in the spare room of the cared-for person’s home is partly a solution to the recipient’s needs and partly a solution to the worker’s housing costs, and the blurring of wage and shelter that the au pair margin displayed appears again at the centre of the metropolitan market.

The pattern is worth extracting because it shows the care wage being set by more systems than the care system. A wage that cannot house its earner within reach of her work is a wage that assumes she lives somewhere cheaper on somebody else’s subsidy, a partner’s wage, a family’s spare room, a sending country’s cost of living, and the assumption is one more form of the general rule that the sector runs on transfers it does not record. Urban policy meets the care question here whether it wishes to or not: the zoning decision, the transport fare, and the supply of modest housing near the hospitals and the nurseries are care policies in everything but name, and a city that wonders why its care homes cannot recruit while its planning system forbids the workforce an address within the borough is asking a question it has already answered.

The carer’s body: lifting, infection, and the occupational bill

Occupational health statistics tell the sector’s story in a dialect of backs and shoulders. Caring is heavy manual work conducted on loads that cannot be palletised: people, who shift, resist, fall, and must be caught, in bathrooms and bedrooms never designed as workplaces, frequently alone, at hours when assistance is thinnest. The injury rates for musculoskeletal damage among care workers stand, in the occupational literature, alongside those of construction and warehousing, with the difference that the caring load’s weight is a person’s dignity as well as a person’s mass, and the techniques that spare the lifter must also spare the lifted. Add the biological exposures of the work, infection as a routine condition rather than an industrial accident, the violence that confused or distressed recipients can direct at the nearest body, and the sleep destruction of the night economy described earlier, and the sector’s true compensation package comes into view: a low wage, plus a probability of leaving the workforce injured that the wage never priced.

The significance for the framework is the destination of the cost. The injured carer’s care is then required in her turn, delivered by the same system at the same ratios, and her lost earnings and pension accrual reappear as claims on the household and the state. Occupational injury in care is thus a mechanism by which the sector’s underfunding finances itself across time: save on the staffing that would have made the lift a two-person lift, and pay later, elsewhere, in the currency of another woman’s damaged spine and the benefits that keep her household afloat. The pattern is the austerity circuit internalised in a single body, and it is why the workforce’s own safety campaigns belong in the economic analysis and not in a footnote on welfare. A ratio is an ergonomic fact before it is a budget line.

The school as care infrastructure: the hours nobody counts

One institution performs so much care that its caring function has become invisible through familiarity: the school. For the length of the school day, the state minds the children of the working population, and the entire structure of dual-earner employment rests on that custody as surely as on any nursery. The function’s invisibility is exposed at its edges, in the mismatch between the school day and the working day, the holidays that no employment contract recognises, the sick child whose exclusion from the classroom transfers the care instantly and entirely to a parent whose employer experiences the transfer as an attendance problem. Feminist analysis of education policy has long treated the school as part of the care infrastructure, and the point sharpens the article’s general thesis: the boundary of the care system is not where the formal services end. It runs through every institution that holds people safely while their households work, and every one of those institutions, when its hours or staffing are cut, discharges its load into the same households through the same front doors.

Teachers occupy the instructive middle of the sector’s hierarchy: credentialled, publicly employed in the main, paid above the care floor and below the professions whose entry their qualifications resemble, and engaged in work whose caring content, pastoral, safeguarding, the noticing of the child who has changed, expands as other services retreat. The profession’s own complaints, of workload composed increasingly of the welfare functions shed by other agencies, are the pressure table’s transfers recorded in a staffroom idiom. When the children’s services are rationed, the school becomes the service of first resort because it is the one universal institution that sees every child, and the teacher becomes the assessor, the referrer, and the holder of the case in the interval, which is to say the years, before another agency’s threshold is met. The pattern is generalisable and worth stating as this article’s recurrent law: cuts to any row of the care system do not abolish the need; they promote the nearest universal institution into an unfunded provider, and its workforce into carers with other job titles, until that workforce’s own capacity tears and the need moves on again.

Dying, and the care that investment language cannot see

The human capital frame was shown earlier to value the nursery and discount the hospice, and the end of life is the place to test every framework this article has used, because dying strips care to its essence: labour whose entire product is the quality of a person’s remaining time, with no future earnings to justify it and no recovery to measure. The hospice movement built its philosophy on exactly that product, presence, pain controlled, fear attended, the family held inside the circle of concern rather than processed as visitors, and its historical achievement was to demonstrate that the worst-organised part of the care system could be reorganised around the relational core once a movement insisted on the core’s value. Its chronic funding position, dependent in many systems on charitable giving and volunteer labour to a degree no other branch of health provision tolerates, is the value system’s confession: the care of the dying is supported as a charity precisely because it cannot be justified as an investment, and the reproduction framework’s insistence that the maintenance of people is an end and not only a means is nowhere more plainly the difference between the two accounts.

Volunteering deserves its sentence here, because it is the fifth site no table in this article has rowed: the vast unpaid labour that flows not through kinship but through organised goodwill, the hospice volunteer, the hospital visitor, the drivers and sitters of a thousand small charities. Its political economy is genuinely double. It expresses a social solidarity that no wage system has yet manufactured and that a humane society would not wish to lose; and it functions, in the budget’s quiet arithmetic, as a subsidy that lets the funded system set its establishment below the level its own standards imply, with the volunteer’s devotion filling the margin in the manner of every other buffer in this analysis. The honest position holds both truths and declines to resolve them into either celebration or exposure. What the framework adds is the question it asks everywhere: what happens to the recipient on the day the buffer fails, and who, in the present arrangement, has been made responsible for the answer.

The queue: rationing by delay

Where price and eligibility do not ration care, time does. The waiting list is the care system’s most democratic instrument and its most deceptive, formally equal, substantively sorted by the capacity to endure, to chase, to pay for an interim, or to deteriorate in the politically useful way. Delay is a transfer like any other this article has tracked: the months spent waiting for an assessment, a place, an operation’s aftercare, are months in which the need does not wait, and the labour of meeting it is performed by the household at the household’s price. The queue therefore belongs in the pressure table’s logic as surely as the fee and the budget line, and its length is a measure of the system’s underfunding translated into the one currency, other people’s days, that treasuries are least required to count. Reformers who announce a service’s expansion while its queue lengthens are describing the stock and ignoring the flow, and the household waiting at the queue’s far end has a precise understanding of the difference, since it is supplying the flow from its own stock of hours in the meantime.

The theoretical root: Lise Vogel and the reproduction of labour power

Behind the applications of this article stands one theoretical construction that deserves naming in the application itself, because the application’s confidence derives from it. The American scholar Lise Vogel, in a work first published in the early nineteen eighties and recovered for a later generation by the social reproduction literature, proposed that the tradition’s scattered insights about housework, wages, and women’s position could be unified by a single category: the reproduction of labour power. On her account, the daily and generational renewal of the capacity to work is a material process with its own labour, its own sites, and its own necessary place in the circuit of capital, neither an external precondition nor a cultural residue. Women’s oppression, in the strong version of the thesis, is rooted in their assigned position within that process under class society: the sex that bears the children and is made responsible for the renewal work is subordinated not by ideology alone but by the organisation of a material necessity that capital requires and does not administer directly.

The construction’s value for an application like the present one is that it disciplines the claims. It explains why the household’s labour is simultaneously indispensable and invisible, why the state is drawn into reproduction’s management despite its reluctance, why the wage of the paid carer is shadowed by the unpaid labour of the household, and why changes in household form redistribute rather than abolish the work: each of the article’s phenomena is the reproduction process viewed from a different site. It also marks the framework’s limits honestly, since the unification has critics within the tradition who find the category stretched, the bearing of children and the washing of dishes gathered under one term at some cost in precision. The objection echoes the one recorded for the cluster’s framework article and is noted here for the same reason: the strongest version of a theory is the one that has priced its own weaknesses. What the application needs from the theory is not unanimity but the transfer mechanism and the invisibility thesis, and those survive the internal criticism in working order.

A class fraction in formation: the care workforce considered politically

Step back from the bedside and the ward and consider the care workforce as a political actor in formation, because the framework’s final question about any labour force is what its position disposes it to do. The workforce is enormous, feminised, disproportionately migrant, and employed at the point where the state’s fiscal contradictions meet the population’s most intimate needs, which gives it a latent leverage out of proportion to its organisation: a society can postpone a delivery, but it cannot postpone the night shift. Its grievances are simultaneously industrial and domestic, the wage at work and the load at home being, as the double day section showed, frequently the same woman’s two fronts. Its natural allies, the recipients and their families, are the whole population at the points of greatest vulnerability, an alliance of a breadth most labour movements only dream of, and correspondingly difficult to assemble, since the allies meet the workforce across the counter of a rationed service and are taught by the rationing to see each other as rivals for the scarce hour.

Political traditions have noticed the fraction at intervals and mostly mishandled it, courting its votes with the applause this article has anatomised, or addressing it as a clientele of the welfare state rather than as the welfare state’s producers. The Marxist feminist tradition’s wager, from the wages campaign onward, has been that the workforce’s self-organisation changes the question itself: once the people who perform reproduction name it as labour and organise as labour, the invisibility on which the whole cost structure quietly depends is withdrawn, and every budget in the system must be rewritten in the knowledge that the subsidy has a voice. The wager is unproven at scale and contradicted by no decisive experiment, which is the honest status of most consequential political hypotheses. What the analysis of this article can say is narrower: the structural conditions for the workforce’s importance grow stronger with every year of demographic change, its indispensability rising with the dependency it serves, and a political economy that has learned to follow the transfers will not be surprised, when the fraction moves, at the size of the ground it moves on.

Young carers: the workforce that is still at school

At the very bottom of the care system’s hierarchy of visibility stands a workforce that no table in this article has rowed, because its members are children. In households where a parent’s illness, disability, or addiction consumes the family’s stock of adult capacity, the caring devolves onto sons and daughters, sometimes of primary school age: medication prompted, meals cooked, siblings dressed, a parent’s moods monitored with the hypervigilance that children develop in houses where the weather is set by an adult’s condition. Research on young carers, a field built largely by the sociologists who first counted them against official disbelief, documents the pattern’s signature costs: schooling interrupted not by absence alone but by exhaustion and preoccupation, friendships rationed by the impossibility of inviting anyone home, a childhood organised around responsibilities the surrounding institutions prefer not to see, because seeing them would oblige the institutions to act.

The young carer is the pressure table’s reductio. Every transfer this article has traced, from state to market to household, assumed the household contained an adult to receive the load, and where that assumption fails the load does not float; it settles on the smallest shoulders in the building. The case disciplines the analysis in two directions. It shows the household buffer’s true elasticity, which is not the elasticity of a resource but the desperation of a family, absorbing into a child’s development what no budget would fund. And it shows why the system’s invisibility is functional all the way down: a society that counted its young carers honestly would have to record, in its own statistics, that a portion of its care provision is performed by minors, at the price of their schooling, in the years when the state simultaneously compels that schooling as the foundation of their futures. The two facts cannot be held in one ledger without embarrassment, so the ledger keeps them apart, the school recording the absence, the clinic recording the parent’s needs, and nobody recording the child who connects them.

Support, where it exists, arrives in the idiom of the whole system: assessment, threshold, and the performance of crisis, with the added cruelty that the young carer’s competence, the very maturity that adults praise in her, is used as evidence that the arrangement is working. Maturity in a ten-year-old is not a solution to observe; it is a cost that has learned to present as a virtue, and the research literature’s slow insistence on the point, against the sentimental reading of the resilient child, is of a piece with this article’s treatment of devotion everywhere else. Wherever the care system is praised for the extraordinary qualities of those who hold it up at no charge, the praise should be audited for the transfer it is concealing.

When households decline to reproduce: fertility as the system’s verdict

The reproduction framework ends, logically, at reproduction itself, and the wealthy economies’ fertility patterns pose its sharpest long-run question. Where the costs of children are privatised, in housing, in forgone earnings, in the time famine this article has anatomised, and the supports are thin, households respond in the only register fully within their control: they have fewer children, later, or none. Demographers debate the weights of the causes, values, housing, labour market insecurity, the price of care among them, and the debate is recorded here without a verdict on the weights. What the framework contributes is the observation that the response, whatever its motives in individual lives, functions at the level of the system as a slow withdrawal of the very labour of reproduction, a reduction in the future supply of workers, carers, and taxpayers on which every projection of the care economy silently depends.

A strand of commentary names this withdrawal a strike, and the metaphor should be handled with care, since most people who forgo children are not conducting politics but conducting their lives under constraint. Yet the metaphor catches something the neutral language of preference conceals: that childbearing in the present arrangement is an act of unpaid production for a collective benefit, the next generation of everyone’s pension contributors and everyone’s nurses, performed at private cost by specific households, and that a rational response to a bad bargain is to decline it. The system’s answers reveal its priorities in sequence. It exhorts. It offers small payments that acknowledge the principle while trivialising the sum. It imports the grown children of other societies through the care chains and the wider migration system, externalising the cost of the raising as this article has shown. And it discovers, in each ageing budget, that the bill for reproduction deferred is larger than the bill for reproduction supported, arriving with the same inevitability as the cost disease itself, because the two are cousins: both are what happens when an economy treats the maintenance of people as a free good and then meets, in time, the invoice.

The circle closes where the article began. Care work is low-paid because the economy assumes it will be done anyway, out of love, duty, and necessity, by people with no alternative purchaser for their devotion. The assumption holds until it does not, and the signs of its failing are not dramatic: a vacancy rate that will not close, a maternity ward’s worth of unborn statisticians’ projections, a daughter’s decision, made quietly and repeated millions of times, about how much of her life the arrangement may have. The framework’s final service is to connect those quiet decisions to the payslip with which the analysis opened, and to insist that they are one phenomenon, the price of care asserting itself at last in the only currency the system could not devalue: its own future supply of people.

Before the welfare state: orders, charities, and the inheritance of devotion

The wage pattern’s roots run deeper than the factory system, and a historical glance explains why the ideology of devotion clings to this sector with a tenacity it shows nowhere else. Before states funded care, the care of strangers was organised substantially through religious vocation: nursing orders, monastic infirmaries, charitable sisterhoods whose members were maintained at subsistence by their institutions precisely because their labour was defined as a calling rather than a trade. The arrangement was not hypocrisy; it was a coherent economy in which the carer’s keep, her community, and her salvation were the compensation, and money wages were beside the point by design. When the modern hospital and the modern state inherited the work, they inherited the workforce’s founding bargain as a cultural default: the expectation that caring labour comes with its own reward built in, an expectation no other branch of the public service had to overcome, since no one supposed that a road-mender’s soul was the road’s true payment.

The charity sector carried the same inheritance into the voluntary provision that still surrounds the formal system, and its traces are legible in arrangements a purely economic eye finds puzzling: the hospice funded by appeals while the ward beside it is funded by taxation, the prestige that attaches to unpaid service on a charity’s board and the obscurity that attaches to the paid coordinator who makes the service exist. Historians of welfare describe the long transition as a mixed economy’s slow sorting, functions migrating from church and charity to state and market at different speeds in different places, leaving the map of provision a palimpsest in which a district’s services may still bear the shape of a Victorian endowment. For the analysis of wages the lesson is the depth of the discount’s roots. The pricing of care as devotion is not a recent managerial trick. It is the survival, inside a wage economy, of an older economy of vocation, and it survives because it remains useful: every era’s funder discovers, with an air of innocence, that the people doing this work would largely do it for less than a living, having been formed by institutions that taught them to call the doing its own reward. The tradition this cluster represents has always insisted that such formations are historical products rather than facts of the soul, and the history of the orders is its Exhibit A: devotion, too, has a labour process, a recruitment system, and a cost structure, and it can therefore be analysed, and its modern secular echo can be priced.

That echo reaches the present workforce in a form worth one more paragraph, because it explains a feature of the sector’s politics that puzzles outsiders: the workforce’s frequent reluctance to press claims in the language of self-interest, its preference for framing every demand as being on behalf of the cared-for. Partly this is tactics, learned in the booby-trapped bargaining described earlier. Partly it is the inheritance working as designed, a formation in which the carer’s own needs are the last legitimate topic. The framework’s respect for the workforce consists, in part, in refusing to mistake that formation for the workforce’s nature, and in hearing the claim made on behalf of the residents as what it also is: a wage claim wearing the only clothes in which this sector permits a wage claim to appear in public. When carers say the ratio is unsafe for the residents, they are telling the truth, and they are also, in the same sentence, describing their own working lives, and a political economy worthy of its subject holds the two truths in one hearing.

The basement trade: domestic cleaning and the care system’s unmarked floor

Below the recognised care sector lies a trade that shares its economics and escapes its name: paid domestic cleaning and household help. The cleaner’s work is the maintenance of the domestic environment rather than of persons, and for that reason it falls outside most definitions of care, including, in its occupational statistics, the one this article has used. Yet the household that employs her experiences no boundary: the same afternoon’s labour may run from scrubbing a bathroom to sitting with a convalescent while his wife shops, and the labour market that supplies her is the same graded, feminised, heavily migrant market, with the same presumption that the skill walks in with the woman and the same pricing of reliability and trust as unpayable virtues. Sociologists of domestic work have treated the trade as the care system’s basement, the floor on which the whole structure’s labour relations appear in their least regulated form: cash payments, no contracts, isolation in private homes, and an intimacy of access, to the household’s dirt, secrets, and standards, that makes the employment relation personal in ways that protect and expose the worker in the same gesture.

The basement matters to the analysis for what it reveals about the floors above. Every mechanism this article has described operates there without mitigation: the productivity question is farcical, since the cleaner’s hour is as incompressible as the carer’s and her employers know to the minute what an hour buys; the association with unpaid housewifery is total, since the trade is quite literally the purchase of a housewife’s task list; the chains run through it, the migrant cleaner in the wealthy suburb remitting to the household whose own floors are cleaned by a poorer neighbour. And the trade’s informality is not a pre-modern residue awaiting formalisation’s arrival. It is actively reproduced by the purchasers’ preference for arrangements without paperwork and by tax and immigration systems whose enforcement stops politely at the front door. The private household is the last workplace the inspectorate does not enter, and the analysis should register why: the state’s reluctance to police the home is the obverse of its reliance on the home, the same reluctance that keeps the household’s unpaid labour unpriced extending its protection to the paid labour the household buys. Follow the reluctance and the whole architecture stands revealed as one design: a zone of life defined as private, inside which labour may be extracted, waged or unwaged, under exemptions from the norms that govern every other workplace, because the economy’s reproduction has been assigned to that zone and the assignment must not be audited too closely.

Reform literatures periodically propose to formalise the basement, with voucher systems that subsidise declared domestic employment and pull the trade into social insurance. The experiments are instructive whatever a reader thinks of them, because their uptake patterns show the purchasers’ true demand: formalisation succeeds where the subsidy makes the declared hour cheaper than the undeclared one, and stalls where it does not, demonstrating that the informality was never a cultural lag but a price. The cleaner herself, offered the choice, divides according to her position in the chains: the documented worker gains protection she can use, the undocumented one loses the only market open to her, which is why the trade’s politics, like the sector’s everywhere, refuses the simple alignment of virtue with regulation. The framework does not resolve the dilemma. It locates it, in the gradient of security that runs through every row of the care system, from the registered nurse at the table’s top to the woman with the bucket whom the table, like the economy, nearly forgot to count.

A reader’s audit: applying the table at home

The genre of application licenses one direct exercise before the close. Take the pressure table and audit a single household known to the reader, which in practice means the reader’s own. List the caring it consumes: the children minded, the elders watched, the illnesses nursed, the domestic order maintained, and beside each item write who performs it, under which row of the table, and at whose expense of earnings, sleep, or health. Then ask the audit’s three questions. Which of these arrangements would survive the withdrawal of the one person whose labour is simply assumed? Which are priced, and does the price resemble the value the household would set on the work if it had to bid for it against strangers? And where, in the household’s own accounts, in its conversation about money, does the largest item of all, the unpaid item, appear at all? Most readers who perform the exercise honestly discover that their household, like the economy it belongs to, runs on a subsidy it does not record, supplied by a person it loves, under an arrangement it has never consciously chosen. That discovery, replicated across enough households, is what the tradition means by making the invisible visible, and it is why an article about wage theory ends in the reader’s kitchen. The structures of political economy are not elsewhere. They are the terms on which the reader’s own people will be cared for when they are frail, and on which the reader, in turn, will be asked to care, and the analysis has done its work if those terms can never again be mistaken for the weather.

The audit has a forward-looking page as well. Every household in the reader’s acquaintance is a small care system with a demography of its own: its members will age along paths that are broadly predictable, its children will leave along paths that are proudly encouraged, and the distance between the two paths is where the household’s future care will have to be found. Households that have run the audit tend to make different arrangements from those that have not: they talk about money and frailty earlier, they treat the carer’s pension and health as household infrastructure rather than private luck, they notice which sibling is being silently appointed by geography and default, and they sometimes, not always, redistribute the appointment while redistribution is still a choice rather than a grievance. None of this is politics in the grand sense, and all of it is the material from which a politics of care would have to be built, since the grand sense is only the sum of arrangements like these, made visible to one another. The framework that began with a payslip ends, fittingly, with a conversation it cannot have for the reader, in a kitchen it cannot enter, about a bill that is coming due in every house on the street.

The bill, presented honestly

The argument can now be gathered without being shrunk. Care work, paid and unpaid, is the labour of maintaining people, and a capitalist economy depends on it as a condition of production while organising its cost off the books wherever the books can be arranged to exclude it. The paid sector’s wages are low for two compounding reasons: because the work’s skill was trained unpaid in households and is priced as nature, and because the work sits in the productivity trap, unable to cheapen its core hour without destroying its product, so that every funding stream attached to it, fee, budget, or family income, is under permanent and rising strain. The strain does not vanish when it is moved. It travels through the four sites of the pressure table, household, public service, market, and migrant chain, and it lands, in every generation and under every policy fashion, disproportionately on women, and among women disproportionately on the poorest and the least secure. The household form that the classic analysis assumed has dissolved, and the framework survives its dissolution because its object was the transfer, not the form. What has not dissolved is the necessity the form once organised, now larger, longer-lived, and more clinically complex than at any point in the tradition’s history.

A reader who accepts the analysis is not thereby committed to any remedy in the map this article drew, and the article has been careful to keep its diagnoses and its sympathies from masquerading as one another. But the analysis does foreclose one position, the position that the present arrangement is natural, efficient, or free. It is none of the three. It is a specific distribution of a real and rising cost onto specific and identifiable people, maintained by an accounting convention that declines to see them, and it persists not because no alternative can be imagined but because the alternatives all begin by presenting the bill to parties who do not pay it now. Every serious politics of care is an argument about the bill’s destination. The least this article can do, and the most the genre of application permits, is to make sure the bill is itemised, addressed, and no longer deliverable to the old address by default: the unpaid day of the woman nearest the need, whoever she is, in whichever country of the chain she stands, at whichever hour of the night the need arrives.

For students and exam candidates

The examinable core of this article is the compound explanation of low pay, and answers should reproduce both halves and their division of labour: the first half, the pricing of household-trained skill as nature, explains the cheapness at the point of hiring; the productivity trap accounts for why that cheapness persists across ownership forms and deepens over time. Quote the care pressure table as your organisational device, since its four rows let you discuss households, public provision, markets, and migration as one system of transfers rather than four topics. The strongest objection to rehearse is the household-change objection, and the model treatment is the one given above: concede the ethnography, defend the transfer mechanism, and note that the framework predicted the growth of the paid sector and the chains. Carry the topic into the framework article, social reproduction theory explained, for the theory this application presupposes, and use the VaultBook study notes prepared for this series to build a card for the pressure table, since the table is the artifact an examiner can ask you to reconstruct. Avoid the two losing moves: the purely attitudinal answer, which cannot explain persistence, and the recitation of figures, which the measurement section has shown to be convention-laden and which this series’ protocol forbids.

For teachers and seminar leaders

The productivity trap is the most teachable idea in the article, because it converts a moral commonplace, society does not value carers, into a mechanism students can test against cases they know: the quartet that cannot be halved, the nursery ratio, the relative price of a haircut against a manufactured good across their own lifetimes. Stage the seminar as a contest between the attitudinal and structural accounts and make each side specify what evidence would change its mind; the falsification passage is written to be lifted into that exercise. The counter-case section supplies the discipline’s necessary discomfort, and students should be pushed to state the household-change objection in its strongest form before being shown the transfer answer, since the answer is only impressive against the real objection. The misreadings catalogue works well as a closing diagnostic: give the seminar a piece of journalism on the sector and let them find which of the five errors it commits, an exercise that reliably produces the recognition that the errors are not carelessness but the default settings of the debate.

For researchers, historians, and journalists

For researchers, the article’s open files are the measurement problems and the transfer mechanisms: time-use methodology and its undercounting of supervisory care, the convention-dependence of every aggregate figure in the field, the empirical career of the care penalty across labour markets, and the still-thin evidence on what staffing ratios do to relational quality as distinct from safety. Historians will find the live questions in the transitions: the construction and dissolution of the breadwinner settlement, the socialisation experiments and their reception by the populations they served, the professionalisation campaigns and the tiers they created. Journalists are pointed, as the persona path intends, to the remedies map and to the fiscal mechanism, since the sector’s stories become intelligible when the reporter follows the cost rather than the announcement: ask of every reform which row of the pressure table it relieves and which it burdens, of every shortage what wage would end it, and of every saving where the work went. The series treatment of marxist state theory today supplies the general theory of the fiscal seam on which care provision sits, and pairs with this application for background on any story where a budget line becomes a family’s evening.

Between kin and market: neighbours, mutual aid, and the informal middle

The pressure table’s four rows do not exhaust the map, because between the household and every formal system lies a penumbra of neighbourliness, friendship networks, congregations, and mutual aid that carries a real share of the total caring in any community. The neighbour who watches a child for an hour, the friend who sits with a dying man so his wife can sleep, the network that organises meal rotas after a birth: this labour is unpaid and unchosen by any institution, given in the currency of regard, and it is the part of the system most often romanticised by commentators who wish to argue that communities, left alone, will provide. The framework’s treatment is the one it applies to the grandmother: the giving is real, its distribution is patterned, and its reliability as infrastructure is the question. Mutual aid flourishes where people have time, proximity, and stability, which is to say it flourishes least where wages are lowest, hours longest, and housing most transient, so that the communities with the thinnest formal provision frequently have the thinnest informal provision too, the two deficits sharing a cause in the exhaustion of the same population. A politics that prescribes community as the answer to the care deficit is prescribing the scarcest resource in the deficit’s own heartland.

The informal middle also has a shadow side the romantic account omits: it runs on social credit, and those without credit, the newcomer, the stigmatised, the family whose troubles are judged rather than pitied, draw on empty accounts. Formal provision, for all its rationing and gatekeeping, at least entitles the unlovable. The historical function of the public services in this respect is underappreciated in the literature of community celebration: the clinic and the home help did not only supply labour, they supplied impartiality, a floor beneath which a person’s care did not depend on being liked. The mature position, and the one the table’s architecture implies, treats the informal middle as a good to be supported with time and money rather than a substitute for the funded rows: give neighbours the hours and the respite that let neighbourliness survive, and it will; build a system that presumes it, and it will perform exactly as the household performed under the same presumption, by consuming its most devoted members first.

The language of care: how words set the wage

Attend, finally, to the vocabulary in which the sector is discussed, because the pricing of the work begins in the words available for it. The language of care is saturated with the domestic and the devotional: angels, saints, heroes, the work described as a labour of love, its performance as dedication, its practitioners as having a gift. Each term performs the same quiet operation, relocating the activity from the economy to the heart, where payment becomes almost an embarrassment, a category error, as though money would soil what love sanctifies. Compare the vocabulary of occupations whose pay is not in question: nobody calls a pilot’s skill a gift, or suggests that a surgeon’s dedication should be netted against the salary, and the absence of devotional language around those professions is not an accident of their prestige but a component of it. Professions are paid in the currency of competence; vocations are paid in the currency of meaning, and the exchange rate between the currencies is set by the party doing the paying.

The workforce’s own linguistic counter-moves repay study. The insistence on job titles that name tasks rather than virtues, the union sloganry that translates devotion back into hours, the professional bodies’ campaigns against the angel imagery in recruitment material, each is a small act of repricing conducted in the only medium where the original pricing was done. Even the word care itself has been fought over, the disability movement’s preference for assistance being the sharpest case, since the word’s warmth can function as an anaesthetic for the power relations inside the relationship it names. A reader trained by this article will hear the vocabulary of any care debate as evidence: when the speeches fill with angels, check the budget line; when the minister reaches for the language of love, ask after the ratio. The words are not decoration on the structure. They are the structure’s public interface, and they are maintained, like everything else in this analysis, because they are useful to the arrangement they describe so tenderly.

A last distinction: caring about, caring for, and the division between them

One distinction from the ethics of care literature organises much of what has been said and earns a closing statement. The philosopher Berenice Fisher and the political scientist Joan Tronto, in a formulation the field has adopted, separated caring about, the noticing of need and the assumption of responsibility in the abstract, from caring for, the hands-on work of meeting it, with further phases for the giving and receiving of the care itself. The division of the phases across a society is a map of its power. Those who care about, who fund, commission, inspect, and empathise, are positioned high in every hierarchy this article has surveyed; those who care for are positioned low, and the two groups are divided, with statistical regularity, by sex, by class, and by the borders of the care chains. Tronto’s name for the advantaged position, privileged irresponsibility, the capacity to care about while being excused from caring for, is the precise term for the arrangement the whole system rests on: the many are excused, the few are assigned, and the excuse is dressed as the natural order of things.

Tronto’s later work on caring democracy draws the political conclusion the application genre may report without endorsing as its own: that a society’s quality can be judged by how it distributes the phases, and that democracy itself is incomplete while the hands-on phase is assigned by inherited hierarchy rather than shared as a common obligation. Set beside the Marxist account, the ethical framework and the material one illuminate each other’s blind spots. The ethicists see the moral texture the economists miss, the meaning of attentiveness as a human good; the materialists see the cost structure that decides whose attentiveness will be extracted and at what price. The reader who holds both will understand the sector better than the reader of either alone, which is the note on which a long application should end: the phenomenon is one, the lamps are several, and the wage with which this article began is low in every light.

Frequently Asked Questions

Q: Why is care work so low-paid?

Two mechanisms compound. First, the competence the job requires is largely acquired without wages inside households, so employers can hire trained ability while paying for what looks like an untrained disposition. Second, the activity itself cannot be made much more productive per hour, because shortening the hour of attendance degrades the service, so the sector lacks the growing fund out of which other industries finance rising wages. The first mechanism sets the wage low at the point of hiring. The second keeps it low against every subsequent pressure, under public and private ownership alike, which is why the pattern survives changes of government and of management fashion.

Q: Is care work productive for capital?

In the strict taxonomy of the tradition, the answer depends on where the labour is exchanged and what it produces, and that taxonomy is treated in the series article devoted to productive and unproductive labour. What can be said without the taxonomy is functional: caring labour, paid and unpaid, maintains and replenishes the workforce that every employer draws on, so capital as a system depends on its continuous performance whether or not any individual carer’s hour yields a profit for her employer. A hospital cleaner employed by a contractor, a nanny employed by a family, and a mother employed by nobody stand in different relations to the value form while doing continuous work on the same human material.

Q: What is the care penalty?

The term names a finding from comparative earnings research: occupations whose content is face-to-face caring pay less than occupations rated at similar levels of skill, education, and effort whose content is not caring. The penalty is defined against comparison, not against a feeling of unfairness, and it persists in the studies after the usual explanations, hours, credentials, workforce composition, have been accounted for. Researchers dispute its size and its decomposition, how much is cost structure, how much the gender coding of the work, and this article follows the series protocol in reporting the direction rather than a figure, since the estimates move with the methods used to produce them.

Q: Can care work be fully commodified?

Substantially, but not completely, and the residue is instructive. Markets already organise a large and growing share of caring, and they handle its tasks, its scheduling, and its logistics after a fashion. What resists full commodification is the relational core: the accumulated knowledge of a particular person, availability at the moment of unscheduled need, and the attentiveness that no contract can specify in advance. A commodity must be definable independently of who supplies it, and care defined that way loses the attribute purchasers value most. The practical result is a permanent fringe of household labour, goodwill, and improvisation around every marketised system, on which the system silently depends while describing itself as complete.

Q: What happens when care is marketised?

The cost structure does not change, so the pressures relocate. A provider that cannot raise productivity in the attended hour must find its margin in the wage bill, the staffing ratio, the skill mix, or the fee, and competitive pressure pushes toward the first three because they are the least visible to a purchaser who cannot judge quality before, or sometimes even after, buying. Continuity of carer thins, visits shorten to timed tasks, and ownership can financialise, extracting value from the property rather than the provision. None of this requires bad faith. It is the predictable expression, inside a market form, of a product whose quality cannot be inspected and whose core input cannot be cheapened.

Q: How does austerity shift care back to households?

Through the funded hours. When public services are cut or relet at lower rates, the visits shorten, the eligibility threshold rises, and the day centre closes, but the needs those services met are set by bodies and do not shrink in step. The unmet residue is absorbed by the nearest available unpaid labour, usually a woman in the same family, whose day lengthens to contain it. The saving recorded in the budget is therefore partly a transfer rather than a reduction: costs reappear as forgone earnings, damaged health, and later emergencies in other budgets. Feminist research on adjustment programmes named the household the shock absorber of fiscal policy, and the mechanism described here is that phrase at the level of a single family’s week.

Q: Who does paid care work?

Predominantly women, and within that majority, disproportionately women whose bargaining position is weakened by something beyond the job itself: migrants whose status may depend on an employer, workers recruited through agencies onto fragmented hours, mothers whose availability is rationed by their own unpaid obligations. The workforce is internally graded, from registered professionals with protected titles down to informal helpers invisible to every statistic, and the grades are connected, since the cheapest tier’s existence disciplines the wages of all the others. Men are a minority throughout, concentrated toward the technical and managerial roles, a distribution that tells its own story about how the sector values the work against the worker.

Q: What is the care drain?

The phrase adapts the older idea of a brain drain to the specific traffic in caring labour. Wealthy economies recruit nurses and carers from poorer ones, gaining trained staff whose upbringing and education were paid for elsewhere; the sending economies lose experienced hands from health and household systems that were already thin. The drain is measured not only in professionals. Every migrant carer’s departure sets off a substitution at home, a relative or a poorer paid helper taking up her caring, so the cost travels down a chain and settles where refusal is least possible. The term’s purpose is accounting: to make a transfer appear on the page that the receiving economy’s books record as pure gain.

Q: What policies are proposed for care?

The proposals sort into families, each answering a different diagnosis. One family would wage or credit the household carer, making the work visible through payment and pension entitlement. A second would expand public provision toward a universal service, accepting the rising budget share as the honest price of the standard. A third keeps the market and fences it, with wage floors, mandated ratios, and limits on financial extraction. A fourth redistributes time, through shorter working weeks and leave designed to be taken by fathers as well as mothers. A fifth, from outside the critical traditions entirely, trusts demand and technology to raise the sector’s wages unaided. The article maps the positions without adopting one; what the analysis insists on is that each family is a decision about where the cost of care is to land.

Q: Has rising female employment changed the analysis?

Yes, and the change is registered in the article rather than waved away. The household the classic texts assumed, a full-time male earner and a full-time unwaged carer, is now a minority form in many economies, participation has risen across the long run as the research of economic historians documents, and household technology has abolished specific drudgeries. What rising employment changes is the location of the caring, not its necessity: the work moves between households, markets, states, and migrant labour, which is precisely the movement the framework exists to track. The concession the objection earns is real but specific: descriptions of the domestic division of labour written for the old household form need rewriting for the new one, and the article rewrites rather than repeats them.

Q: How does the productivity trap explain a pay pattern across unlike care jobs?

Because the trap is a property of the activity, not of any one job title. The nursery worker, the dementia carer, the district nurse, and the live-in helper share one feature beneath their differences of setting and status: the value of their hour lies in attended presence, and compressing the presence cheapens the product. Occupations built on that feature inherit the same ceiling on productivity growth and therefore the same pressure on wages, whatever their uniforms or funding streams. Jobs at the sector’s edge, laundry, catering, transport, escape the trap insofar as their tasks genuinely mechanise, and the border between the two kinds of work, visible in any hospital’s pay structure, is the trap’s fingerprint.

Q: What should a student separate between care’s low pay and care’s low status?

Separate the price from the regard. Low status is the work’s reputation, the assumption that it is simple, natural, and suitable for those fitted to nothing else. Low pay is its price in the labour market, set by the mechanisms this article details. The two reinforce each other, since badly paid work is presumed worthless and work presumed worthless is paid badly, but they are not the same injury and they do not yield to the same remedy. Honours and rhetoric treat the first and cost nothing. Only funding treats the second. An answer that keeps the distinction visible can explain why a society may sincerely celebrate its carers, in speeches and ceremonies, while the wage packet does not move.

Q: Why can care not be speeded up the way manufacture can?

Because in manufacture the worker’s time and the product are separable: the same shoe, made faster, is the same shoe, and the saving is real. In care the time is a constituent of the product. A consultation, a bath, a settling to sleep, done in half the time, is not the same service delivered efficiently; it is a different and usually worse service, with the difference falling on a person who may be unable to report it. Some surrounding tasks, records, laundry, routing, can be speeded without loss, and should be. The core cannot, and the attempt to treat the core as though it were the periphery is the sector’s characteristic management error, producing measured efficiency and unmeasured harm in fixed proportions.

Q: How does the care pressure table order the sector’s employers’ choices?

The table arranges the four sites of care, household, public services, market provision, and migrant chains, and asks of each who works, who pays, and what gives under pressure. For an employer it functions as a decision order: revenue that cannot grow with productivity leaves four variables to absorb the strain, the fee, the wage, the staffing ratio, or the subsidy demanded from the state, and the table shows where each absorption lands. Its use is diagnostic rather than decorative. Presented with any reform, the reader locates the row it relieves and the row it burdens, and the claim that the pressure has been abolished can be checked against the rows the announcement does not mention.

Q: What separates a structural explanation of low pay from a moral one?

A moral explanation stops at the attitude: the work pays badly because it is despised, full stop. A structural explanation specifies a mechanism that would produce the pattern even among employers of goodwill, and it exposes itself to evidence in ways the moral account does not. The cost structure described in this article predicts where pay pressure will appear, which ownership forms will dilute staffing, what fiscal consolidation does to unpaid hours, and what would refute the whole account, namely a care service that durably raised output per worker at constant quality and shared the gain in wages. The moral account is not false; attitudes are among the mechanisms, through the pricing of skill as nature. It is incomplete, and its incompleteness shows in its prescriptions, which reach for esteem when the binding constraint is arithmetic.