Care Does Not Clock Out

Paid care has a market price. Unpaid care has hours, schedule control, interruption, and responsibility that ordinary receipts leave elsewhere.

Care enters public statistics through more than one door. Childcare market surveys record prices charged by providers. Time-use surveys record what people report doing across a day. Eldercare estimates identify providers and, for those giving care on a sampled day, the hours involved. Each record is useful. None contains the whole shift.

This matters because paid and unpaid care can substitute for, overlap with, or enable paid work without becoming the same kind of burden. A provider price is money. Three hours and fifty-four minutes of reported eldercare is time. Being responsible for a child while cooking or working may appear as secondary care. Availability, interruption, worry, and schedule control can remain partly outside all three.

The answer is not to invent a heroic total that prices every family relationship. It is to stop treating the market receipt as proof that the rest of the care economy clocked out.

Childcare prices are available at the county level

The National Database of Childcare Prices organizes state market-rate survey results into county-level estimates.

County detail matters because the practical market for daily care is local, not national.

State surveys differ in methods and years; a county estimate is not a quote from every provider or a measure of open slots.

Childcare price comparisons must preserve provider type

The federal database distinguishes center-based care from family childcare homes.

Combining provider types can hide meaningful differences in schedule, setting, regulation, and price.

Provider type does not by itself establish quality, convenience, or suitability for a particular child.

The age of the child changes the childcare price record

The database separates prices for infants, toddlers, preschool-age children, and school-age children where source data permit.

Staffing ratios and care arrangements change with age, so one generic childcare price is a weak household estimate.

Age categories and availability vary across state survey instruments and should be checked before comparison.

The childcare database spans multiple market-rate survey years

The Department of Labor’s harmonized files cover childcare price observations from 2008 through 2022.

A longer record supports inflation and local-change questions that a single-year ranking cannot answer.

The series is assembled from state surveys rather than one identical annual national instrument.

The federal childcare-price compilation is broad but not perfectly complete

The Women’s Bureau reports usable childcare price information for 48 states, the District of Columbia, and Puerto Rico in the compiled database.

Naming the missing coverage prevents a broad national resource from being mistaken for a complete census.

Coverage and field completeness differ across years and jurisdictions.

About 38.2 million people provided unpaid eldercare in 2023–2024

BLS estimated that 14 percent of the U.S. civilian noninstitutional population age 15 and older provided unpaid eldercare over a three- to four-month reference period.

Caregiving is large enough to belong in work, time, health, and household-margin analysis.

The measure covers unpaid care for someone age 65 or older because of an age-related condition; it is not all disability or childcare care.

The handoff from price to time

The first five records describe a childcare-price database built from state market-rate surveys. County detail, provider type, child age, survey year, and coverage are not clerical decorations. They determine whether two numbers belong in the same comparison. A price for infant care in one provider setting and year cannot be quietly promoted to the price of “childcare” everywhere.

Even a perfectly labeled price remains an offer rather than a completed arrangement. The database does not say whether a provider has an open slot, covers the needed hours, can serve the child’s needs, remains reachable from home or work, or accepts the payment arrangement available to the family. Those conditions can matter as much as the posted amount, but the cited price record does not measure them.

The eldercare records enter through time rather than market price. Their population includes unpaid providers who gave qualifying care during a three- to four-month reference period. A smaller share did that work on the sampled day, and the reported hours belong to those participating that day. The method can describe scale and daily participation without claiming that care follows a predictable shift.

This is why care resists one total. A paid slot can replace some hours and leave coordination, transport, nighttime responsibility, emergencies, and other work untouched. Unpaid care can be freely chosen, deeply valued, exhausting, or all three before lunch. Measurement should not settle that meaning. It should make the time and conditions visible enough that public policy cannot treat them as air.

Only part of the eldercare-provider population gives care on a given day

BLS found that 28 percent of eldercare providers engaged in eldercare on an average day in 2023–2024.

The daily denominator prevents the provider population from being treated as if everyone performs care every day.

Average-day participation does not show peak episodes, overnight responsibility, or being on call without a recorded activity.

Eldercare providers who gave care on a day averaged 3.9 hours

Among providers who performed eldercare on a given day, BLS estimated an average of 3.9 hours of care activities.

The time is substantial, but it belongs to the participating-day group rather than every provider.

The average can hide a wide distribution and follows ATUS activity-classification rules.

Time-use data preserve some supervision that occurs alongside another activity

ATUS records secondary childcare when a respondent has a child in their care while reporting a different primary activity.

This protects some simultaneous supervision from disappearing behind cooking, work, travel, or another main activity.

Secondary childcare is not a complete measure of attention, responsibility, mental load, or all care for adults.

Care needs a receipt with both money and time

Federal price and time-use systems separately document market childcare prices and unpaid eldercare activity.

A household can face a high cash price, a high unpaid-time burden, or both; collapsing them loses who supplied the care.

No federal source combines these measures into one universal care-burden score, and bent.WTF does not invent one.

What the ten records show together

The childcare database gets closer to real purchasing conditions by preserving county, provider type, child age, survey year, and coverage. Those fields prevent a national average from impersonating a local quote. They still do not reveal whether a slot exists, matches the needed schedule, accepts the child, or can be reached reliably.

The eldercare records answer a different question. About 38.2 million people were identified as unpaid eldercare providers in 2023–2024, while 28 percent provided care on an average day. Among that participating-day group, average care time was 3.9 hours. Mixing those denominators would make a large population look like a daily headcount or make a participating-day average describe everyone. Neither move survives contact with the footnote.

A care receipt needs separate rows for provider price, fees, travel, paid hours, unpaid hours, secondary responsibility, missed work, schedule changes, and whether reliable care was actually available. Some rows will be blank because the present sources do not measure them. The blank is more honest than assigning all care one hourly wage and calling the matter settled.

Publish the schedule, not only the price

Better public measurement would keep local prices comparable across provider type, age, and year while adding the questions families actually face: hours offered, open slots, wait lists, closures, travel, and schedule fit. Time-use work should continue to preserve secondary care and make its activity rules plain. Caregiver analysis should distinguish the full provider population from those providing care on the diary day.

No one dataset needs to observe every emotional or relational dimension of care. It does need to stop at its boundary without suggesting that what lies beyond is negligible. Paid care appears as a transaction. Unpaid care appears as someone else’s Tuesday.

Care does not clock out when the survey diary closes. The public account should at least admit who is still on call.

Evidence boundary

This report uses ten source-reviewed federal records on county-level childcare prices and American Time Use Survey eldercare and secondary-childcare measures. It preserves provider types, child-age categories, survey years, coverage, average-day denominators, and activity rules. It does not estimate open childcare slots, value unpaid care in dollars, or measure every on-call, emotional, or scheduling burden. Human read-aloud, accessibility, observed-reader, and publisher review remain separate release gates.