How Child Care Demand calculates a market

Resident demand starts by counting children aged 0 to 4 who live inside the catchment. We then use the newest Australian approved-care rate as a starting point. The report shows each number and keeps service-location enrolments separate, so the same child is not used to make both estimated need and observed use look bigger.

Parent employment comes from ABS Census 2021 G47A and G48A family tables at SA2 usual-residence geography. The model apportions employed-parent and total-parent counts by catchment overlap, then divides the totals. Well-covered samples use the local observed rate. Smaller samples are stabilised with an explicitly named state rate, and sparse catchments use a labelled state fallback. The report shows the sample, coverage, fallback and confidence. It does not claim SA1 precision for this SA2 source. We show parent employment because work can affect a family's need for care. It does not change the estimate yet because current evidence has not proved how much it should change it.

The Australian Government recorded 753,040 children aged 0 to 4 using CCS-approved care during the March quarter of 2026. ABS estimated that Australia had 1,517,435 children aged 0 to 4 at 30 June 2025. Dividing 753,040 by 1,517,435 gives 49.6%.

Imagine 100 Australian children aged 0 to 4. About 50 used approved care during the quarter. We use that as the starting point because it is current and matches the same ages as the local child count. A child is counted after at least one charged session, whether they attended once or every week.

This 49.6% rate is current for Australia, but it is not a measured rate for this address. Public regional CCS tables show where services are located, not where each child lives. We do not pretend those service locations are a local resident rate. That is why the Australian source is high confidence, while the estimate for one address remains low confidence. The older ABS June 2017 care-use survey is kept only as history and no longer changes the central estimate.

Worker evidence starts with two observed ABS Census measures: people working at workplaces in the catchment and employed people living in the catchment. Their positive difference is shown as a conservative net-inflow lower bound. It is not treated as gross inbound commuting because workers can travel both into and out of the same catchment.

Worker-linked places are estimated only when official origin-to-destination flows identify gross inbound workers, at least 80% of those workers have origin-area parent and paid-care evidence, and an independently validated care-near-work selection rate is loaded with cohort size, RSE, confidence interval and passed state, metro and regional hold-outs. The care-near-work source is the ABS 2017 Childhood Education and Care Survey item LDCQ19, using the distinct-child union of close to a parent's work and on the way to a parent's work. Until every gate passes, worker-linked places are unavailable. Even when validated, they are shown separately from quarterly resident users. The previous Greater Sydney-derived fixed-share result is retained as a labelled comparison, never as national evidence.

Quarterly distinct users are not average daily places, so the report does not add worker-linked places to them.

Every report records the model version, observed and derived values, source vintages, confidence, fallback reason, inclusion decision, and sensitivity range. Missing evidence produces an unavailable or context-only result, not a hidden national fallback.

The origin child-worker factor uses ABS Census 2021 G47A and G48A. It counts each child aged 0-4 once when at least one parent is employed, including employed male- and female-lone-parent families, then divides by employed residents in that worker origin. This avoids counting a child twice when both parents work.

Captured demand uses the latest complete Department of Education Child Care Subsidy quarter loaded into Child Care Demand, apportioned from SA3 areas by their overlap with the report catchment. Each report names the quarter used. It is an observed enrolment signal for children aged 0 to 5 at service locations, separate from the resident estimate for ages 0 to 4.

These groups differ in age and geography; the source does not show how many local residents attend nearby centres each day. We therefore withhold the old capture-rate and open-place calculations, including in saved reports. A daily-place gap needs compatible attendance and occupancy evidence.

The commercial ratio uses approved places from services identified as long day care. Outside school hours care and standalone preschool or kindergarten services are excluded because they do not provide equivalent supply for the same family decision.

Reports also show an inclusive secondary ratio using every nearby service with published approved capacity. This broader figure is context only. It is not the primary long day care feasibility measure. Missing place counts remain unknown rather than being guessed.

Observed occupancy requires occupied place-time divided by available place-time for the same services, age groups and operating period. Licensed capacity is the regulatory ceiling. Staffed capacity is what a service can safely offer with its rooms and workforce. Booked sessions reserve places and may include absences. Attendance records who was present; attended hours measure time present. Charged session hours can be longer.

We have not obtained a commercially reusable, validated local occupancy series. Neither quarterly unique children nor CCS capture, advertised vacancies or children-per-place ratios supply that missing measure. Regional observations cannot establish a centre’s occupancy.

Table 16’s footnote defines spare capacity as 100% minus typical places divided by maximum places offered. We use that percentage definition despite the table’s “Number” heading. The report’s page 6 explains typical places as children attending at one time on a typical day. We publish the source’s average spare capacity, without converting it into an occupancy estimate. Dividing rounded national average places would produce a different statistic.

The survey’s national Centre Based Day Care response was 8,909 of 9,006 services (98.9%). Table 16 has weighted counts of 8,938 specified and 68 unspecified services; these are not raw respondents. It provides no SA2 breakdown or confidence intervals. Small non-zero counts below five are suppressed. Values describe May 2024, even in newly generated reports.

Reports use Mapbox drive-time polygons at 5, 10, and 15 minutes. The selected catchment is saved with the report. Where SA1 population and address evidence are available, population is allocated dasymetrically so uninhabited land does not receive the same weight as residential areas. The model falls back to SA2 overlap where SA1 coverage is incomplete, with a guard against remote-area over-allocation.

The commercial trade area is narrower than the largest reporting ring. It uses the full 5-minute core plus 25% of the incremental 5-to-10-minute band. The same rule is applied to young children and long day care places.

Childcare development applications are included only when they are relevant to future supply in the report catchment. Proposed places are weighted by planning status: pending 0.40, approved 0.75, under construction 0.90, stale 0.20, and needs review 0.25. Evidence confidence, freshness, and likely timing can reduce those weights further.

The report distinguishes current places, published possible places, and weighted likely future places. A blocked or incomplete planning source is shown as a coverage limitation, not interpreted as a zero pipeline.

The standard pipeline view uses a three-year horizon. Where an official state projection for children aged 0 to 4 is loaded, it informs projected resident quarterly care use. Otherwise the report uses the observed child-population trend. Annual child growth is capped between minus 10% and plus 10% to stop sparse or unstable inputs overwhelming the market view. Worker-linked places are not added to this quarterly user projection.

The report names the projection source, source years, annual rate, and applied rate. A forecast is an assumption-led scenario, not a guarantee of enrolments or centre openings.

Local staffing uses the ABS 2021 Census SA2 occupation-by-qualification extract. Usual residence is the primary local availability view, while place-of-work counts describe competition for workers where that evidence is loaded. Estimated educator requirements use a transparent proxy of one educator for every 5.5 long day care places.

National Workforce Census 2024 and Jobs and Skills Australia NERO SA4 figures are shown as regional or state context only. They do not become a site-level staffing score when local qualification evidence is missing.

The school signal counts official primary and combined-school locations inside the saved report catchment and measures the nearest primary-capable school from the report address. Proximity is local family context. It does not prove childcare demand.

Primary enrolment totals use compatible observations from the same year and measure. Combined schools contribute only a published primary subtotal, not their whole-school count. Preschool-inclusive totals and restricted domestic or delivery-mode subsets remain separately labelled in the school evidence. Missing values are not treated as zero.

Trends compare exact three- or five-year periods only where school identity, census, measure and educational scope are comparable. The dates and contributing schools are shown with the result. An unavailable trend does not mean enrolments were unchanged.

ACARA data is not used. Its published terms do not grant the commercial reuse needed for this product. States without a loaded source stay marked unavailable in the report.

Planning coverage is recorded by jurisdiction, source, geography, and date window. A report can therefore distinguish a live match, an audited search with no childcare applications, a blocked source, and an area still awaiting enrichment. See the live coverage ledger for current jurisdiction status.