The binding constraint on universal childcare in the United States is not appropriations but labor: full coverage for ages 0–12 requires roughly 2.8 million educator positions against a 2020 pre-pandemic baseline of ~1.05 million (current employment is closer to 1.10 million), in an occupation among the lowest-paid BLS tracks. Evidence from five national programs shows funding that outruns workforce produces queues, litigation, or price inflation rather than care. Instrument choice determines pass-through; Senate procedure permits financing and forbids mandates; and the durable American precedents are state-level and bipartisan. We conclude feasibility is a design and sequencing problem — pay the workforce first, build supply second, extend benefits third — and stamp our uncertainties in the text.
The headcount: two-thirds of the kids are missing from the debate
School-age children are 64% of the population in scope, and they are absent from essentially every cost estimate we collected, which stop at age five. Their problem is after-school hours and roughly 75 school-free weekdays a year; unmet afterschool demand is at an all-time high, with 4 in 5 middle schoolers who want a program unable to get one.
Census PEP Vintage 2024 (committed pipeline) · America After 3PM 2025The worker problem: triple the workforce, or halve the quitting
Full coverage requires ~2.8 million educator FTEs (range 1.9–3.9M) against a 2020 pre-pandemic baseline of ~1.05 million (current employment is closer to 1.10 million). Ramp-rate arithmetic: ~800,000 hires per year, of which 579,000 merely replace departures — cutting turnover in half saves more hiring than the entire expansion requires. The credential pipeline issues over 50,000 per year, renewals included — still well short of the ~798k/yr ramp. Childcare's median wage is $15.41/hour, among the lowest-paid occupations BLS tracks.
FTE model (committed script; Census + BLS inputs) · CDA Council · Cleveland Fed worker-flowsEvery failed precedent also underpaid. No country has attempted the buildout at genuinely competitive wages; at parity, the constraint is untested rather than proven fatal.
A Phase 2 steelman found the word doesn't clear the filing's own test — a constraint is binding only if relaxing it alone helps and relaxing the others wouldn't. Loosening ratios to what Florida's own statute already permits, with no pay rise at all, cuts required hiring to ~501k/year — matching what halving turnover saves. No scorecard dimension prices regulatory burden and no scored architecture touches ratios, so the alternative was never tested, not ruled out.
Instrument choice: vouchers inflate, supply grants suppress
Australia's demand-side subsidy watched fees outrun it (+4.4% year-over-year in one quarter). The United States accidentally ran the opposite experiment: $24B in provider-side stabilization grants under which costs fell — and whose expiry (September 2023) was followed within two quarters by rising prices and a jump in parents unable to find care, 24% to 31%.
CEA analysis · Australian Dept. of EducationNo peer-reviewed evaluation of the $24B experiment exists yet — four papers mention the program; none evaluates it. The finding stands at single-source-family confidence.
Senate procedure: finance survives, mandates die
Reconciliation's Byrd rule declares six categories of provision extraneous; the one that governs here strips provisions whose budgetary effects are "merely incidental" to their non-budgetary components. The closest on-point determination is the $15 minimum wage: the Parliamentarian advised in February 2021 that it did not qualify, it was dropped from the American Rescue Plan before floor consideration, and when it was offered again as an amendment that March a point of order was sustained against it on merely-incidental grounds, the waiver failing 42-58. A childcare wage mandate faces the same objection; wage funding is budgetary and survives that test — though CRS cautions every such determination is case-specific, and money-moving provisions can still fall to the rule's outyear-deficit test, which is why reconciliation-built programs get written with sunsets. Separately, CBO's scoring of the House-passed BBB title ($381.5B, 2022–31) embedded substantial state non-participation — the holdout problem is the official base case, and any national design requires a federal backstop.
Senate parliamentarian precedent (ARP 2021) · CBO 57630 · KFF expansion trackerThe comparative record
| Case | Design | Outcome |
|---|---|---|
| QUEBEC '97 | Flat-fee supply-side | 30,688 waitlisted after 25 years; staffing and funding both named as causes |
| GERMANY '13 | Legal entitlement | Courts award money, not slots |
| AUSTRALIA | Demand subsidy | Fee inflation; the later rate cap had "only limited effectiveness" (ACCC) |
| CANADA '21 | $10/day federal-provincial | 194K of 284K spaces; ~$19/day; educator shortage |
| GA/OK/FL | Universal pre-K | Decades durable in red states |
| NYC '14 | Municipal sprint | 60,000+ seats in two years |
| US DOD | Competitive-rate pay, no parity mandate | 34–50% turnover (FY2022) and staffing-driven waitlists despite market-rate pay |
Architecture: three designs survive every ranking
Eleven candidate architectures plus a cash comparator were scored on fourteen anchored dimensions under four objective weightings (labor-supply-first, child-development-first, equity-first, equal), then re-scored blind. Three designs occupy the top ranks under all weightings: a public option for markets private operators decline, a scaled Head-Start-style delivery model, and K–12 extension for ages 3+ and school-age wraparound. The fourth seat flips — supply-first under three weightings, caregiver-choice under equity. A federal fallback remains the holdout-proofing feature, not a top-four architecture on its own. The two most fashionable designs — a pure demand-side allowance and federalized employer cost-splitting — rank last; unconditioned cash is mid-table. Ages differ: schools for the after-3pm-and-summer problem; funded mixed delivery for preschool; home-based options, including paid kin, for infants.
Scorecard: 12 × 14 × 4, anchored scales, independent re-score (47 cells moved) — committedA Phase 2 steelman found three of the fourteen scoring dimensions anchor their scale on the instrument category itself — "pure transfer" scores 1, "direct provision" scores 5 — so a cash or voucher design scores near the floor by definition, before any evidence. Drop those three dimensions and the top four changes under all four weightings: supply-first (a6) exits and caregiver-choice (a11) enters under three, while under equity-first a11 leads outright and the cash comparator — the design the anchor text cannot score above the floor — takes the fourth seat. What survives is narrower than the headline: the public option (a8) and scaled Head Start (a3) hold top-four places under every weighting in both specifications. "Three designs survive every ranking" does not; one of the three is an artifact of how the instrument was scored.
Sequencing: pay, build, then promise
Visible-first where capacity exists (school-age seats, pre-K), supply-first where it does not (ages 0–2, compensation). Fund retention immediately — it is the arithmetically dominant lever — with dedicated revenue rather than annual appropriations; extend demand-side eligibility only as staffed capacity arrives, so money buys care instead of price increases. Reconciliation-built sunsets recreate the 2023 funding cliff at scale and are a design defect, not a compromise.
The honesty box
Nobody has tried fair wages at scale. "Workforce binds" is proven only for underpaying programs. The price evidence is one source deep. The $24B experiment awaits independent evaluation. Big chains aren't full. Metro markets show price-rationed slack at ~71% occupancy; the physical shortage is rural and infant-age. Twice checked, not final. An independent structurally-blinded re-score moved 47 cells and broke the old flip-free top-four; three designs are stable, the fourth seat flips by objective, and a single re-score is weaker evidence than this project once treated it as. The child-development weighting doesn't measure child development. No scorecard dimension scores an effect on children; child_dev_first is implemented as coverage counts, which assumes the disputed proposition rather than testing it — though setting that term to zero, rather than inverting it, leaves the top four unchanged. The Head Start Impact Study is absent from the record. Our own protocol commissioned it by name; scaled Head Start ranks in the top four under every weighting without it ever being engaged. Quebec's child-outcome literature is cited only for its waitlist. The canonical Baker–Gruber–Milligan findings turned up on our first search pass and were logged as a citation count, never their result — while newer work (Montpetit et al. 2026) finds the behavioral harms don't translate into lower attainment or earnings.