Date: 2026-08-03. Draft recommendation from the session's evidence; milestones and leading indicators to be formalized.
Pass-2 amendment (red-team attack 4): the sequence below is restated as "visible-first where capacity exists, supply-first where it doesn't." GA/OK prove demand-first works at the 4-year-old tier where school capacity existed; the supply-first logic governs 0–2 and compensation, where capacity does not. The pure supply-first architecture's political fragility (durability score 2) is real and must be mitigated by pairing it with fast visible wins (school-age seats, retention bonuses landing in educators' paychecks in year one).
The constraint order the evidence supports
- Workforce binds first and everywhere. Five independent designs (Canada, Quebec, Germany, DC, the US market itself) show money outrunning educators (ws04, ws15). Within workforce, the evidence splits the problem: retention responds fast to wage funding (DC: measurable turnover reduction); recruitment at scale is pipeline-bound and slow (Canada: years behind with money in hand).
- The instrument matters as much as the amount. Supply-side operating grants suppressed prices at $24B scale; demand-side subsidies inflate fees (Australia) (ws09).
- The cliff is measured, not theoretical. Access deteriorated within two quarters of the stabilization expiry (ws09). Reconciliation's sunset-forcing (ws11) would rebuild this cliff at larger scale — temporary-funding ramps are a design defect, not a compromise.
Draft sequence (v0)
Year 0–1 (no new legislation required): consume the remaining administrative headroom (post-2024-rule CCDF items, §45F monitoring under its new parameters); state-level: zoning preemption, school co-location authority. Cheap, fast, small.
First bill — supply and compensation, not eligibility. A stabilization-style operating-grant program with compensation funding (DC-PEF-style, financed not mandated — the Byrd-survivable form, ws11), rural-weighted capital (deserts >70% rural and worsening, ws05), and pipeline funding (credentialing + background-check throughput). Reconciliation-compatible because it is money. Dedicated revenue from the start — the DC lesson (ws04): proven programs without dedicated revenue fight annual survival battles.
Second bill(s) — demand-side expansion, after capacity exists. Sliding-scale (no cliffs — the near-poor burden peak, ws08), phased age-band eligibility. Demand money arriving after supply money inverts the Australia/Quebec failure order.
Segment sequencing (per ws14): school-age first is newly credible — 64% of children, existing buildings, the OK funding-formula precedent, red-state-durable — and it delivers the largest fast labor-supply effect. Then 3–4 consolidation. 0–2 last and differently — supply-first plus caregiver-choice, because its market is home-based and its workforce is scarcest.
The employer-channel fork (flagged, unresolved): §45F just expanded — H12.3 (crowd-out) is now running as a live experiment. Track claim volume; if uptake surges among large employers, the constituency-erosion scenario gains evidence and the sequencing case against further employer-channel expansion strengthens.
Leading indicators to formalize
Educator headcount growth vs. slot growth (the Canada tell); fee inflation vs. subsidy growth (the Australia tell); waitlist emergence in price-capped segments (the Quebec tell); chain participation rates at public rates (the rate-participation tell); FCC licensure trend reversal (the 0–2 supply tell).