GUBMENTPlain talk · policy frontier
Filings / Childcare / Sources / §14 Scorecard — Anchored Scales (autho
GBMT-1 · Research record · No. 1

§14 Scorecard — Anchored Scales (authored before scoring)

childcare/research/scorecard/scales.md
This is a working research document from the childcare filing, published as written — including the parts later corrected. It is the underlying record for Whitepaper No. 1, not a summary of it.

Method note. The protocol's 23 dimensions are consolidated to 14 for this pass (deviation #5): the omitted dimensions (time-to-first-child, ramp rate, admin burden, build-vs-buy, employer-constituency effect, fault-line trigger, recession resilience, evidence base) fold into the retained ones or lack pass-level evidence; they are scored narratively in the rationale file where evidence exists. Scores 1–5.

CORRECTED 2026-08-04 (Verification Protocol, Phase 1). This note previously read: "Every cell in the CSV carries an evidence tag resolvable in the rationale file." That was false. scores.csv contains 168 cells (12 architectures × 14 dimensions). The rationale file's "Key cell citations" section tags 9 of them. The remainder are uncited analyst judgement against these anchored scales.

This matters more than a normal correction, because the series method (M6) requires that "every cell cites a workstream finding," and this filing's scorecard is what produces its central architecture recommendation. The honest description is: anchored scales authored as stated, applied by a single analyst, with ~5% of cells carrying an explicit evidence tag. The scales are real and the scoring may well be defensible — but it is not evidenced at the cell level, and the record claimed it was. See verification-log.md.

Dimension 1 means 3 means 5 means
relief_workforce No compensation/pipeline funding at all Partial or indirect wage support Funds compensation at parity + pipeline, DC-PEF-style
relief_supply Builds no capacity; pure transfer Holds existing capacity (stabilization-style) Builds new capacity incl. rural/nontraditional
coverage_02 Reaches almost no 0–2s Reaches subsidized/poor 0–2s only Universal 0–2 offer incl. home-based settings
coverage_34 No 3–4 provision Partial (income- or state-limited) Universal 3–4
coverage_512 Ignores school-age Before/after in some districts Universal wraparound + summer
procedural Needs 60 votes AND fails Byrd if reconciled Reconcilable with major strips Clean reconciliation fit or no legislation needed
federalism Coverage collapses in holdout states Strong incentives, no fallback Holdout-proof (direct-federal or fallback)
durability Annual appropriations, no constituency Broad constituency OR dedicated revenue Both + institutional entanglement
passthrough_risk (5=low) Pure demand-side into inelastic supply Mixed instruments Supply-side operating grants / direct provision
participation_risk (5=low) Depends on incumbents accepting below-cost rates Cost-based rates negotiated Doesn't depend on incumbent participation
preference_fit Center-only for all ages Mixed delivery, weak home-based support Funds the arrangement families actually choose per band
integrity_risk (5=low) Netherlands-shaped: mass demand-side clawback exposure Standard grant oversight Direct provision / small-N grantee audit surface
distributional Benefits concentrate on high-income/served markets Proportional Concentrates on near-poor burden peak, deserts, nontraditional hours
evaluability No evaluable variation Observational only Staged/lottery rollout built in
← All Childcare research documents Sources digest Read the whitepaper