Encode historical MO and NY CHIP premium schedules for 2020-2026

[MaxGhenis]

Follow-up to #8086.

Problem

The CHIP premium parameters for Missouri and New York were dated 2021-01-01 with the most-recent (2025–2026) dollar amounts, so chip_premium at period=2024 returned modern values rather than the historically accurate 2024 amounts. This matters for the us-data MOOP imputation subtraction described in #8089: chip_premium evaluated at the CPS reference year (2024) needs to return what CPS respondents actually paid in 2024.

Changes

Missouri (MO HealthNet for Kids, Appendix E / IM-4(PRM))

Three effective-dated schedules encoded in premium/tier_1.yaml, tier_2.yaml, tier_3.yaml:

Period	Family-size-1 premium (tier 1 / 2 / 3)	Source
2020-01-01 – 2024-06-30	$15 / $48 / $117	Appendix E (04-20)
2024-07-01 – 2025-06-30	$18 / $58 / $141	IM-4(PRM) (07-24)
2025-07-01 onward	$18 / $60 / $147	IM-4(PRM) (07-25)

The 2020 schedule was stable through June 2024 (confirmed by archived 2020 and 2022 charts showing identical dollars, and a Missouri Budget Project November 2023 brief citing the same $82 value for a family of 3 at 201% FPL).

New York (Child Health Plus)

Restructured premium/per_child.yaml and premium/family_cap.yaml:

Period	Structure	Per-child rates	Source
pre 2023-02-01	6 brackets, free below 160% FPL	$0/$9/$15/$30/$45/$60	2020 snapshot
2023-02-01 onward	5 brackets, free below 222% FPL ($9 tier eliminated)	$0/$15/$30/$45/$60	Feb 2023 snapshot
2026-02-17 onward	Same rates, thresholds micro-shifted (2.22→2.23 etc.) reflecting FPL rounding	$0/$15/$30/$45/$60	current page

Wisconsin

No change. The per-child table has been stable since Release 19-01 (April 2019); only the 5%-of-income family cap amounts have moved release-to-release, and that cap isn't implemented in the model yet. Existing 2021-01-01 entries are historically accurate for 2024.

Tests

Existing MO and NY tests at period=2024 updated to reflect actual 2024 values (e.g., MO family of 3 tier 1 = $25/mo = $300/yr pre-July 2024, not the $31/mo = $372/yr July-2025 rate). Added period=2026 MO test and period=2022 NY test to verify transitions.

70/70 CHIP tests pass locally. make format / ruff check clean.

Not included

• Missouri April 2021 and April 2023 charts — not archived on archive.org or mo.gov; evidence (identical dollar amounts in the 2020 and 2022 captures; Missouri Budget Project Nov 2023 brief citing matching numbers) indicates the premium dollar amounts were stable. Only the annual FPL refresh changed. Flagged as inferred in the parameter comments.
• Wisconsin release history — confirmed unchanged but not encoded as separate date entries since the per-child table didn't change.

Test plan

• 70 CHIP tests pass locally
• Format / lint clean
• CI passes

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 70.65%. Comparing base (05d5d97) to head (8eb635e).
⚠️ Report is 26 commits behind head on main.

Additional details and impacted files

@@             Coverage Diff              @@
##              main    #8091       +/-   ##
============================================
- Coverage   100.00%   70.65%   -29.35%     
============================================
  Files           18     4519     +4501     
  Lines          353    65640    +65287     
  Branches         4      331      +327     
============================================
+ Hits           353    46380    +46027     
- Misses           0    19260    +19260     

Flag	Coverage Δ
unittests	70.65% <ø> (-29.35%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.