Terminate clean-energy credits under Pub. L. 119-21

[MaxGhenis]

Summary

Applies the One Big Beautiful Bill Act (Pub. L. 119-21, the 2025 reconciliation law) terminations of four clean-energy tax credits.

• Residential clean energy credit (IRC 25D, § 70506) — Sets the applicable percentage to 0 for expenditures made after 2025-12-31 and removes the now-dead IRA phasedown entries (2033: 0.26, 2034: 0.22, 2035: 0). The credit's formula already reads applicable_percentage, so no code change was needed.
• Energy efficient home improvement credit (IRC 25C, § 70505) — Turns off in_effect at 2026-01-01 (property placed in service after 2025-12-31). The formula already gates on in_effect, so no code change was needed. The 25C(h) product-identification-number (PIN) requirement remains unmodeled — see Update section 25C energy efficient home improvement credit for current law #8703.
• New and used clean vehicle credits (IRC 30D/25E, §§ 70502/70501) — Adds eligibility/in_effect parameters under clean_vehicle/new/ and clean_vehicle/used/ and gates the two eligibility variables (new_clean_vehicle_credit_eligible, used_clean_vehicle_credit_eligible) on them. Because the downstream credit, credit-limit, and potential variables are all defined_for the eligibility variables, gating eligibility to False zeroes the entire credit chain — no hardcoded values in formulas.

Clean vehicle acquisition-date caveat

The statute terminates the vehicle credits for vehicles acquired after September 30, 2025 (an intra-year date, confirmed against 26 U.S.C. §§ 25E(g), 30D(h)). PolicyEngine models annual periods, so this cutoff is approximated as the credit ending for tax years after 2025 (2026-01-01), rather than mid-2025. This is documented in both in_effect.yaml files and the two formula comments. It is not exact: vehicles acquired in Q4 2025 are treated as eligible here even though they are not under current law. Representing the intra-2025 cutoff exactly would require an acquisition-date input at sub-annual resolution (the adapter/design ask in #8760).

Tests

Added YAML tests asserting each credit is nonzero in 2025 and zero in 2026 (residential clean energy potential, EEHIC potential, and new/used clean vehicle eligibility), and updated the stale residential-clean-energy 2033 phasedown test (previously asserted 26%, now 0). Added changelog.d/clean-energy-terminations.fixed.md.

Ran locally against Python 3.14 (policyengine-core test ... -c policyengine_us):

• residential_clean_energy: 5 passed
• energy_efficient_home_improvement: 44 passed
• clean_vehicle: 30 passed

Parameter tree loads cleanly; boundary values verified: 25D applicable percentage 0.3 (2025) → 0 (2026); 25C in_effect True (2025) → False (2026); new/used vehicle eligibility.in_effect True (2025) → False (2026).

Fixes #8694
Addresses #8703 (PIN requirement unmodeled)
Addresses #8760 (acquisition-date sunset approximated at annual resolution; adapter/design ask open)

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.00%. Comparing base (f9e58e7) to head (3d9c11d).
⚠️ Report is 69 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff            @@
##              main     #8901   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files            3         2    -1     
  Lines           55        44   -11     
  Branches         0         2    +2     
=========================================
- Hits            55        44   -11     

Flag	Coverage Δ
unittests	100.00% <100.00%> (ø)

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

☔ View full report in Codecov by Harness.
📢 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.