Compute the Arkansas CDCC from IRC § 21 as in effect January 2, 2013 (Form AR2441)

[hua7450]

Summary

Implements Arkansas's static conformity to the 2013 IRC for its Child and Dependent Care Credit, fixing the TY2021 ARPA leak found in the #8826 verification — plus a base correction proven by the state's own form during implementation.

Closes #8880

Regulatory basis (all primary sources read directly)

Ark. Code Ann. § 26-51-502:

• (b)(1): "Title 26 U.S.C. § 21, as in effect on January 2, 2013, is adopted for purposes of determining the allowable credit…"
• (b)(2): "The amount of credit shall be twenty percent (20%) of the federal credit allowable."

2021 Form AR2441 (read from the rendered form): line 3 caps qualified expenses at "$3,000 for one qualifying person or $6,000 for two or more" (not ARPA's $8,000/$16,000); the line 8 rate table runs $0–15,000: .35 down to "43,000 – No limit: .20" (not ARPA's 50%); line 10: "Multiply line 9 by .20"; the header applies the § 21(e)(2)/(4) married-filing-separately rules. Arkansas computed its own pre-ARPA credit in 2021 — the 2021 AR1000F instructions (Line 35) direct filers to AR2441, not the federal form.

2022 Form AR2441: structurally identical — and, like the 2021 form, contains no federal tax liability limitation line. The state base is the § 21 credit allowable (pre-cap), matching (b)(2), not the liability-capped federal credit.

Changes

1. New ar_federal_cdcc — the pre-ARPA § 21 replica for 2021, computed from federal CDCC parameters pinned at the 2020 instant (the same pinning pattern as va_child_dependent_care_deduction_cdcc_limit, id_cdcc_limit, and the CA 2021 replica): expenses including disabled-adult care_expenses, per-qualifying-person caps, the head/spouse earnings cap (§ 21(d)(2) deeming included — the $250/$500 monthly amounts are unchanged since 2003), the 35→20% AGI-phased rate, and the cdcc_filing_status_eligible gate. second_start is .inf at the 2020 instant, so no ARPA second phase-out leaks in.
2. ar_cdcc_potential — 2021 uses the replica; all other years use cdcc_potential (pre-liability-cap) instead of the liability-capped cdcc, per the form and statute. In non-2021 years the 2013 and current § 21 produce identical values, so the pin itself only binds in 2021. Also converts the deprecated documentation field to reference and documents the unmodeled § 26-51-502(c) refundable credit (care at a Department of Education-approved early-childhood facility, Form AR1000EC; mutually exclusive with this credit — we don't track facility approval at the moment).

Reproductions (hand-computed on Form AR2441)

Scenario	Before	After
TY2021, AGI $50,000, one child, $4,000 expenses	$400 (20% × ARPA $2,000)	$120 (AR2441: $3,000 × .20 × .20)
TY2021, AGI $14,000, $3,000 expenses	inflated	$210 ($3,000 × .35 × .20)
TY2021, MFS not separated	positive	$0 (AR2441 header rule)
TY2022+, federal credit zeroed by federal liability, $1,000 potential	$0	$200 (no liability line on AR2441)

Tests

New ar_cdcc_potential.yaml with the four cases above (line-by-line AR2441 comments); existing ar_cdcc.yaml inputs updated to the corrected base. Full AR income tax tree passes locally: 198 tests.

Independent of #8855/#8878/#8884 — no shared files.

🤖 Generated with Claude Code

[codecov]

Codecov Report

❌ Patch coverage is 93.75000% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 95.23%. Comparing base (f9e58e7) to head (7f13085).
⚠️ Report is 46 commits behind head on main.

Files with missing lines	Patch %	Lines
...ov/states/ar/tax/income/credits/ar_federal_cdcc.py	92.59%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##              main    #8885      +/-   ##
===========================================
- Coverage   100.00%   95.23%   -4.77%     
===========================================
  Files            3        2       -1     
  Lines           55       42      -13     
  Branches         0        2       +2     
===========================================
- Hits            55       40      -15     
- Misses           0        1       +1     
- Partials         0        1       +1     

Flag	Coverage Δ
unittests	95.23% <93.75%> (-4.77%)	⬇️

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.