Repository navigation
Fix the 2020 unemployment compensation exclusion: cutoff test, per-spouse cap, dependents - #10175
Conversation
- Compare taxable_uc_agi, which already leaves out unemployment compensation, with the $150,000 cutoff instead of subtracting the compensation from it a second time. - Exclude up to $10,200 of each person's own compensation, so each spouse on a joint return excludes up to $10,200 (worksheet lines 8 and 9). - Keep a dependent's compensation on the dependent's own return: the tax unit's taxable compensation sums only members who are not dependents, and a dependent's exclusion is tested on a new dependent_taxable_uc_agi. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Hypothesis batches of single, joint, joint-with-dependent and single-parent tax units near the $150,000 cutoff check: per-person bounds, the cutoff on the filers' and each dependent's own return, a differential between taxable_uc_agi and AGI less the taxable compensation, that dependents' compensation never moves the filers' return, and spouse symmetry; and that nothing is excluded in 2021. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The tax unit's taxable unemployment compensation now sums only the head and spouse, in every year, so Indiana's deduction (IC 6-3-2-10) no longer deducts a dependent's compensation that is not in the filers' federal AGI. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…ment_compensation_addback; supply Case 7's taxable allocation as an input Both review rounds noted that the helper adds more than nontaxable compensation for a person recorded as both a dependent and the head. Case 7 no longer pins the federal exclusion's allocation, which #10175 changes. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…ts' estate losses Address the independent review of #10175: - California's Schedule CA subtraction and Wisconsin's unemployment compensation subtraction now read tax_unit_taxable_unemployment_compensation, and Rhode Island adds back the new tax_unit_tax_exempt_unemployment_compensation, so a dependent's own-return amount no longer lands on the filers' return. - Add estate_income to the dependent helper's loss sources: limited_business_loss deducts estate and trust losses for filers, and the helper's own comment says it lists the sources loss_ald deducts. This reaches the dependent's IRC 86 and 85(c) modified AGIs and the dependent's IRA deduction. - Fix two form-line citations (Form 1040 line 5b, Schedule 1 line 17) and state that the differential invariant covers compensation that is not negative. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
#10187 fixes the same three state lines (ca_agi_subtractions, ri_additions, wi_unemployment_compensation_subtraction) more broadly and has its own review, impact run and Axiom issues. Restore them to main here and drop tax_unit_tax_exempt_unemployment_compensation, which only Rhode Island used. #10175 should land after #10187, so the dependent routing never meets the old pooled state lines. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…egative Address round 2 of the independent review of #10175: - dependent_income_and_deductions now limits a dependent's business losses (self-employment, farm, rental, partnership/S corporation and, since the last commit, estate and trust) to business income plus the IRC 461(l) threshold, as limited_business_loss does for the filers. The threshold is infinite while the limit is suspended (2018-2020), so the 2020 exclusion test keeps the whole loss. - taxable_unemployment_compensation is floored at zero. irs_gross_income counts positive amounts only, so a net repayment never reduced gross income; a negative taxable amount let CA, PA and MA subtract it anyway. The exempt part carries the repayment, as before this PR, which Wisconsin's homestead income reads. - The property test now draws net repayments, and the differential invariant holds for every input. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
Hand-off to the US + core hub. Session bebaaf35, head
|
… in Montana's elderly homeowner/renter credit income (#10163) * Count SSI, TANF and workers' compensation in Montana's elderly credit income MCA 15-30-2337(9)(a) counts all nontaxable income, including (vi) cash public assistance and relief. Schedule 2EC line 16 (2024, 2025) names SSI and workers' compensation, and the 2014-2023 instructions list SSI payments and workers' compensation as disability benefits. Add ssi, tanf_person (the SPM unit's TANF per capita, so the household sum counts it once) and workers_compensation to every period of the gross income sources, with worked Schedule 2EC / Form 2EC cases for 2016-2025. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Add property tests for government assistance in Montana's elderly credit For a claimant, another adult and a child in four arrangements (one return or two, one SPM unit or two, and more assistance), gross household income adds every member's SSI and workers' compensation and the TANF once; the credit matches an independent reading of Schedule 2EC lines 18 to 30, does not depend on the arrangement, stays within the cap and never rises with more assistance. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Format the Montana government assistance property tests Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Pin SSI to zero in Montana elderly credit cases that test other income Gross household income now counts SSI, and the model computes SSI for a claimant aged 65 or older with little income (take-up defaults to true). Three gross household income cases and three property-test builders are about other items, so they now input no SSI; SSI has its own cases in the government assistance tests. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Count guaranteed income payments; move the cases to variable-named files Review follow-ups: gi_cash_assistance is outside federal AGI and 15-30-2337(9)(a) counts all nontaxable income (a government-funded pilot is also "any other government assistance" on Schedule 2EC line 16), so it joins every period of the sources. The worked cases move into mt_elderly_homeowner_or_renter_credit.yaml (and the SNAP guard into the gross household income file), with a guaranteed income case, the 2016 worksheet total corrected to line 10 and the references consolidated to one per document, with the 2009 MCA text for the 2009 period. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Count disability benefits and unemployment compensation outside federal AGI in Montana's elderly credit income (WIP) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Test the 2020 cutoff above the band the federal test double subtracts; add changelog Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Add a grid invariant test: disability benefits and unemployment compensation count once in every arrangement and year Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Pin SSI in the one-claimant property test; format reference titles #9900's one-claimant property test (now on main) builds claimants aged 65 or older with little income, for whom the model computes SSI, which gross household income now counts; its reference leaves SSI out, so the builder inputs none. Reference titles follow the repo format: no page in a single-page title, and quoted titles ending in "#page N-M" for several. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Name line 5 for disability pensions, describe the add-back's role edge, and test each list period's wiring Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Rename the add-back to mt_elderly_homeowner_or_renter_credit_unemployment_compensation_addback; supply Case 7's taxable allocation as an input Both review rounds noted that the helper adds more than nontaxable compensation for a person recorded as both a dependent and the head. Case 7 no longer pins the federal exclusion's allocation, which #10175 changes. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
PR Review: Fix the 2020 unemployment compensation exclusion: cutoff test, per-spouse cap, dependentsSummaryI checked the law before reading the code. The statute and form.
The PR matches the law on all three fixes:
Code and tests.
Open items.
CriticalNone. Should addressS1. Two new terms are unpinned: deleting either fails no test. (a) The business income term in the dependent's IRC 461(l) limit.
(b) The floor on the compensation removed in
SuggestionsG1. Extend the axiom line to the dependent-loss change.
G2. Refresh the PR body before squash. The squash commit will carry it.
G3. Follow-up issue:
G4. Invariant 2 restates the formula.
Verified correctLaw and form.
Code.
Hand checks (each person's own exclusion = min(own compensation, 10,200) if their return's line 6 < 150,000):
Other checks.
Axiom line assessmentLine: rulespec-us#1719 is valid.
The encoded-correct claim holds.
Coverage: see G1 for the § 461(l) part. CI statusAll 27 checks pass at That run predates several main-side guards: #10287 (source-conflict ratchet), #10317, #10408 and the #10439 Rest sharding. I checked them on the merged tree instead (see Branch status). Branch statusPosition: 7 commits ahead and 479 behind main, merge base
On the merged tree (one small run at a time), these pass:
Siblings (note only):
0 critical, 1 should and 4 suggestions are open at this head. Verdict: REQUEST_CHANGES |
DTrim99
left a comment
There was a problem hiding this comment.
Requesting changes: 0 critical, 1 should (two new terms are unpinned: business_income in the dependent's 461(l) limit, and the max_(0, ...) removed on UC in dependent_taxable_uc_agi; the report gives a case for each). All three fixes match § 85(c) and the 2020 worksheet, and on a tree merged with current main the #9558/#10317/#10408 guards, the ratchet and the PR's 45 cases pass. Full report in the comment above.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Pin business income in the dependent loss limit and preserve exclusion AGI after a dependent's net repayment. Add California and Rhode Island cases for dependents' separate returns, and remove the property assertion that repeats the exclusion formula. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
David, here's where each item of your review stands at
The branch merges cleanly with main, and all 36 checks pass at 🤖 Generated with Claude Code |
Commit 8147027 adds the cases and the test change this review asked for, and the description and two new issues cover the rest. My reply above maps each item to its file and line.
|
US + core hub merge audit, us#10175 at
🤖 Generated with Claude Code |
Main's #10175 gives each person their own IRC 85(c) exclusion and counts unemployment compensation in full in the dependent income measure, which dependent_adjusted_gross_income starts from. The dependent's own return leaves out the excluded part, so the new variable now subtracts it, and Utah relief counts it as nontaxable income for dependents and filers alike, as main does. A 2020 case and a 2021 case pin it. #10175 also nets a dependent's estate losses, so this PR no longer changes dependent_taxable_ss_magi, and the changelog drops that line. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* Count strike benefits other than gifts in federal gross income Publication 525 counts strike and lockout benefits a union pays as income unless the facts clearly show the union meant them as gifts, and Form 1040 reports them on line 1h from 2022. Publications 596 and 503 make the taxable benefits earned income. - Split strike_benefits into tax_exempt_strike_benefits (the gift part, an input) and taxable_strike_benefits (the rest). - Add taxable_strike_benefits to the federal gross income sources, EITC earned income, adjusted_earnings and the dependent earned income sources. - Pennsylvania subtracts it again (72 P.S. 7301(d)(vi)); Mississippi, Iowa, Alabama and Arkansas add it to their own income bases; Utah, Montana and Arizona now add only the gift part as nontaxable income, so nothing counts twice. Refs #9919 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Record strike benefits as earned income in the source-conflict backlog Section 32(c)(2)(A)(i) and Treas. Reg. 1.32-2(c)(2) don't name strike benefits; Publications 596 and 503 count taxable strike benefits as earned income. Add us:strike-benefits-earned-income (gap-fill) and point us:dependent-deduction-worksheet-earned-income at this PR, which adds line 1h to the worksheet's earned income. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Map taxable strike benefits to SPM resources in the parity test Every federal gross income source must reach SPM resources or carry a reason it doesn't (#9919). strike_benefits, the whole union payment, is in market income, so it counts the cash behind taxable_strike_benefits. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Add taxable strike benefits to the Arkansas line-comment map The code-health test pins each Arkansas gross income source to its AR1000F line for the year. Taxable strike benefits go on other income (Form AR-OI): line 20 on the 2015-2017 form and line 22 from 2018. Refs #9919 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Answer the first review: Utah dependents, pre-2022 years, shared cap, AMT and PA - Utah homeowner's and renter's relief counts an adult dependent's own adjusted gross income (59-2a-101(15)): dependent_gross_income less their own above-the-line deductions. - The dependent earned income list takes strike benefits from 2022, when Form 1040 line 1h first puts them in line 1z; before that the worksheet counted line 1 and Schedule 1 lines 3 and 6. - Utah, Montana and Arizona read one variable, strike_benefits_excluded_from_gross_income, for the part federal gross income leaves out, so a reform that drops taxable_strike_benefits from gross income still counts the whole amount once. - The kiddie tax support test and the AMT child exemption cap leave strike benefits out of earned income (IRC 911(d)(2); Form 8615). - Pennsylvania Tax Forgiveness eligibility income counts strike benefits (PIT Guide, Tax Forgiveness, pages 5-6). - The property test draws an adult dependent with wages and strike benefits, and sums Montana's per-person measure over the household. Refs #9919 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Answer the second review of strike benefits - Utah relief counts an adult dependent's own adjusted gross income (new dependent_adjusted_gross_income: losses netted within 1211(b), above-the-line deductions taken, taxable social security counted once), leaves out a parent's or grandparent's income under 59-2a-101(15)(b), and counts no nontaxable unemployment for a dependent. - California's AMT child exemption (Schedule P line 8, earned income for personal services) leaves out strike benefits, as the federal cap does. - Pennsylvania subtracts strike_benefits_in_gross_income, the part federal gross income holds under any source list, so a reform that lists the whole amount still keeps it out of PA taxable income. - A negative gift input no longer adds to taxable strike benefits. - The kiddie tax comment cites the 2017 Form 8615 instructions the model applies; the dependent worksheet comments name the pre-2022 lines. - New tests: source-list reforms across UT, MT, AZ and PA; strike benefits are not pay for personal services (federal and California AMT child measures); Utah dependents, losses and parents; a 2022 dependent filing test. The PA candidate cites the Tax Forgiveness and Gross Compensation guides. Refs #9919 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Answer the third review of the strike benefits PR Utah household income: the tax-unit AGI fallback honours the head's parent or grandparent flag, and the documentation says the flags describe a member's relation to their own tax unit's filer. A dependent's estate and trust losses now net in their modified AGI, as loss_ald nets them for filers. Pennsylvania: the compensation comment cites 72 P.S. 7301(d) ("received for services rendered") and drops the code-health exception; the Tax Forgiveness record cites 61 Pa. Code 101.1 and 123.3. Invariant 6 gains explicit examples that fail if any AMT or kiddie tax subtraction is dropped, and the source-reform test adds a head with an adult dependent and records why federal earned income keeps strike benefits under an exclusion reform. New YAML for dependent_adjusted_gross_income and three Utah cases. Refs #9919 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Answer the fourth review of the strike benefits PR strike_benefits_in_gross_income returns zero when no strike benefit source is listed, rather than add()'s empty result, which let an earlier-year input carry over. The Utah household income documentation names the multi-unit cases the parent and grandparent flags miss, and a dependent AGI test case uses a qualifying child. The source-reform docstring sets out what a reform for S. 2779 needs beyond the gross income sources (EITC election, refundable CTC phase-in under IRC 24(d)(1), the dependent earned income list). Comment and record fixes: Pennsylvania nontaxable list wording, the Gross Compensation anchor, Form 8615's 2022 worksheet sentence, and the Montana test citation. Refs #9919 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Answer the fifth review of the strike benefits PR - amt_exemption: the 2017 Form 8615 instructions name Form 1040 line 7 only where they total lines for a child whose trade or business uses no capital; say so, and quote their earned income words. - Utah household income: state that a flagged member the model selects as claimant has their own income left out. - The dependent AGI Social Security case gives the student a disability, since survivors benefits past 19 need one (42 U.S.C. 402(d)(1)(B)). - The S. 2779 note: adjusted_earnings takes earned income from IRC 32(c)(2) through Treas. Reg. 1.21-2(b)(3) and IRC 129(e)(2), so the bill's election reaches it too. - candidates.yaml cites the 2022 Form 8615 Alternate Worksheet. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Leave the gift part of strike benefits out of North Dakota renter's refund income #10697 merged first, so this PR reconciles the two. N.D.C.C. 57-02-08.1(5)(c) leaves out "any amount excluded from income by federal or state law", and IRC 102(a) excludes the strike benefits a union pays as a gift (Pub. 525). nd_renters_refund_federally_excluded_income now subtracts strike_benefits_excluded_from_gross_income, capped at the benefits received, so renter's refund income holds the part federal gross income holds under current law and under source reforms. The ND invariant test moves strike_benefits from the components gross income omits to the partly excluded ones and draws strike benefits with a gift part; the source-reform test adds North Dakota; ND YAML cases cover the gift part and a gift input above the benefits received. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Take a dependent's 2020 unemployment exclusion out of their AGI Main's #10175 gives each person their own IRC 85(c) exclusion and counts unemployment compensation in full in the dependent income measure, which dependent_adjusted_gross_income starts from. The dependent's own return leaves out the excluded part, so the new variable now subtracts it, and Utah relief counts it as nontaxable income for dependents and filers alike, as main does. A 2020 case and a 2021 case pin it. #10175 also nets a dependent's estate losses, so this PR no longer changes dependent_taxable_ss_magi, and the changelog drops that line. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Answer the sixth review's nits - The household North Dakota check now pins the part of strike benefits in federal gross income per scenario (8,500, 0, 10,000), so it no longer compares the model with itself. - us-pa:tax-forgiveness-strike-benefits quotes Schedule SP line 10's cafeteria plan item, which supports the payments-into-a-program reading. - The amt_exemption comment follows the 2017 Form 8615 wording on lines 7, 12 and 18. - The dependent AGI case cites 42 U.S.C. 402(d)(1)(B)(ii) and (G). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Fix the 2020 unemployment compensation exclusion, including the AGI it tests against the cutoff, each spouse's cap and dependents' separate returns. Also net dependents' estate and trust losses and apply their own excess business loss limit.
Law and calculation
Under IRC 85(c), taxpayers with exclusion AGI below $150,000 exclude up to $10,200 of their own 2020 compensation. Each spouse receives a separate cap on a joint return. The IRS leaves compensation out when calculating that AGI and applies a strict cutoff in its 2020 exclusion worksheet, lines 1–9.
For example, spouses who receive $20,000 and $5,000 exclude $10,200 and $5,000 when eligible, leaving $9,800 taxable. A taxpayer whose exclusion AGI reaches $150,000 receives no exclusion.
Changes
taxable_unemployment_compensationfor each person fromtotal_unemployment_compensation. Compare the filers'taxable_uc_agiwith the cutoff without subtracting the compensation a second time, and test dependents against their owndependent_taxable_uc_agi.tax_unit_taxable_unemployment_compensation, matching the filers' federal gross income. Indiana's unemployment deduction and the TAXSIM output receive that return total.tax_exempt_unemployment_compensation.Outside 2020, the exclusion amount remains zero. We still exclude dependents from the tax unit total and apply their own business loss limits.
Consumers and related work
Three related PRs merged first:
For 2020, we still count a dependent's full compensation in the qualifying relative gross income test, though IRC 85(c)(1) keeps the excluded part out of their gross income. Directly substituting taxable compensation would create a cycle through filing status and head of household eligibility. #10912 tracks that test, and TheAxiomFoundation/rulespec-us#2230 queues § 461(l) and estate and trust loss netting on a dependent's own return.
Validation
The eight changed YAML files hold 55 cases, 28 more than on main, and the new property module runs two tests.
Pull request run 43603299648 at
d3f91beaf6passed all 36 jobs on GitHub's merge with main3ab9b19a11.Five one-line changes each make a new case fail: dropping business income from a dependent's loss limit or the floor on their repayment, or counting dependents' amounts in the California subtraction, Rhode Island addition or return total.
Microsimulation impact
We estimate that this branch changes none of the outputs below in 2026 on the default dataset, apart from float rounding of under 2 cents in four tax units' AGI and two households' net income. We compared
mainata93cc58d7ewith this branch atd3f91beaf6.taxable_unemployment_compensationtaxable_social_securityadjusted_gross_incomestate_income_taxincome_taxhousehold_benefitshousehold_net_incomeCongress limited the exclusion to 2020, a year this run doesn't cover, so the household tests above show that change.
axiom: TheAxiomFoundation/rulespec-us#1719 queued; TheAxiomFoundation/rulespec-us#2230 queued (§ 461(l) and estate and trust loss netting on a dependent's own return; no corpus scope in the pinned release names § 461).
us:statutes/26/85already encodes § 85(c)(1) correctly: per-recipient $10,200 cap, joint return summed per spouse, strict<, with teststemporary_2020_exclusion_applies_per_spouse_on_joint_returnandtemporary_2020_exclusion_denied_at_threshold. It takes the § 85(c)(2) AGI and the recipient relation as free inputs, and TheAxiomFoundation/rulespec-us#1719 queues encoding both, with companion tests from the same worksheet cases.🤖 Generated with Claude Code