Skip to content

Read the same year's income for the Ohio homestead exemption - #11161

Merged
hua7450 merged 2 commits into
mainfrom
oh-homestead-same-year-income
Nov 20, 2026
Merged

hua7450 merged 2 commits into
mainfrom
oh-homestead-same-year-income

Conversation

@DTrim99

@DTrim99 DTrim99 commented Nov 19, 2026 •

Copy link
Copy Markdown
Collaborator

Part of #10812 (state credit income audit).

What was wrong

oh_homestead_exemption_total_income read oh_modified_agi at period.last_year. Inputs carry forward in policyengine-core, not backward, so the prior year has no data in a single-year household run. The read then returns 0, and the homestead exemption's income limit never binds: every aged, disabled or surviving-spouse owner passed the income test whatever their income.

What changes

  • Total income reads the same year's Ohio modified AGI. R.C. 323.151(C) (PDF page 1) defines total income as modified adjusted gross income under R.C. 5747.01 "of the owner and the owner's spouse for the year preceding the year in which application for a reduction in taxes is made". The model now uses the same year's Ohio modified AGI as a proxy for the prior year's, as or_tax_before_credits_in_prior_year does, so the exemption for year Y tests year Y's income. A known prior-year figure goes in through oh_homestead_exemption_total_income (eligible Case 8). The variable's documentation and the formula comment say that the law uses the prior year and the model uses the same year as a proxy. The authenticated statute PDF is added as a second reference; the existing reference is kept.
  • The income limit itself is unchanged: R.C. 323.152(A)(1)(b)(iii) (PDF page 1) requires total income that does "not exceed" the adjusted limit (40,000 in 2025, 41,000 in 2026 in income_limit.yaml).
  • I checked the other Ohio homestead variables (oh_homestead_exemption, oh_homestead_exemption_eligible, oh_homestead_property_tax_reduction, oh_homestead_exemption_qualifying_surviving_spouse). None reads income for another year. Age, ownership and residence timing are left as they are.

Who is affected

Ohio homeowners who qualify by age, permanent and total disability or as a qualifying surviving spouse, in single-year household runs without prior-year income:

  • Those whose same-year Ohio modified AGI is above the limit become ineligible, and their oh_homestead_exemption and oh_homestead_property_tax_reduction fall to 0. On main they always passed the income test.
  • Those at or under the limit are unchanged.
  • A household that supplies oh_modified_agi for two years is now tested on the exemption year's value instead of the prior year's.

No other variable reads the homestead variables, so Ohio income tax and household net income don't change.

Tests

Moved from #10812 (into oh_homestead_exemption_total_income.yaml):

  • Case 3, one year of inputs: 100,000 of wages at age 70 gives Ohio modified AGI 100,000, total income 100,000 (main: 0), and the owner is ineligible (100,000 > 41,000; main: eligible).

Added (in oh_homestead_exemption_eligible.yaml), both from wages so they run through oh_modified_agi:

  • Case 15: same-year modified AGI of 41,000, at the 2026 limit, is eligible.
  • Case 16: same-year modified AGI of 41,001, one dollar above the limit, is ineligible.

Existing tests changed:

  • oh_homestead_exemption_total_income.yaml Case 1: inputs unchanged (oh_modified_agi 35,000 in 2025, 99,000 in 2026); expected value 35,000 → 99,000, the 2026 value, since the 2026 exemption now reads 2026 income. Renamed as the same-year proxy default.
  • oh_homestead_exemption_total_income.yaml Case 2: inputs unchanged (39,000 in 2025, 60,000 in 2026); expected value 39,000 → 60,000 for the same reason. Renamed from "current-year ... is ignored", which is now the reverse of the rule.
  • oh_homestead_exemption_eligible.yaml Case 8: renamed from "prior-year total income controls eligibility" to say that a total income input overrides the same year's modified AGI. Inputs and the expected value (ineligible) are unchanged.

Not in this PR

The #10812 review lists no other Ohio gap. The audit notes two points this PR leaves alone:

  • Owners who received the reduction for tax year 2006, or for 2013 (2014 for manufactured homes), and qualifying surviving spouses of 2013/2014 recipients, get it with no income test under R.C. 323.152(A)(1)(b)(i)-(ii) (same PDF page 1). PolicyEngine has no input for prior receipt, so it applies the income limit to them too. With the limit now binding, the model can deny some of these grandfathered owners.
  • R.C. 323.151(C) combines both spouses' income. For spouses who file separately, the model reads one tax unit's modified AGI; the audit didn't trace that case.

axiom: n/a: same-year proxy only; the legal rule in R.C. 323.151(C) is unchanged, and rulespec-us has no R.C. 323.151 module

Tested on Modal: run w1-oh-a, 227/227 passed (every Ohio test file, every partner test file, and test_source_conflicts.py).

🤖 Generated with Claude Code

@codecov

codecov Bot commented Nov 19, 2026 •

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.00%. Comparing base (8999568) to head (68d5ce0).
⚠️ Report is 238 commits behind head on main.

Additional details and impacted files
@@            Coverage Diff            @@
##              main    #11161   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files            1         1           
  Lines           24        11   -13     
=========================================
- Hits            24        11   -13     
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.

@DTrim99
DTrim99 requested a review from hua7450 November 19, 2026 22:31

@hua7450 hua7450 left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks, David. The single-year fix is right: on main a household run with no prior-year input read 0 and qualified everyone, and Cases 15 and 16 now pin the <= at 41,000 through the real income chain. The 2025 and 2026 limits (40,000 and 41,000) are dated by tax year, as the Department's table has them. One fix before I approve, and it is wording only.

Before merge

  1. Describe this as a same-year proxy, not a re-booking. R.C. 323.151(C) counts income "for the year preceding the year in which application ... is made", so tax year Y tests modified AGI of Y-1 against the limit for Y. The formula now tests Y against Y, with the limit, amount and age all still read for the tax year. The documentation says the model "books the homestead exemption to that income year", which would need next year's limit and amount: 2025 modified AGI of 40,500 qualifies for tax year 2026 in law (limit 41,000), but the model tests it against 40,000. Please reword the documentation, the formula comment, the YAML header and the changelog the way or_tax_before_credits_in_prior_year does ("defaults to the current year's ... as a proxy"), and say that a known prior-year figure goes in through oh_homestead_exemption_total_income (Case 8 already shows the override).
    • Case 2's new name ("prior-year Ohio modified adjusted gross income under the limit does not qualify a filer over it") states the reverse of 323.151(C). Cases 1 and 2 should be named as the proxy default.
    • The changelog says microsimulation read an empty prior year. With the default 2024 dataset, 2025 and 2026 runs on main had prior-year data, so the fragment should claim single-year runs only.

Optional

  • Owners exempt from the income test. With the limit now binding, owners who had the reduction for tax year 2013 or 2006 (323.152(A)(1)(b)) are denied. Only the PR body records this; a comment at the income test in oh_homestead_exemption_eligible.py would keep it with the code.
  • An issue for the departure from the statute's income year, so it stays tracked.
  • Tests. Case 3 in oh_homestead_exemption_total_income.yaml covers the same class as Case 16 in oh_homestead_exemption_eligible.yaml, which is the boundary case.

DTrim99 and others added 2 commits November 20, 2026 10:51
The total income for the homestead exemption read Ohio modified AGI for the
prior year. With only one year of inputs, as in household runs and
microsimulation, that read returned 0, so the income limit never applied.
R.C. 323.151(C) uses the year before application; PolicyEngine books the
relief to that income year, so the exemption now reads the same year's
Ohio modified AGI.

Moves the one-year-of-inputs case from the #10812 audit, adds cases at and
one dollar above the 2026 income limit, and restates the two-year total
income cases for the same-year read.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…same-year proxy

- Reword the oh_homestead_exemption_total_income documentation and formula
  comment: R.C. 323.151(C) uses the year before application, and the model
  defaults to the current year's Ohio modified AGI as a proxy, which an
  explicit input overrides (same pattern as or_tax_before_credits_in_prior_year).
- Rename total income test Cases 1 and 2 as the proxy default; inputs and
  expected values are unchanged.
- Limit the changelog fragment to single-year runs and use the proxy framing.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@DTrim99
DTrim99 force-pushed the oh-homestead-same-year-income branch from 71a1a1c to 68d5ce0 Compare November 20, 2026 15:53
@DTrim99

DTrim99 commented Nov 20, 2026

Copy link
Copy Markdown
Collaborator Author

Thanks, Ziming. Fixed in 68d5ce0, rebased on main.

Fixed (your "Before merge" item)

  • Same-year proxy wording. The documentation, the formula comment, the YAML header and the changelog now describe the same-year read as a proxy for R.C. 323.151(C)'s prior year, following or_tax_before_credits_in_prior_year. They point to oh_homestead_exemption_total_income for a known prior-year figure (eligible Case 8 shows the override). The claim that the model books the exemption to the income year is gone.
  • Case names. Cases 1 and 2 in oh_homestead_exemption_total_income.yaml are named as the proxy default. Inputs and expected values are unchanged.
  • Changelog. It now claims single-year runs only.
  • PR body. Updated to match.

Skipped (optional, not in this round): a comment on owners exempt from the income test under 323.152(A)(1)(b), a tracking issue for the income-year departure, and the duplicate Case 3.

Verification: the Ohio homestead exemption tests pass locally (32/32) after the rebase. An independent check confirmed all three items and found no other changes. Ruff is clean.

@DTrim99
DTrim99 requested a review from hua7450 November 20, 2026 15:54

@hua7450 hua7450 left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks, David. The documentation, formula comment, YAML header and changelog now describe the same-year read as a proxy for R.C. 323.151(C)'s prior year and point to oh_homestead_exemption_total_income for a known prior-year figure. Cases 1 and 2 are named as the proxy default with main's inputs, and the changelog claims single-year runs only. Comparing the PR's own diff before and after the rebase, those wording changes are the only ones. Approving.

@hua7450
hua7450 merged commit d053a54 into main Nov 20, 2026
40 checks passed
@hua7450
hua7450 deleted the oh-homestead-same-year-income branch November 20, 2026 17:16
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants