Repository navigation
Drop the fixed Illinois AABD entry from the person-adds ratchet - #10092
Merged
Merged
Conversation
#9902 made il_aabd_expense_exemption_person count only the person's own withholding, so it no longer adds the tax-unit state_withheld_income_tax. test_known_violations_only_shrink (#9897) then fails on main because the entry is still listed in KNOWN_VIOLATIONS. Remove it. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This was referenced Oct 11, 2026
Contributor
Author
|
Merge audit: head 404ee0c. |
MaxGhenis
added a commit
that referenced
this pull request
Oct 11, 2026
Brings in #10092, which drops main's stale il_aabd entry from the person-adds ratchet that failed this branch's Rest job. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
MaxGhenis
added a commit
that referenced
this pull request
Oct 11, 2026
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
MaxGhenis
added a commit
that referenced
this pull request
Oct 11, 2026
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This was referenced Oct 11, 2026
MaxGhenis
pushed a commit
that referenced
this pull request
Nov 2, 2026
Brings the branch up to date with 513 commits on main. Main has since dropped the stale Illinois AABD entry from the person-adds ratchet (#10092), which was the only failing check on this branch, and merged the IRMAA lookback fallback (#10106), whose test comment this branch now makes stale. The merge had no conflicts. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
MaxGhenis
added a commit
that referenced
this pull request
Nov 5, 2026
…10112) * Add 2015-2023 Medicare IRMAA brackets and 2015-2020 Part B premiums The Part B and Part D income-related monthly adjustment schedules started in 2024, so the backdating floor applied 2024 thresholds and amounts to benefit years 2015-2023, and the standard Part B premium applied its 2021 value to 2015-2020. Encode each year's schedule from the Federal Register Part B premium notices and CMS's Part D income-related adjustment releases: thresholds frozen through 2017, the MACRA brackets for 2018, the 85% tier from 2019 and CPI-U indexing from 2020, with the shorter married-filing-separately schedule (a single tier in 2018). Add YAML tests for every tier boundary in 2015-2023 and property tests for the statutory structure (ordering, joint and separate schedules, Part B adjustments relative to the standard premium, Part D tier proportions). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Merge main into irmaa-historical-brackets Brings the branch up to date with 513 commits on main. Main has since dropped the stale Illinois AABD entry from the person-adds ratchet (#10092), which was the only failing check on this branch, and merged the IRMAA lookback fallback (#10106), whose test comment this branch now makes stale. The merge had no conflicts. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Cite the Federal Register page of each IRMAA table and premium Each title now gives the page of the notice that prints the table it sources, not just the notice's first page, and the Part D files cite the same Part B tables for their thresholds, which 42 U.S.C. 1395w-113(a)(7)(A) takes from 1395r(i). The base premium file also cites the 2021-2023 notices, which state the standard premiums already in it. Every value cited was checked against the text of these pages. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Name each 2015-2023 IRMAA test case's source table A reviewer can check each expected value against one page: the Federal Register table for Part B and the standard premium, and the CMS memo or fact sheet table for Part D. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Draw IRMAA lookback properties from every published year The lookback test's comment said years before 2024 are backdated from 2024, which stops being true once these brackets land, so its monotonicity and zero-below-threshold properties now draw from 2015-2026. They keep 25 examples each, so this adds no run time. Each schedule property also states the provision it checks. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Merge remote-tracking branch 'origin/main' into pr-10112 * Keep two IRMAA lookback cases below the 2015-2016 threshold Main's #10106 cases 6 and 7 assert no IRMAA for a $100,000 MAGI in 2015 and 2016. They were written when 2015-2023 read the backdated 2024 schedule, whose first threshold is $103,001. With this branch's schedules, 2015 and 2016 use the $85,000 threshold that 42 U.S.C. 1395r(i)(6) holds at its 2010 level, so $100,000 lands in the 65 percent tier and both cases failed in CI. The cases test the lookback, not the tier, so lower the income to $80,000 to keep each name true: 2015 MAGI is 80,000 and 2016 MAGI deflates to 80,000 x 328,964 / 326,078 = 80,708.05, both below $85,000. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Merge origin/main into irmaa-historical-brackets Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Pin the highest MAGI of each IRMAA tier and drop a stale Medicaid note The 2015-2023 single and joint cases tested each tier's lowest MAGI (X + 1) but never X itself, so a "more than $X" bound stored as X survived every test. Add a person at each X above the first threshold, expecting the amount of the tier below, in Parts B and D. base_premium.yaml now starts in 2015, so the Medicaid comment saying it starts in 2021 is no longer true. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Merge main into irmaa-historical-brackets Cases 6 and 7 keep the $80,000 MAGI below the 2015-2016 threshold, on the per-capita AGI deflator that main now uses. Case 7 expects 80,584.71. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Main's Rest suite fails
policyengine_us/tests/code_health/test_no_group_variables_in_person_adds.py::test_known_violations_only_shrink:Two merges crossed. #9897 (merged 2026-10-10) added the ratchet with this entry listed as a known bug. #9902 (merged 2026-10-11 13:08Z) fixed the bug, making
il_aabd_expense_exemption_personcount only the person's own withholding, but did not remove the entry. Every PR whose CI merges current main now fails the Rest job; #10054's run 38142244518 is one.This removes the entry. The test passes locally on this branch (10 passed) and fails on main at f403afe (1 failed, 9 passed) with the message above. Test-only change; no model output changes.
axiom: n/a: test bookkeeping, no policy change
🤖 Generated with Claude Code