[MaxGhenis]

This updates docs/parity-ledger.md for what merged after its last check (#590): B0.2's plan, protocol and counts (#584), S0.1 (#588), SSI0 (#573), G part 2 with R2 part (b) (#563) and the first part of D4 (#574), along with OP1's random-stream change (#591). At 97003ac2, master after #574, it re-verifies six rows, the critical path's opening and its steps 1 and 3, the section on the 24-month plan and ruling d1338's row. It changes no other file.

Rows at 97003ac2

• Mortality differences (S0.1: source, exposure, access and power audit for SM2–SM4 and ED1–ED2 #588, Serve OP1's adapter draws from DirectRNGStreams by default #591). S0.1 merged as a frozen audit record that admits no gate (docs/design/s0_1_source_audit.md:1-4, 43-50). The tracking cell now says it merged, and the next step starts at SM2; SM3 still follows a ruling (docs/design/mortality_groups_financing.md:397-398). OP1's random-stream test, merged in Serve OP1's adapter draws from DirectRNGStreams by default #591, imports SM1's step and runs it on an invented roster with an invented model (tests/open_population/test_stream_injection.py:91, 5200, 5273-5287, 5344-5350). The sentence on who names SM1 now lists that test. A scoped rg over src/, scripts/ and tests/ finds no projection, runner or gate that calls SM1.
• Education (S0.1: source, exposure, access and power audit for SM2–SM4 and ED1–ED2 #588). The tracking cell says S0.1 merged, and the next step starts at ED1; ED2 still waits on the Microcosm start (docs/design/mortality_groups_financing.md:397, 399).
• Family maximum (Track B G part 2 and R2 part (b): deductions before reduction; SSA's operational saving clause #563). G part 2 applies the 1972 savings clause under SSA's operational reading by default and keeps the literal reading's refusal behind SavingsClauseReading.LITERAL_REFUSAL (src/populace_dynamics/track_b/gross_benefits.py:242-252, 5660-5664, 5820, 5861). It refuses by name an unknown history, a payment procedure its captured sources leave unsettled and a change among the worker and beneficiaries subject to the family maximum (:475-477, 5667-5677; tests/test_track_b_gross_benefits_saving_absence.py:134). It still refuses beneficiaries entitled as auxiliaries or survivors on more than one record (gross_benefits.py:473, 5371-5394). family_benefits_after_deductions deducts first, then redistributes the maximum (:4591, 4602). The worked-example registry grows from 29 to 37 entries, all eight new ones from POMS (tests/data/track_b/gross_benefit_examples.yaml). The row drops its draft wording, and the next step goes straight to I.
• Earnings test (Track B G part 2 and R2 part (b): deductions before reduction; SSA's operational saving clause #563). src/populace_dynamics/track_b/ret_ledger.py no longer defines DEDUCTION_BEFORE_REDUCTION. R2 passes an auxiliary's own earnings-test deduction to G, which readjusts the family's rates under a binding maximum (:1611-1633). R2 refuses with FAMILY_CONTEXT_REQUIRED when the record lacks that month's family facts (:121, 1611). The row says R2 part (b) merged, and the next step names only H and I.
• SSI (Federal SSI interface and source-to-input contract (SSI0) #573, Aggregate-table captures and readers, part 1 (D4) #574). SSI0 calls only policyengine.py's public API (src/populace_dynamics/income_resources/ssi/_child.py:1) and pins policyengine 6.10.2 with PolicyEngine-US 2.2.1 (ssi/pin.json:58-60). It computes potential federal SSI under a frozen annual state, giving each month a twelfth of annual income and holding resources constant for the year (docs/analysis/ssi0/source_contract.md:75-77; ssi/runner.py:99-104). It keeps every attempted case in its denominator (ssi/runner.py:1-11). It refuses any household with a member under 65 (:296-301), couples where either member lacks citizenship or the take-up flag (:249-260), retirement accounts with a balance or periodic-payment eligibility (:331-336) and joint claims by tax dependents (:261-271), among other cases. Its input contract gives each inventoried SSI-chain input, for the PSID, SIPP and Microcosm, a source, an assumption or a refusal, and a source names a route that awaits a later milestone (docs/analysis/ssi0/source_contract.md:5-7; ssi/input_contract.json). Only its own tests call the SSI runner; they use invented households (tests/income_resources/ssi_invented.py:1). The wheel-pinning script loads the child worker only to call its digest helper (scripts/income_resources/pin_ssi.py:22-28, 99-101). D4's part 1 pins POMS SI 01415's supplement payment tables (src/populace_dynamics/income_resources/aggregate_poms.py:1), and its SSA registry holds only Tables 4.B1, 4.B2, 4.B10 and 4.B11, so SSA's SSI tables wait for part 2 (data/external/snapshots/income_resources_d4_v1/ssa_registry.json; aggregate_tables.py:1-16). The next step starts at SSI1 and SSI2 (docs/design/income_and_resources.md:208-210).
• Wealth, pensions and other income (Aggregate-table captures and readers, part 1 (D4) #574). D4's pension reader pins report-only WP8 inputs (src/populace_dynamics/income_resources/aggregate_pensions.py:1). Every entry in its registry carries the role report_only_wp8. The entries come from the Federal Reserve's Financial Accounts, the Labor Department's Private Pension Plan Bulletin and IRS Statistics of Income tables on individual retirement arrangements (data/external/snapshots/income_resources_d4_v1/pensions_registry.json). The tracking cell now lists D4 with the merged work instead of the drafts. Outside D4's own tests, only the birth-evidence exclusion list names its readers, so the status stands.
• Critical path (Track B G part 2 and R2 part (b): deductions before reduction; SSA's operational saving clause #563, Track B B0.2: source, exposure, access and power records for B3 through J #584). The opening lists G part 2 with R2 part (b) among the built rules. Step 1 says B0.2's plan, frozen protocol and counts merged in Track B B0.2: source, exposure, access and power records for B3 through J #584, which replaced the closed, unmerged Track B B0.2: source, exposure, access and power records for B3 through J #554 (docs/design/track_b_b0_2_plan.md:3-8; docs/design/track_b_b0_2_protocol.md:3). It also says the per-gate records and their reviews come next (plan §3, :119-145). Step 3 says G part 2 merged.
• The 24-month plan (Track B G part 2 and R2 part (b): deductions before reduction; SSA's operational saving clause #563). R2 part (b) leaves the list of missing pieces, since Track B G part 2 and R2 part (b): deductions before reduction; SSA's operational saving clause #563 built it.
• Ruling d1338 (Track B G part 2 and R2 part (b): deductions before reduction; SSA's operational saving clause #563). The row drops "not yet built" and "(draft)".
• Header. It now records the 3 November verification. Between 29 October and 3 November 2026, B0.2's plan, protocol and counts merged, and so did S0.1, SSI0, G part 2 and the first part of D4.

Links

One definition joins the list, [pr591]. Definitions go from 89 to 90 and uses from 249 to 246, with none unresolved, none unused, no duplicate definitions and the 4 inline links unchanged. Four next-step cells drop a merged PR (#588 twice, #563 twice), the SSI row's #573 moves from the next step to the status, and the mortality status gains #591.

Merges that change no claim

• Held-out data registry: round 7's follow-ups (unset event path, record ids, zero push boundary) #580 and Held-out registry: deeply nested events raise LookupError; test the 64-digit zero before #592 harden the held-out data registry and its tests (src/populace_dynamics/track_b/holdout_registry.py); the ledger cites the registry only where it records Track B: held-out data registry for every Track B and open-population gate #560's merge.
• Flush the real streams before each B0.2 test captures output #596 adds a fixture that flushes the real streams before each B0.2 count test; no ledger claim depends on it.
• Beyond the SM1 sentence, Serve OP1's adapter draws from DirectRNGStreams by default #591's default stream class (open_population/gates.py, harness.py) changes no claim in the births or immigration rows, which still say OP1 ran on invented data only (docs/design/open_population_evidence/op1_report.md).

Wording that stands

• Track B B0.2: source, exposure, access and power records for B3 through J #554 closed unmerged. The d1802 row still cites it as public record.
• The ledger's other PRs in review (Track B insured status: quarters of coverage; fully, currently and disability insured #562, Retirement-plan rule layer (WP2) #572, Design: HRS public-file addendum to the income-and-resources design #579, SIPP income, assets, pension and SSI readers (D2) #581, Coverage-rule registry (CE1) #583, Registration 18 (A-COLA) build: stage 1 on master, stage 2 and close-out (blind replay) #585 and F1: income tax on benefits split between OASDI and HI (Chapter 1 only) #586) remain open drafts in the 10:50Z list of open PRs from 3 November, so their wording stands.

Checks

Every changed claim cites code, tests or docs at 97003ac2, never a PR body. Merge status comes from master's merge commits, and open status from the 10:50Z list. I made no network or gh calls and ran no test suite, since the change touches one Markdown file.

🤖 Generated with Claude Code

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated
social-security-model	Ready	Preview	Nov 9, 2026 12:35pm UTC