Senior and Disabled medicaid needs more details

[daphnehanse11]

1. We don't have home value as part of asset limits
2. We lump them together for per capita spending
3. We don't include partial medicaid benefits

[MaxGhenis]

Reopening because the SSI asset-limit reform analysis exposed this as a live model gap, not just polish.

Observed severity from the local 2026 enhanced CPS no-SSI-asset-limit run using policyengine-us 1.700.0 / policyengine-us-data 1.115.5:

• SSI recipients: 5.797M -> 7.196M, change +1.399M.
• New Medicaid eligible through SSI-recipient pathway: +618.5k.
• New Medicaid enrolled: +584.1k.
• Medicaid cost: +$20.1B/year.
• The affected new Medicaid group was baseline category NONE; all passed Medicaid immigration eligibility; nearly all were income-eligible for the optional aged/disabled pathway, but 0 were asset-eligible.

The law supports the broad interaction: SSI recipients are generally a Medicaid eligibility group, while non-MAGI aged/blind/disabled Medicaid can still use resources tests. But our implementation is too coarse for analysis of SSI reforms and older/disabled Medicaid access. Priority sub-issues:

1. Calibrate SSI disability/take-up so asset-limit reforms do not treat every broad survey-disabled person as a successful SSI applicant. CBPP estimates a no-limit reform adds 462k SSI participants, versus our +1.399M.
2. Add missing non-SSI Medicaid aged/disabled pathways, especially medically needy/spenddown and Medicaid Buy-In for working disabled people.
3. Audit SSI-recipient Medicaid state handling: 1634 vs SSI-criteria vs 209(b), including Oklahoma and 209(b) states where blanket exclusion may understate coverage.

Keeping this as the umbrella issue for senior/disabled Medicaid completeness.