Calibrate OBBBA deduction aggregates to JCX-45-25 static tax-expenditure lines

[MaxGhenis]

JCX-45-25 (December 2025 tax-expenditure report, post-OBBBA) publishes static, each-alone FY lines for four of the five OBBBA individual provisions — Table 1, individuals column (mapping verified on the employer-FICA-tips credit row and by row-sum checks):

Provision	FY2026 ($B)	FY2027 ($B)	Downstream repeal score on Build P ($B)	Diff vs FY2026
Deduction for overtime pay	21.9	22.4	22.2	+1.5%
Enhanced deduction for seniors	22.4	22.7	18.5	−17%
Deduction for tips	6.9	7.2	1.8	−73%
Deduction for interest on personal vehicle loans	6.2	8.6	1.1	−83%
Non-itemizer charitable	— (folded into charitable TE)	—	3.9	—

Proposal: add these as calibration targets (or validation-surface benchmarks) in the jct.tax_expenditures source family — same schema as #519's five entries. Notes from the decomposition:

• Tips is a yield gap, not a base gap: Wire the W-2 tips dollar target; add tips and overtime coverage probes (#451) #465 wired the W-2 tips dollar target and the income base sits near the BLS anchor (~$34B), yet the deduction's repeal yields $1.8B vs the $6.9B static line. An amount target on tip income can't close this — the deduction-level line (or a repeal-revenue benchmark per QBI base −31%/−40% vs targets and JCT; add per-component repeal-revenue benchmarks (amount residuals can't see yield gaps) #298) is the right ref, and the residual likely lives in eligibility mechanics and where tips sit in the rate distribution.
• Auto-loan interest base is thin outright (qualified_passenger_vehicle_loan_interest is never imputed — OBBBA auto-loan deduction is structurally $0 #252's imputation gap persists into Build P; the deduction resolves but small).
• Overtime already lands within 2% of the static line at current claiming — the ~64%-of-Yale-anchor premium-base gap (Input-coverage readiness quartet: CDCC adult care, SE health premiums, tips, overtime (Build N scope) #451) binds for post-behavior or distributional uses, not the static aggregate, so a static-line target here is a cheap regression guard rather than a fix.
• Senior deduction runs −17% vs static despite age/SSN bases tracking Census — worth a look at the phase-out interaction once a target binds.

These four lines would complete static-frame coverage of the OBBBA individual provisions alongside the #519 ALD/CDCC refs. Source: https://www.jct.gov/getattachment/8c830c45-1680-4f7e-a649-2a0106f6b6e3/x-45-25.pdf