Cap tax exclusion for employer health insurance at $50K. Raises ~$450B over 10 years.
Ten-year deficit effect, conventional-$449.5Breduces the deficit
With macroeconomic feedback and debt service-$503.7BFRB/US-Lite (Federal Reserve calibrated)
Checked against
Published score -$450.0B; this estimate is 0.1% away (calibrated reference). Calibrated to reproduce -$450B (CBO) — agreement is by construction, not an independent test.
Typical error for this kind of policy: -$481.4B to -$417.6B, from 2 out-of-sample tax expenditure cases.
Out-of-sample accuracy, tax expenditure: 2 pre-registered rows miss their published scores by 7.1% on average (median 7.1%), worst 12.8% — 1 of the 2 falls inside the average. On this figure that is -$481.4B to -$417.6B typical, -$507.0B to -$392.0B at the worst row. Read off the out-of-sample tier only — not the calibrated tiers, whose agreement is by construction. It is the observed spread of this model against published scores for this class of policy, not a confidence interval.
Approximate bracket method: this policy cannot run through the return-level microsimulation, so read the shape as directional rather than as a validated table. Groups are ranked by adjusted gross income per return. Average change is in dollars per return; positive means the group pays more tax.
Economic effects
difference from the CBO baseline
GDP, summed over the window-0.56percent-years
Jobs at the largest effect-70K
Deficit effect of the GDP change+$38.9Bless tax collected
Deficit effect of interest on the debt-$93.1Bless debt, less interest
Fiscal year
GDP level
Jobs
Unemployment
10-year rate
GDP change, deficit
Interest, deficit
2026
-0.094%
-60K
+0.05 pt
-0.00 pt
+$6.6B
-$1.5B
2027
-0.110%
-70K
+0.06 pt
-0.01 pt
+$7.7B
-$3.1B
2028
-0.098%
-62K
+0.05 pt
-0.01 pt
+$6.8B
-$4.7B
2029
-0.077%
-50K
+0.04 pt
-0.02 pt
+$5.4B
-$6.4B
2030
-0.058%
-37K
+0.03 pt
-0.02 pt
+$4.1B
-$8.1B
2031
-0.042%
-27K
+0.02 pt
-0.03 pt
+$3.0B
-$9.9B
2032
-0.030%
-19K
+0.02 pt
-0.03 pt
+$2.1B
-$11.8B
2033
-0.021%
-14K
+0.01 pt
-0.04 pt
+$1.5B
-$13.8B
2034
-0.015%
-9K
+0.01 pt
-0.04 pt
+$1.0B
-$15.8B
2035
-0.010%
-7K
+0.01 pt
-0.05 pt
+$0.7B
-$18.0B
FRB/US-Lite (Federal Reserve calibrated). Demand effects fade over five to seven years as monetary policy responds; there is no supply-side channel. The dynamic score is the conventional score, plus the deficit effects of the GDP change and of interest on the debt: -$449.5B + $38.9B - $93.1B = -$503.7B.
Other scoring models
exploratory
Model
Ten-year deficit effect
Method
CBO-Style
-$449.5B
Static + ETI behavioral response + optional FRB/US dynamic feedback
TPC-Microsim Pilot
Cannot score this
Cap Employer Health Exclusion at $50K (TaxExpenditurePolicy) does not map onto current microsim pilot reforms (income-tax rates, CTC, EITC, SALT, std deduction, AMT exemption).
No pilot model returned distributional output; the comparison is aggregate-only.
CBO-Style is the model behind the headline above. The TPC-style microsimulation is a pilot: where the two disagree, the gap tells you how much the answer depends on method. It is not an official range.
Limitations
holdout labels follow the locked post-2026-05-02 regression protocol; they are not retroactive historical out-of-sample claims.
Target revised (target_revisions.eliminate_mortgage.v2) to a published RANGE [-$495.0B, -$367.9B] with Tax Foundation's -$367.9B as the carried anchor. The percentage this row reports is a distance from one modeller's point, not a measurement of accuracy; read `within_published_range` and `distance_to_published_range_billions` instead. No agency has scored repeal: CBO has published no post-TCJA option, JCT publishes the tax expenditure, and there is no FY2027 Green Book.
Yale's own 'close to $1.2 trillion' is not a bound: it is scored against pre-P.L. 119-21 current law, where TCJA's larger standard deduction lapses and the itemising population roughly doubles.
The record's annual_cost = 25.0 is a pre-P.L.119-21 level. JCT's JCX-45-25 puts the capped expenditure at $45.5B in FY2025 rising to $54.9B in FY2029, because raising the SALT cap to $40,000 took itemising claimants from 11.8M to 17.8M returns; Treasury's FY2027 edition gives $23.9B falling to $14.1B on the same statute. Choosing between them is an owner decision with a visible consequence for this row, not a provenance lane's.
The target and this benchmark's twin `eliminate_salt` state opposite baselines: repeal is priced against a permanent $10,000 cap and elimination against CBO's world where the cap has lapsed. Both are now stated rather than hidden; reconciling them needs a baseline-vintage concept the expenditure module does not have.
P.L. 119-21 sec. 70120 replaced the $10,000 cap with $40,000 for 2025-2029 (indexed 1%/yr, phased down above $500,000 of MAGI, never below $10,000) reverting permanently to $10,000 in 2030, so 'repeal the $10,000 cap' describes no live reform for most of the window. JCT's own row for that provision is +$946.2B over FY2025-2034 and is already carried as `pl119_21_salt_cap_40k`.
CBO measures the option on a baseline where the $10,000 cap lapses after 2025. P.L. 119-21 has since replaced that world with a $40,000 cap through 2029 reverting to $10,000 in 2030, so the target's baseline is no longer current law -- a model gap, not a target one, and one that needs a baseline-vintage concept the module does not have.
Reclassified rather than retuned: the module's behavioural magnitude for a charitable benefit-rate ceiling was an unsourced 0.40, and this annual had been chosen so that static x 1.40 lands on -$200.0B. Lane H7 derived the magnitude from CRS R40518's central price elasticity of giving (0.5) through the deduction's own SOI distribution, giving 0.220780, and the row moved 0.3% -> 12.5%; no constant was moved to close it.
The -$200B target is itself `secondhand` and sits in EXAMINED_NOT_REVISED: the 28% limitation it points at is a real Green Book proposal with a real score, but that proposal limits the value of ALL itemized deductions plus several exclusions, so its $645,538M is not a charitable-only cap. The residual is therefore a distance from an unsourced figure rather than a measurement against a document.
Cite asWilse-Samson, Laurence. “Cap Employer Health Exclusion.” Fiscal Policy Calculator, CBO February 2026 baseline, FY2026-2035. https://fiscal.lwilsesamson.com/explore/cap-employer-health-exclusion