Notes and Methodology

1. Purpose of the v.0.1.0 estimate

This estimate implements an institution-level public-data proxy for the post-2025 Section 4968 base. The model builds a public-data proxy universe, filters it under the post-2025 coverage rules, and adds an explicit module for post-2025 net investment income addbacks: student-loan interest and federally subsidized royalty income.

2. Post-2025 law mapping

The model maps current Section 4968 into executable tests. The covered-institution screen applies a 3,000 tuition-paying student threshold, a more-than-50-percent U.S. student-location test, a $500,000 student-adjusted endowment threshold, and a state-college exclusion. It then assigns current-law rates of 1.4 percent, 4 percent, or 8 percent by student-adjusted endowment tier. The model also includes the statutory NII overrides for student-loan interest and federally subsidized royalty income. These amendments apply to taxable years beginning after December 31, 2025.[1]

3. Base bridge and universe construction

The starting public tax anchor remains IRS SOI Publication 6038 for CY2024. IRS reports 45 Schedule O returns and $168,823,327 of tax on net investment income of private colleges and universities. Dividing that reported tax by the 1.4 percent CY2024 rate implies $12.059 billion of old-law taxable NII.[2]

Because public SOI data do not identify taxpayers institution by institution, the model allocates the old-law aggregate NII across a public proxy universe using FY2024 endowment market values from the 2024 NACUBO-Commonfund public table. Candidate rows are private nonprofit institutions with at least 500 FTE students and at least $500,000 of endowment per FTE student, approximating the old-law coverage screen. The dollar allocation is scaled to reconcile exactly to the IRS aggregate.[3]

4. Institution-level calculation

For each covered institution row and each budget-window year, the workbook calculates old-rule NII by applying the CBO-based cumulative growth factor to the retained CY2024 NII allocation. It then adds the scenario-specific expanded NII addback, calculates current-law tax using the institution-specific 1.4 percent, 4 percent, or 8 percent rate, and calculates proposal tax at 21 percent after the scholarship/aid exclusion, behavioral reduction, and loss-adjustment assumptions. The gross receipts estimate is proposal tax minus current-law tax, summed across institutions. CBO growth inputs use PCE inflation plus potential total factor productivity growth;[4] income/payroll offsets use JCT excise-tax offset factors.[5]

5. Expanded NII addback module

The Expanded NII Addbacks sheet provides direct override columns for institution-specific student-loan interest and federally subsidized royalty income. Where direct values are not entered, the public-data proxy applies scenario factors to retained old-rule NII. The central case applies a 0.05 percent student-loan-interest factor, a 0.40 percent royalty factor for doctoral institutions, and a 0.10 percent royalty factor for non-doctoral institutions. These proxy factors are deliberately transparent and editable; they can be replaced with audited-statement, Form 990, technology-transfer, or student-loan-receivable data where available.

6. Results

The central estimate is $10.2 billion of net federal receipts over FY2027-FY2036. The low and high cases are $8.5 billion and $12.6 billion, respectively. The scenario range is driven primarily by scholarship/aid exclusion assumptions, behavioral reductions, current-law tier mix, and the size of the expanded NII addback.

7. Limitations and interpretation

This institution-level public-data proxy is not a return-level audit. The largest remaining uncertainty is that public data do not provide each taxpayer's Schedule O net investment income, statutory daily-average tuition-paying student count, direct-use asset exclusions, related-organization asset and income adjustments, or actual student-loan-interest and federally subsidized royalty addbacks. The model therefore separates direct override columns from proxy formulas so future institution-specific evidence can replace assumptions without redesigning the model.

The row count should not be interpreted as a confirmed IRS taxpayer count. The old-law public proxy contains 46 candidate rows, compared with 45 Schedule O returns in the IRS aggregate. The model reconciles dollars to the IRS aggregate but uses public endowment/FTE data to allocate the base and apply the post-2025 filter.

Appendix A. Scenario assumptions and totals

[1] 26 U.S.C. Section 4968, current post-2025 text, https://uscode.house.gov/view....(title:26%20section:4968%20edition:prelim).

[2] IRS Statistics of Income, Publication 6038 (Rev. 1-2026), Excise Taxes, Calendar Year 2024, https://www.irs.gov/pub/irs-pd....

[3] NACUBO Public NCSE Tables, 2024 NACUBO-Commonfund Study of Endowments, https://www.nacubo.org/Researc....

[4] Congressional Budget Office, The Budget and Economic Outlook: 2026 to 2036, https://www.cbo.gov/publicatio....

[5] Joint Committee on Taxation, JCX-10-25, Income and Payroll Tax Offsets to Changes in Excise Tax Revenues for 2025-2035, https://www.jct.gov/getattachm....