The Expanded Section 4960 Excise Tax Reaches Up to 28% More Nonprofit Employees — and Almost No New Organizations

What changed in the law

Since 2018, Internal Revenue Code Section 4960 has imposed a 21% excise tax on tax-exempt organizations that pay a "covered employee" more than $1 million in remuneration. Under the original rule, covered employees were limited to each organization's five highest-compensated employees (plus anyone previously covered — once covered, always covered).

The One Big Beautiful Bill Act (P.L. 119-21, signed July 4, 2025) struck that five-highest limitation. For tax years beginning after December 31, 2025, ANY current or former employee whose remuneration exceeds $1 million can be a covered employee. IRS Notice 2026-36 (June 5, 2026) sets out the interim framework; public comments on the notice close August 4, 2026.

Two points of precision the coverage often blurs. First, the $1 million threshold did not change — it was and remains the trigger. What changed is who can be a covered employee. Second, Section 4960 excludes remuneration paid to licensed medical professionals for medical services — which matters enormously, because physicians dominate the $1M+ population at nonprofit health systems. That exclusion is why this analysis publishes a range rather than a single number.

The numbers

Law firms, CPA firms, and the trade press have covered the rule change extensively — but qualitatively. To our knowledge this is the first data-based estimate of the expansion's reach. From each organization's most recent Form 990 Schedule J filing (a FY2023–24 snapshot; 67,405 organizations after curation):

The ex-medical column removes employees in clinical and mixed administrative-clinical roles (physicians, surgeons, medical directors, and the like) as a proxy for the statute's medical-services exclusion; the all-in column keeps them. The true figure sits between the two, and for reasons explained in the methodology, both are conservative undercounts.

The structural finding: more employees, not more organizations

The expansion adds covered employees, never newly exposed organizations. Any organization with even one employee over $1 million necessarily had its top earner inside the old top-5 — so the set of exposed organizations is identical under both rules. What changes is intensity: organizations that were already exposed now have more covered employees, in some cases many more.

Where it lands

By sector (ex-medical basis): hospitals and medical organizations see the largest increase in exposed employees (+38.6%), followed by education (+24.9%). By budget size, the effect is dramatically top-heavy: organizations with budgets above $500 million see exposed-employee counts rise by half (+51.1%), while below $25 million in budget the expansion is nearly a non-event (increases of 0–5%).

$500M+ +51.1% (+640 employees)

$100M–$500M +4.1% (+37)

$50M–$100M +14.0% (+39)

$25M–$50M +5.0% (+11)

$10M–$25M +3.9% (+6)

$5M–$10M +1.5% (+1)

$2.5M–$5M +0.0% (0)

Percentage increase in employees over $1M vs the old top-5 rule

Budget bands below $2.5M fall under the analysis's n ≥ 10 cell-publication minimum and are not shown.

Health — Hospitals & Medical +38.6% (+500 employees)

Education +24.9% (+91)

Unknown NTEE +16.9% (+118)

Public Affairs +11.5% (+7)

Community Improvement +2.4% (+2)

Percentage increase in employees over $1M vs the old top-5 rule

Sector cells below the analysis's n ≥ 10 cell-publication minimum are not shown.

What it means for boards and CFOs

For the vast majority of nonprofits, the honest answer to "does this reach us?" is no — unless the organization is a hospital or health system, a university (athletics compensation included), or operates at a nine-figure budget. For organizations in those categories, the planning question has changed shape: under the old rule the covered-employee roster was at most five names plus history; under the expanded rule, every employee crossing $1 million joins the roster — permanently, under the once-covered-always-covered principle.

Boards in the exposed categories should be inventorying who could cross the threshold in tax years beginning after December 31, 2025, and modeling the excise cost of pay packages that approach it. (This analysis is research, not legal or tax advice; organizations should consult their advisors on their specific facts.)

Methodology (summary — full methodology published with the report)

• Source: IRS Form 990 Schedule J e-filed disclosures in the CauseComp corpus; each organization's most recent filing (analysis set: 238,768 compensation rows across 67,405 organizations; 89% of exposed organizations filed FY2023 or FY2024).
• Measure: total reported compensation from the filing organization and related organizations; threshold strictly greater than $1,000,000. Old-rule baseline = the five highest-compensated per organization on the same measure.
• Medical segmentation: employees in clinical and mixed administrative-clinical roles are flagged by role and title (treating mixed roles such as Medical Director as fully medical, the conservative direction), producing the ex-medical primary estimate and the all-in upper bound.
• Robustness: recomputing on compensation excluding nontaxable benefits (closer to the statutory wages-based remuneration definition) moves the headline deltas by less than 5%.
• Why the estimate is a floor: Schedule J only discloses officers, directors, trustees, key employees, and the five highest-compensated employees. A $1M+ employee outside those categories is invisible in the data — so the additional-employee count can only be understated, not overstated. Filing lag points the same direction: compensation has grown since the FY2023–24 filings underlying this snapshot, and the expanded rule takes effect in tax years beginning after December 31, 2025.
• All figures derive from committed, deterministic analysis scripts; aggregates only — this report names no organizations and no individuals.

Caveats log

1. (a) Schedule J compensation ≠ the §4960 statutory remuneration definition exactly (remuneration is §3401(a) wages plus vested §457(f) amounts; Schedule J columns are calendar-year W-2/1099 figures plus deferred and nontaxable components). The ex-nontaxable sensitivity bounds this gap at ~5% of the delta.
2. (b) Schedule J only lists officers, directors, trustees, key employees, and the five highest-compensated employees — a $1M+ employee outside those categories is INVISIBLE in the data. The beyond-top-5 delta is therefore a FLOOR: a conservative undercount of the expansion's true reach. (This is the single most important caveat for the published range.)
3. (c) Filing-year lag: the analysis set is dominated by FY2023–24 filings (89% of exposed orgs), while the expanded rule applies to tax years beginning after 2025-12-31. Comp levels will have grown by the effective date; counts are again conservative.
4. (d) The medical flag is a title/role proxy, not a license check, and treats mixed administrative-clinical roles (e.g., Medical Director) as fully medical. The true statutory exclusion applies only to remuneration FOR medical services, so the ex-medical count is deliberately a floor and the all-in count a ceiling; the truth is between.
5. (e) "Former"-titled individuals (208 with >$1M) are excluded from BOTH rules' counts because old-rule once-covered status is unobservable from a snapshot; including them only in the expanded count would overstate the delta.
6. (f) Titles that cannot be classified ("SEE SCHEDULE O", bare "DIRECTOR") count as non-medical; some hidden medical staff may therefore remain in the ex-medical count. Direction: slightly overstates the ex-medical floor's level, does not affect the all-in ceiling.
7. (g) total_comp_all includes related-organization compensation; an individual compensated by two related filing orgs can appear under both EINs (dedupe is per-EIN). Health-system group structures may produce a small double-count in org-level and employee-level counts.
8. (h) 22% of ex-medical expanded-rule employees sit in orgs with unknown NTEE sector — sector cells understate true sector counts; the headline is unaffected.
9. (i) The panel covers e-filed Form 990s with Schedule J parsed into the corpus (71,130 orgs); paper filers and 990-EZ/PF filers are out of scope.
10. (j) Once-covered-always-covered means real covered-employee rosters are cumulative across years; this analysis compares single-year definitions, which is the policy-relevant "who does the new definition newly reach" question.

About CauseComp

CauseComp publishes compensation benchmarks for U.S. nonprofit organizations, built from IRS Form 990 filings and federal wage surveys — 1.9M+ compensation records. This analysis is free to cite with attribution and a link. The interactive version of every exhibit — benchmarks by role, budget size, sector, and state — is available in the CauseComp product at causecomp.org.

Frequently asked questions