What Is the Hold-Harmless Prohibition in Provider Tax Programs?

In Medicaid financing, the hold-harmless prohibition on provider taxes is a federal rule that denies matching funds on any tax revenue a state collects from health care providers if the state turns around and shields those providers from the cost of the tax. The prohibition sits in 42 U.S.C. § 1396b(w) and its implementing regulation at 42 CFR 433.68, and it is one of three conditions a provider tax must satisfy to count toward the state’s share of Medicaid spending, alongside being broad-based and uniformly imposed.¹²

The concern behind the rule is a specific financing maneuver. A state taxes its hospitals, nursing facilities, or other providers, uses the tax revenue to draw down federal Medicaid matching dollars, and then routes payments back to the same providers so that no one actually bears the tax. On paper the state has funded its share of Medicaid. In practice the federal government has funded both shares. The hold-harmless test is how CMS separates a genuine state contribution from that kind of round trip.

The Three Conditions That Trigger the Prohibition

A provider is considered held harmless if any one of three conditions is present. Tripping a single condition is enough to make the tax revenue impermissible.³

Correlated Non-Medicaid Payments

The first condition looks at money flowing to providers outside the Medicaid program. A hold-harmless arrangement exists if the state, or another unit of government, makes a direct or indirect non-Medicaid payment to taxpayers, and the amount of that payment is positively correlated either to the tax itself or to the gap between what the provider paid in tax and what it received in Medicaid reimbursement.³

“Positive correlation” is read broadly. The regulation covers any positive relationship between payment and tax, including one that is inconsistent over time. Random variation or occasional breaks in the pattern do not defeat the test. If CMS can identify a general upward relationship, the condition is met.

Medicaid Payments That Depend on the Tax

The second condition applies inside the Medicaid program. It is triggered when some or all of a provider’s Medicaid payment varies based on how much tax the provider paid, rather than on the services delivered, or when a Medicaid payment is conditioned on the provider having paid the tax.³

This is where most engineered financing arrangements fail. A supplemental payment pool can be labeled service-based, but if the actual distribution tracks who paid what in tax, CMS treats it as tax-dependent. The agency looks at where the dollars land, not the stated methodology. Medicaid payments are supposed to reflect care delivered to patients, not contributions to the state treasury.

Direct or Indirect Guarantees

The third condition is the broadest. It captures any payment, offset, or waiver that directly or indirectly guarantees a provider will not bear the actual cost of the tax.³ An explicit guarantee might appear in a statute or contract that promises reimbursement levels tied to the tax. An implicit guarantee comes from the overall structure of a program: if providers face no real financial risk from the tax, the guarantee condition is met whether or not anything is written down.

The Safe Harbor Percentage

The guarantee prong runs on a numerical presumption. A tax at or below 6 percent of net patient revenue is presumed not to involve an indirect hold-harmless arrangement. Above that percentage, the arrangement is presumed to hold providers harmless, and the state must rebut the presumption to keep the federal match.²

The 6 percent figure is no longer the ceiling in every state. The Working Families Tax Cuts legislation, enacted in 2025, changed the safe harbor for state fiscal years beginning on or after October 1, 2026. States that have not adopted the Medicaid expansion are frozen at whatever percentage was in effect for each class of services on July 4, 2025; where a non-expansion state had no tax on a class as of that date, its safe harbor for that class is zero. Expansion states face a gradual reduction of the threshold over several years.²

CMS guidance issued in 2025 set the transition schedule. Taxes on managed care organization services must comply by the end of the state fiscal year ending in 2026. All other permissible tax classes have until the end of the state fiscal year ending in 2028.⁴

The 2026 Rule on Tiered Taxes

In February 2026, CMS finalized a rule that sharpens what counts as an indirect hold-harmless when a state uses tiered tax rates. Some states had built structures that imposed higher rates on the Medicaid-related portion of provider activity than on the non-Medicaid portion. The math cleared the existing statistical tests, but the practical effect was to concentrate the tax on Medicaid dollars, which then came back to providers through federally matched Medicaid payments.⁵

The rule adds a bright-line prohibition at 42 CFR § 433.68(e)(3). A tax is not considered generally redistributive if any of the following is true:

The tax rate applied to any provider’s Medicaid-related units is higher than the rate applied to that provider’s non-Medicaid units.
Providers grouped by higher Medicaid volume pay a higher rate than providers grouped by lower Medicaid volume.
The tax uses proxy terminology that achieves the same targeting effect without naming Medicaid, such as references to “joint federal and state health care programs” or to income levels closely tied to Medicaid eligibility.

CMS declined to exempt any provider class, including managed care organizations and nursing facilities, and it rejected proposals to carve out nonprofit emergency medical services agencies, stating that the authorizing legislation does not permit provider-type exceptions.⁵

What Happens When a Tax Fails the Test

When CMS determines that a provider tax violates the hold-harmless prohibition, the revenue from that tax is subtracted from the state’s total reported Medicaid expenditures before the federal match is calculated. The federal government matches only what remains after the reduction.⁵ The same treatment applies to revenue from a tax that fails the broad-based or uniformity standards without an approved waiver.²

The dollar impact is larger than the tax itself. A state collecting $100 million through a prohibited tax does not simply lose $100 million; it loses the federal match on that amount, which in most states runs another $100 million to $300 million depending on the state’s Federal Medical Assistance Percentage. The state then has to close the gap from its own general revenue or reduce Medicaid spending.

Appealing a Disallowance

A state that disagrees with a CMS determination can appeal to the HHS Departmental Appeals Board. The notice of appeal is due within 30 days of the final written decision and must include a copy of that decision, the dollar amount in dispute, and a brief statement of why the state believes the decision is wrong.⁶

Filing generally freezes enforcement. The federal government cannot implement the disallowance or withhold funds while the appeal is pending, subject to narrow exceptions where a statute or regulation specifically authorizes recovery during the appeal period. The state must also have exhausted any preliminary dispute resolution process required by applicable regulations before the Board will hear the case. The 30-day clock is firm; missing it forfeits the administrative challenge.

1
eCFR. 42 CFR 433.68 – Permissible Health Care-Related Taxes
2
Office of the Law Revision Counsel. 42 USC 1396b – Payment to States
3
eCFR. 42 CFR 433.68 – Permissible Health Care-Related Taxes – Section: Hold Harmless
4
Centers for Medicare & Medicaid Services. CMS Issues Guidance to Strengthen Oversight of Medicaid Financing
5
Federal Register. Medicaid Program; Preserving Medicaid Funding for Vulnerable Populations – Closing a Health Care-Related Tax Loophole
6
eCFR. 45 CFR Part 16 – Procedures of the Departmental Grant Appeals Board