By Michael Kinnucan, Health Policy Advisor
March 2024

Key Findings
  • The Senate and Assembly one-house budgets propose a Managed Care Organization (MCO) provider tax that would be cost-neutral for the State, but would generate $4 billion in increased Federal Medicaid revenue.
  • This complex proposal, inspired by a recent move in California, may face regulatory challenges at the Federal level.
  • Despite these challenges, an MCO tax will allow the state to draw down significant federal funds at no cost to state taxpayers, avoid home care wage cuts, and set the stage for further reforms.


The debate over Medicaid spending in New York has rarely been more divided. Governor Hochul, in her executive budget, declared Medicaid spending growth unsustainable and proposed sharp cuts to home care spending, financially distressed hospitals and several other items, drawing sharp criticism from advocates and industry groups. Meanwhile, 1199 and the Greater New York Hospital Association are campaigning for significant rate increases for hospitals and nursing homes.

The legislative one-house budgets come out firmly for higher Medicaid spending, restoring most of the governor’s cuts and offering significant rate increases. But how will they pay for it? The Senate and Assembly budget memos propose to raise $4 billion a year through an obscure mechanism: A tax on Medicaid managed care plans, the private insurance companies which administer most of the state’s Medicaid program. To quote the Assembly’s budget summary:

The MCO tax generates $4 billion in receipts from Managed Care plans. This revenue [will] be used by the State to repay the tax obligation for each plan through their capitated rates. This repayment generates an additional $4 billion in federal funding to then be used by the State as the non-federal share of investments in the Medicaid program.[1]

Despite appearances, this proposal won’t cost MCOs a dime. The Assembly and Senate are proposing to tax them while reimbursing them for the tax through a rate increase — and yet somehow this circular procedure will generate federal revenue.

How can that possibly work? Answering this question requires a journey into one of the more obscure corners of Medicaid policy.

Provider Taxes in the 1980s: Generating Federal Revenue Out of Thin Air

Medicaid is jointly funded by the state and federal government. Specifically, in New York, for each dollar of State spending on Medicaid, the federal government kicks in a dollar of federal funds. The more a state spends, the more the federal government contributes.

In principle, this system creates an opportunity for state governments. Suppose a state imposes a $100 tax on a given hospital, then spends that $100 on Medicaid payments to the hospital. When the money is spent as a Medicaid payment, it draws down another $100 in federal matching funds, so the hospital receives $200. The state has not lost any money (it taxed $100 and spent $100), but the hospital has gained $100 (it paid $100 in taxes and received $200 in Medicaid payments). Such a tax provides states with a way to effectively generate Medicaid dollars out of thin air, by imposing a “tax” on healthcare providers that ends up generating revenue for those same providers.

Congress never contemplated this strategy when it enacted Medicaid in 1965, but by the mid-1980s, as states faced mounting fiscal pressure due to a weak economy and growing Medicaid rolls, arrangements like the one described became increasingly common.[2] A state would impose a tax on, for example, nursing homes with high Medicaid utilization, spend all the revenue raised on Medicaid nursing home rates, and draw down a federal match on the spending — essentially funding Medicaid on the federal dime.

Of course, even a tax that generated “free” federal revenue in the aggregate might still cost money for some specific providers. A tax on all hospitals that was used to fund a Medicaid hospital rate increase would increase funding for the hospital industry overall, but hospitals with few Medicaid patients might pay more in taxes than they received from the rate increase. However, states had ways of structuring taxes to address this problem: for example, they might levy the tax only on providers with a high concentration of Medicaid patients (what we will call “high-Medicaid providers”), or make the tax proportional to Medicaid revenue. In some cases, states even guaranteed that every provider paying the tax would receive at least as much in rates as they paid in taxes, a so-called “hold harmless” provision.

The 1990s: Congress Steps In and New York Sues Bill Clinton

Unsurprisingly, these provider taxes drew criticism. Fiscal conservatives complained that states were manipulating the Medicaid system to draw down more federal money. Still, a total ban on the practice proved politically unpalatable: healthcare advocates naturally supported the taxes as an admittedly circuitous route to sustaining Medicaid, and many governors (of both parties) had come to rely on provider tax revenue to balance their budgets. Thus in 1991 Congress passed legislation which restricted the use of provider taxes but did not fully ban it.[3] The rules were complex, but in essence the law requires that provider taxes be[4]:

  • Broad-based: A provider tax must fall on an entire category of providers — it cannot, for example, single out only high-Medicaid providers.
  • Uniform: A tax must fall equally on all providers — it cannot be higher for high-Medicaid providers than for others, for instance.
  • Without “hold harmless” provisions: States cannot guarantee that every provider will receive at least as much as they pay.

These rules came with two major caveats, however. First, the rules would not apply to any tax raising less than 6 percent of provider revenues — these taxes would remain totally unrestricted.

Second, Congress allowed the Department of Health and Human Services to waive the “broad-based” and “uniform” requirements for specific state taxes as long as the tax was judged to be redistributive — shifting money from non-Medicaid providers to Medicaid providers.

In essence, the federal government will match provider tax funding as long as the tax creates “losers” as well as “winners”: a state can’t just tax the providers who treat a lot of Medicaid patients and will benefit from the associated Medicaid rate increase, it needs to also tax providers with few Medicaid patients who will not experience the increase. And the federal government has some discretion in determining what taxes meet this requirement. This complex situation has led to ongoing controversy over specific tax structures.

Controversy was especially acute in New York, which had adopted a variety of provider taxes in the 1980s. Once the 1991 law passed, it was not clear whether these provider taxes would meet the new requirements or whether the federal government would grant a waiver; the State was left in suspense for several years[5] as the Clinton administration finalized implementing regulations. Meanwhile, the State doubled down on its use of provider taxes by implementing a new set of taxes[6] on hospitals and health insurance as part of Governor Pataki’s 1996 legislation deregulating hospital prices.

New York Senator Al D’Amato attempted to resolve this uncertainty by inserting a provision into the federal budget which would grant New York a retroactive exemption from the 1991 law’s requirements, effectively “grandfathering in” all existing New York provider taxes. President Clinton issued a line-item veto[7] of this provision, leading the Supreme Court to declare line-item vetoes unconstitutional[8] in the landmark Clinton v. City of New York ruling in 1997.[9]

The California Loophole

The 1991 reforms to provider taxes ensured that they would be an ongoing source of state revenue — and state-federal controversy. Provider taxes of various kinds are a near-universal[10] feature of state Medicaid policy: 49 states and the District of Columbia have at least one provider tax, and these taxes generated 17 percent of state Medicaid funds as of 2018. Democratic administrations have tended to be more willing to approve state provider taxes than Republican administrations, and the use of provider taxes grew especially rapidly during the Obama administration. The Trump administration proposed[11] regulations which would have significantly restricted the use of provider taxes, but these regulations were never finalized.

Taxes on Managed Care Organizations (MCOs) have proven to be a particular site of conflict. Most states, including New York, administer their mainstream Medicaid programs through private insurance companies, or “managed care organizations.” States are allowed to levy provider taxes on MCOs and 18 states already do so.[12] To comply with federal regulations, however, these taxes must be broad-based and uniform — that is, they must apply not only to Medicaid MCOs but to private insurance plans like those purchased by small businesses for their employees.[13] States may seek a waiver of the broad-based and uniform requirements, but only if the tax is redistributive, moving money from the private insurance market to support Medicaid.

These requirements would appear to set narrow limits on states’ ability to levy a politically palatable MCO tax. In order to tax Medicaid MCOs (which will support a tax since it will pay for itself in newly-generated federal Medicaid revenue), states must also tax private MCOs (which will not see a benefit from the tax and therefore oppose it).

California in particular has long wrestled with these requirements. It imposed an MCO tax in 2010 that applied only to Medicaid MCOs, but CMS rejected[14] this tax in 2014. The state passed a restructured tax[15] in 2016, but this new tax was constrained by federal requirements and raised only a relatively modest $1.4 billion a year. Last year, however, California enacted legislation that greatly expanded[16] its MCO tax, raising nearly $5 billion a year in federal revenue. The tax levies a charge per person on individuals covered by both Medicaid and non-Medicaid insurance in California, but the tax is set at $182.50 per person for Medicaid enrollees and just $1.75 per person for private insurance enrollees[17] — a clear violation of the requirement that the tax be “uniform.” California was thus required to apply to the federal Center for Medicare and Medicaid Services (CMS) for a waiver of the uniformity requirement by proving that the tax is redistributive — moving funds from private insurance to Medicaid.

On its face the tax is evidently not redistributive; virtually all the money it raises comes from Medicaid and is spent on Medicaid (drawing down federal matching dollars). But California found a way to structure its tax so that it complied with the technical statistical test of redistributiveness used by CMS — and so, in December of last year, CMS was forced to very grudgingly approve it.

The New York State legislature appears to be taking inspiration from this approval: If California can do it, surely New York can too? This may well be the case, but it is important to recognize that CMS is extremely skeptical of the California plan. CMS’s formal approval letter[18] is worth quoting at length (emphasis added):

California’s tax on managed care organizations, due to its tendency to derive revenues from Medicaid, does not appear consistent with the definition of “generally redistributive”… California’s tax derives revenues mainly from Medicaid services (instead of non-Medicaid services) and uses these revenues as the state’s share of Medicaid payments. Accordingly, we are concerned that this tax program fails to be “generally redistributive in nature.” However, CMS is approving California’s request for a waiver of the broad-based and uniformity requirements because the state’s proposal meets the applicable statistical test

For the reasons described above, the result of the statistical tests, in these instances, do not appear consistent with either the definition of generally redistributive or reflective of the expected results based on the intended design of the statistical test. Therefore, CMS intends to develop and propose new regulatory requirements through the notice-and-comment rulemaking process to address this issue […] Please be advised that any future changes to the federal requirements concerning health care-related taxes may require the state of California to come into compliance by modifying its tax structure.

In essence, CMS is saying that California’s tax violates the spirit of the regulations — it is not truly redistributive — but the state has complied with the letter of the law, so CMS must approve it. But there’s an important caveat: CMS adds that it now plans to update the relevant regulations in order to close the loophole California has used — and the new regulations would likely require California to repeal or alter its MCO tax.

That news is not as bad as it may sound. California’s proposal was approved through 2027, and the state will get to keep the money it generates even if regulations change in the future. Meanwhile, new regulations can take several years to develop and implement and are subject to the vagaries of both politics and litigation. It is plausible that, if Biden is re-elected, his administration will decide not to tighten provider tax requirements after all. Meanwhile if Trump is elected, his administration will likely once again pursue a much more aggressive tightening of provider tax regulation — which will restart the regulatory clock and provoke widespread opposition. In the long term, it is quite possible that California will find its provider tax ruled noncompliant and will need to plug a $5 billion annual hole in its budget — but that will not happen until at least 2027, and possibly later.

California and the New York Model

Which brings us back to New York. It is likely that in proposing an MCO tax with “higher rates imposed on Medicaid Managed Care plans compared to non-Medicaid plans,” the legislature is looking to California for inspiration: California figured out how to generate more federal revenue for its Medicaid program, so why shouldn’t New York do the same?

Given the lack of detail in the legislature’s proposal, it is difficult to evaluate whether this model would work in New York and how much revenue it might generate if it did. A specific concern in New York is how this new tax might interact with our Health Care Reform Act (HCRA) taxes, which already raise substantial revenue for Medicaid and which are “grandfathered in” to federal compliance through the D’Amato provision. Would CMS allow New York to leave HCRA in place and impose this new tax alongside it, or would imposing an MCO tax require a broader conversation about HCRA reform? The issue is complex in part because, unlike California’s model, HCRA taxes really are redistributive — they impose a higher rate on commercial insurance and use this to fund Medicaid services. And like California, New York might risk setting itself up for budget problems in the future if CMS tightens regulations on the MCO tax.

Still, the legislature’s proposal is clearly preferable to the drastic cuts to home care and lack of support for safety net hospitals in the governor’s budget. These cuts would create a crisis in access to care for hundreds of thousands of New Yorkers. The best way to avoid these cuts in the long term would be through programmatic reforms like the elimination of the wasteful Managed Long-Term Care program and a more systematic approach to hospital funding reform, but these changes will likely take more than a single budget cycle. The MLTC reform in particular could save billions of dollars per year. An MCO tax to draw down more federal revenue is a smart way to bridge the gap in the meantime.

[1] Assembly One-House Summary, page 102,

[2] Congressional Research Service, “Medicaid Provider Taxes,” Aug. 5, 2016,

[3] “Medicaid Voluntary Contribution and Provider-Specific Tax Amendments of 1991 (Public Law 102-234).” Health care financing review vol. 13,3 (1992): 131-5.

[4] The exact rules are quite complex; a good (although somewhat dated) summary can be found in Congressional Research Service, “Medicaid Provider Taxes,” updated August 5, 2016.

[5] Stevenson, Richard. “How New York Walked Into a Veto of its Medicaid Item.” New York Times, August 16, 1997.

[6] Citizens Budget Commission, “Six Things to Know About New York State Health Care Reform (HCRA) Taxes.”

[7] CNN, “A Historic Veto.” August 11, 1997.


[9] New York’s HCRA taxes on insurance and hospital services continue to be exempted from federal requirements under the “D’Amato provision” to this day, raising about $7 billion in revenue, most of which goes to fund the state’s Medicaid program. Occasional proposals to reform the taxes have noted that any reform effort would cause them to lose their “grandfathered” status and would require federal approval. However, HCRA raises somewhat different issues from most provider taxes and is not directly relevant to the legislature’s current proposal.

[10] Medicaid and CHIP Payment and Access Commission, “Healthcare-Related Taxes in Medicaid,” January 2020.

[11] Center for Medicare and Medicaid Services press release, “Trump Administration Proposes Historic Steps to Strengthen Oversight and Fiscal Integrity of the Medicaid Program,” November 12, 2019.

[12]Kaiser Family Foundation, “States with an MCO Provider Tax in Place.”,%22sort%22:%22desc%22%7D

[13] Taxes need not apply to self-funded employer insurance plans, like those used by the vast majority of large employers.

[14] Heath, Sarah. “California Managed Care Organization Tax Proposal Axed.” Revenue Cycle Intelligence, undated.

[15] Gorn, David. “State Legislature Passes New MCO Tax, Rescues Over $1 Billion for Medi-Cal.” California Healthline, March 1, 2016.

[16] Bluth, Rachel. “California Strikes Huge Deal Unlocking Billions for Health Care.” Politico, June 24, 2023.


[18] Approval letter available at