This is the second in a 3-part series explaining how the Republican replacement plan for Obamacare will address the 3 biggest flaws related to the Medicaid expansion:
- the enhanced federal matching rate (Part 1);
- the failure to alter Medicaid’s open-ended matching rate structure (Part 2); and
- the lack of integration with private health insurance, especially the new health care Exchanges created under the law (Part 3).
Obamacare’s second biggest flaw is that even as it upended the half-century long consensus over who should be in Medicaid, the law inexplicably left intact a feature of Medicaid that we have known for decades fuels excess spending: its open-ended matching rate.
So long as states put up a dollar to fund Medicaid, Uncle Sam is obliged to match it with anywhere from $1 to $3 federal dollars depending on that state’s unique matching rate. It has been proven empirically that this formula fuels higher spending. An analysis by Thad Kousser at UC Berkeley showed that all other things being equal, shifting a state from the lowest to highest federal matching rate increases discretionary Medicaid expenditures by 22 percent.
Many Medicaid providers might well think: “well what’s wrong with that? Don’t we want more Medicaid spending?” But if we look beyond the narrow interests of providers and instead look at this from a social or taxpayer perspective, this arrangement has created serious efficiency and equity problems.
Medicaid Financing Encourages Fiscal Irresponsibility
In part 1, I already explained how the traditional Medicaid matching formula has long encouraged wasteful spending and discouraged states from economizing by creating a one-way ratchet effect that fuels ever-rising Medicaid spending. There is no need to repeat that argument here. But the eye-popping consequence of such incentives is that in 2014 we spent 93 times as much on Medicaid as we did less than a half century earlier in 1970 (Table 1).
Equally troublesome are the perverse incentives this formula creates for states to take from one another and to transfer obligations to future generations. The formula encourages states to expand their programs as much as possible, knowing that at least half the cost will be exported to neighbors. This gold rush mentality likewise minimizes whatever concerns state policy makers might otherwise have about the burden being placed on future generations by profligate entitlement spending. The enhanced federal matching rate undoubtedly greatly amplified these incentives, but this was a baked-in feature of Medicaid financing from the get-go and helps explain why Medicaid has ended up benefiting wealthy states in a manner never intended by its designers (more on that below).
If we can justify an expenditure only when someone else pays for it, then we probably ought to reconsider incurring that obligation . We can do better than to simply rationalize “Well, sure it’s wrong, but everyone else is doing it.” Most of us had mothers who deftly countered the grade-school argument “Well Johnny did it” by asking: “Well if Johnny jumped off a bridge, would you?”
Medicaid financing essentially is a self-inflicted “tragedy of the commons” . The “tragedy of the commons” was an idea penned by Garrett Hardin (a professor of biology at UC Santa Barbara) just 3 years after the passage of Medicaid. Professor Hardin’s concern was pollution, but his idea applies with equal force to any instance in which private costs diverge from social costs.
Just as the “commons” encourages rational farmers to keep adding more sheep to the commons because they have failed to consider the adverse impact of their actions on other farmers, the FMAP encourages every state to increase Medicaid spending more than it would otherwise. Every state has to balance its spending on Medicaid against its spending on K-12 education, higher education, transportation, c0rrections, public assistance and other priorities. Such decisions are greatly distorted when states know they can receive $1 to $3 from Uncle Sam for every state dollar put into Medicaid. The enhanced match rate distorts things drastically greater by giving them 9 federal dollars for every state dollar used for the expansion.
The FMAP literally encourages Americans to pick each other’s pockets by making narrow decisions that will benefit their own state while exporting half or more of the cost of such decisions onto the backs of other taxpayers. But this fiendishly clever mousetrap also allows each state’s neighbors to pull the same stunt. Consequently, the insidious nature of the matching grant formula is that Americans spend more on Medicaid than they would if either the entire amount were financed by states or by the federal government. That is, we collectively overspend on Medicaid relative to what we would do if we more transparently and honestly balanced Medicaid spending against available tax resources in light of competing priorities.
Medicaid is Crowding Out Education and Other Critical State Priorities
For decades, Medicaid has been the Pac-Man of state budgets, slowly but steadily eating away a rising share of them. In the last year of the Clinton administration, Medicaid accounted for less than one-fifth of state spending. By 2014–the first year of Medicaid expansion–that share already exceeded one quarter.
The result has been a steady decline in the share of state budgets devoted to education, transportation infrastructure and public safety.
Much of this has resulted from an explosion in Medicaid enrollment over the past two decades or so.
- A recent Mercatus study showed that monthly Medicaid enrollment grew from 31.7 million in June 2000 to 55.0 million in June 2013, an increase of 73 percent, far outstripping overall US population growth of 12 percent.
- Political scientist Frank J. Thompson has estimated that enrollees per poor person grew in all states from 1992 to 2008, ranging from an 8% increase in VA to 126% in OK (NC grew by 62.5%) [Table 2.3].
- The Council of State Governments reports that as of 2010, 16.3% of the nation’s population was enrolled in Medicaid or CHIP, ranging from a low of 9.4% in Utah to a high of 24.4% in New York.
- Charles Gaba at ACAsignups estimates that as of last October, an additional 17.9 million had enrolled in Medicaid since March 2010; this implies that today, more than one in five Americans is on a program originally designed for poor people (13.5% of Americans have incomes below the federal poverty level.
Medicaid Is a Boondoggle for Wealthy States
Pre-Obamacare. Even though the matching formula is intended to provide more help to poor states compared to wealthier states, in actual fact, Medicaid has always been a huge boondoggle for wealthy states. The amount of federal Medicaid subsidies per poor person varies dramatically across states, yet it is the richest states, not the poorest, who end up benefiting from this. Note that the three highest income states—DC, Connecticut and Massachusetts—are among the 6 states receiving the largest amount of federal Medicaid dollars per poor person .
In contrast, Georgia falls in the bottom quintile of states in terms of per capita income, yet Washington DC gets nearly 5 times as much federal funding per poor person despite being the highest income state in the country.
Even before the Obamacare Medicaid expansion was implemented, 24.4% of New York’s population was enrolled in Medicaid or CHIP despite its having the nation’s 6th highest per capita income. This Medicaid-dependent share of the population is the nation’s highest and is nearly one fifth higher than in Mississippi—long known as the nation’s most destitute state. You might think that because of its high income, New York taxpayers surely end up financing their own lavish Medicaid program. In fact, however, New York collects $1.35 in federal Medicaid revenues for every $1.00 in federal Medicaid taxes paid by its own citizens . Does this make sense to anyone?
But the reason it happens is that because there is a floor on the federal match rate for Medicaid (called the federal medical assistance percentage or FMAP), even the wealthiest states get a 50% discount on their Medicaid spending courtesy of Uncle Sam. This is far higher than the FMAP such states would merit based on their per capita income (for example, but for the 50% floor, the regular FMAP for Connecticut would be only 10.81%! Thus, the state is getting 4.6 times as much federal assistance to cover Medicaid as would be warranted based on “need”). This gives these states strong incentives to expand coverage to many people who are not poor, while shifting half of the cost onto the backs of taxpayers in states with far less discretionary income.
Mr. Badger notes there are 13 “floor” states that are protected by the FMAP floor: Alaska (9), California (12), Connecticut (2), Maryland (7), Massachusetts (3), Minnesota (14), New Hampshire (10), New Jersey (5), New York (6), North Dakota (4), Virginia (11), Washington (13), and Wyoming (8) [I have put each state’s 2012 ranking on per capita income]. Technically, District of Columbia, which has the highest per capita income, would qualify for a 50% FMAP, but by law, it is instead set at 70%.
Fig. 3 codifies how much excess funding the top 10 wealthiest states receive as a consequence of the 50% FMAP (for purposes of exposition, this chart focuses exclusively on the regular FMAP for 2017 and leaves out the enhanced matching rate: including the latter would have made the picture even worse. Also, according to the strict matching formula, Washington D.C. actually should have a negative match rate 6.62%, but its actual regular FMAP is 70%; there’s no way to display this on the chart, so I arbitrarily set the excess amount to 500%).
This serious bias in Medicaid spending favoring wealthy states was well known to policy scholars prior to passage of Obamacare . Thomas Grannemann (Centers for Medicare and Medicaid Services) and Mark Pauly (Wharton School) in their pioneering book called Medicaid Everyone Can Count On, published just prior to the passage of the ACA. They argue that the matching grant structure had diverted federal dollars into optional coverage in high-income states, thereby compromising the adequacy of the coverage for low-income enrollees in low-income states. Yet the law not only did nothing about it, but actually made things much worse.
Post-Obamacare. Now the situation is even worse since many of these wealthy states had already previously covered non-disabled adults: 13 states had extended coverage to childless, nondisabled adults and 12 other states had extended coverage to adults with dependent children up to 133 percent of the federal poverty line . In all cases, the states received the standard FMAP under these waivers, but now they get a 90% match to cover the identical individuals, creating an enormous unexpected windfall.
An analysis by Urban Institute researchers showed that 10 states actually would make money on the Medicaid expansion, i.e., see a net reduction in the total amount of Medicaid spending financed with state dollars compared to what they would have spent under Obamacare had they not expanded. Not every one of these lucky states is wealthy, but it is telling that not a single state in the bottom quintile of states ranked by per capita income is a part of this group, whereas 4 of the 10 states in the top quintile are.
What Does AHCA Do About the Medicaid’s Open-Ended Federal Matching Rate?
Structure of Per Capita Federal Payment Caps. The AHCA continues completely eliminates the federal matching rate formula starting in 2020 by giving states options for a per capita cap or block grant. Using a 2016 base year (that will preclude states from gaming the new system by ratcheting up spending prior to 2020), this cap would vary by eligibility category and state, reflecting the enormous variation that now exists on both dimensions.
The per capita cap would be adjusted annually by the medical component of the Consumer Price Index (CPI) for all eligibility groups except the aged and disabled (for whom the annual adjustment would be the medical CPI plus 1 percent).
The block grant option is available to states only for the eligibility groups that are not disabled or aged, and because it would be adjusted annually by the regular CPI only, my guess is that most states would opt for the per capita cap, since this allows Medicaid spending to rise or fall based on the number of people in each eligibility category. In short, for most states, this will be the more generous and less risky option for federal Medicaid financing.
Fiscal Impact of Per Capita Caps. Because it uses a 2016 base year and because of the inherent incentives within Medicaid that fuel spending to grow faster than either the economy, population and even medical inflation, the per capita funding formula is intended to guarantee federal budget savings even in the very first year the caps become effective.
That said, Mr. Roy has pointed out that in the foreseeable future, the inflation adjustment contained in the AHCA essentially is twice as generous as the formula proposed by President Clinton in 1995 when he endorsed very similar spending caps on Medicaid. The Clinton planned would have allowed per-enrollee spending to grow only by the rate of real (inflation-adjusted) growth in GDP. Currently, real GDP growth is about 2%, compared to roughly 4% annual growth in medical inflation.
Estimates of the savings these caps might produce vary widely:
- According to Forbes Opinion Editor Avik Roy, the projected 10-year savings would range from $5 billion to $107 billion depending on whether one believes the estimate of the Medicare actuary or CBO regarding future medical inflation during this period).
- Vernon K. Smith estimates the aggregate cuts to federal Medicaid funding under the AHCA would be 15% in 2020, rising to 26% by the year 2026, but does not decompose these to show how much is related to the caps themselves as opposed to changes in enrollment etc.
- In contrast, Galen Institute Senior Fellow Doug Badger estimates that 82% of the projected savings would be the result of lower enrollment in Medicaid, 18% ($137 billion) would be the result of eliminating the enhanced matching rate for nondisabled adults and only 0.4% ($3.4 billion) would be the result of the caps themselves.
So the good news is that shifting to per capita caps would save money by controlling Medicaid spending at a more sensible rate than the current open-ended matching rate does. The downside is that the new system largely will preserve rather sizable inequities across the states.
Residual Defects in the New Medicaid Financing System Under AHCA
As the foregoing suggests, the AHCA is not perfect. To better understand its remaining defects, it is worth pondering what an alternative system might look like.
What if Medicaid Were Entirely Federally-Financed? The matching formula appeals to federal policymakers since it allows them to address a serious problem without bankrolling this entirely out of the U.S. Treasury (raising taxes is painful for politicians at every level of government, as well it should be in a nation that purports to be “of the people, by the people and for the people”). But the result is that federal Medicaid spending no longer is under the control of federal policymakers but instead is driven to a considerable extent by what states decide to do.
One caveat to the supposition that spending would be lower if this were entirely decided by Congress is that for many decades (going back to the 1930’s), federal policymakers have been content to let federal spending outstrip tax collections in the vast majority of years. During the Obama administration, the federal budget deficit exceeded one quarter of all federal spending . This means that spending outstripped tax collections by one-third. Unless we think that Medicaid spending is a lower priority than other components of the federal budget (the opposite appears to be true), then we can reason that had federal policymakers been fully in charge during the Obama presidency, Medicaid spending would have been at least one third higher than it otherwise would have been had state policymakers been given unilateral authority over spending levels.
Note that unless such deficits are made up in future years, we effectively are passing along a sizable fiscal burden onto future generations. The chances of this generation paying off its own federal debts (much less those of past generations) are vanishingly small: over the past 75 years ending in 2016, budget deficits have averaged 15.4% of spending, yet during that same period, we have run federal budget surpluses exactly 12 times and only once did the resultant surplus exceed 15% of spending .
In short, the solution to the Medicaid “commons” problem cannot possibly be to hand the problem to Uncle Sam, as Washington has a demonstrated track record of handling the nation’s fiscal affairs completely irresponsibly for the lion’s share of the past 75 years.
What if Medicaid Were Entirely State-Financed? In contrast to Uncle Sam, most states face strict balanced budget requirements. For the sake of argument I will assume that state policymakers have maximized whatever revenue it can raised based on the ability and willingness of its taxpayers to be taxed.
If so, then if Medicaid were financed entirely by states, policymakers would be forced to think hard about whether it is worth giving up one dollar on K-12 education in order to find another dollar to fund Medicaid. Whatever trade-offs they made about these difficult decisions presumably would reflect the preferences of voters in that state. We have already seen in part 1 that there is a huge diversity across states in their willingness to pay for health care of the medically indigent. Consequently, the decentralized collective decisions of 50 states and the District of Columbia are far more likely to result in the socially optimal level of Medicaid spending than if this were decided in Washington, D.C. .
Of equal importance, because of balanced budget requirements, state policymakers at least do not have the same perverse incentive to immorally pass along fiscal burdens to future generations. In a perfect world, we would leave Medicaid spending to individual states, with the federal role limited to providing assistance to whatever number of states are deemed to be unable to adequately finance the medical needs of their poorest citizens.In a perfect world, we would leave Medicaid spending to individual states, with the federal role limited to providing assistance to whatever number of states are deemed to be unable to adequately finance the medical needs of their poorest citizens.
Joint Responsibility for Financing Medicaid in a Federal System. In a perfect world, we would leave Medicaid spending to individual states, with the federal role limited to providing assistance to whatever number of states are deemed to be unable to adequately finance the medical needs of their poorest citizens. From the standpoint of efficiency and equity, it actually doesn’t make sense for Uncle Sam to be bankrolling Medicaid benefits in the nation’s wealthiest states such as Connecticut, Massachusetts and New Jersey. Such states have ample resources to provide more than adequate Medicaid programs for whatever number of their citizens they elect to assist.
The next most perfect arrangement would have Uncle Sam ensure a floor on Medicaid spending by giving states an equal per capita amount per eligibility category, adjusted perhaps for legitimate geographic variations in health spending. That would ensure a “decent minimum” across all states regardless of their ability to pay. States then would be able to top up these amounts by whatever amount was politically palatable. But such decisions would not be distorted by the temptation to rely on citizens in other states to help bankroll whatever level of generosity each state wished to accord its most impoverished citizens. As well, states would have every incentive to spend Medicaid resources wisely knowing that each dollar of waste or excess would be borne by its own citizens.
The Reality of AHCA Medicaid Financing. AHCA’s per capita caps deviate from what might be ideal in two important ways. First, federal funds are provided to all states rather than just those in financial need. Second, the caps freeze in place current disparities in federal funding that have arisen over decades and go far beyond legitimate geographic variations in health spending.
Such legitimate variations include cost of living differences (rents, energy costs, wage differences among office personnel etc.) that are beyond the control of the health system. Other variations that arise from differences in utilization patterns or payment rates to providers really ought to be the responsibility of individual states. That is, it’s perfectly fine if the citizens of New York decide their doctors should get paid more or that relatively more births should be via C-section, but the ideal financing system would not artificially distort such decisions by offloading their costs on residents outside of New York’s borders.
On equity grounds alone, no one can really justify when NY, for example, deserve capped payments that are 40% higher than Mississippi’s (this despite the fact that New York’s match rate is only 50% and Mississippi’s is 74%). Ideally, we don’t want a formula that permanently rewards states such as New York for their past aggressiveness in pursuing Medicaid dollars. But the AHCA in its current form does not rectify this problem. The best that can be said is that a capped funding formula provides a ready mechanism to slowly but steadily compress the amount of federal funding variation across states that is unrelated to their ability to pay or legitimate geographic differences in medical spending.
In short, while far from perfect, when compared to where we have been over the past 50 years, the AHCA is a sizable step in the right direction of providing a Medicaid financing structure that is efficient, equitable and sustainable over the long term .
Update #1: June 17, 2017
I corrected the title to the chart 10 Lucky States That Literally Made Money By Expanding Medicaid. I also added a chart 10 Wealthy States That Benefit From FMAP Floor of 50% so that readers could better see my point about how wealthy states are benefiting from how Medicaid is structured.
READ CHRIS’ BOOK, The American Health Economy Illustrated (AEI Press, 2012), available at Amazon and other major retailers. With generous support from the National Research Initiative at the American Enterprise Institute, an online version complete with downloadable Powerpoint slides and companion spreadsheets has been made available through the Medical Industry Institute’s Open Education Hub at the University of Minnesota.