for Expansion in the Time of Coronavirus Jonathan Gruber, Ph.D., and Benjamin D. Sommers, M.D., Ph.D.
M edicaid — and how to pay for it — has deal for states considerably in its Paying for Medicaid
become a recurring theme in several cur- Medicaid expansion, covering
newly eligible adults with 100% rent critical policy debates. Fourteen U.S. federal funding from 2014 through states have not yet expanded the program under the 2016, and then scaling back to 95% in 2017, 94% in 2018, 93% Affordable Care Act (ACA), most- structure since its creation in in 2019, and 90% thereafter.2 ly because of concerns about the 1965 has been a joint state and So how has this system worked potential impacts on state bud- federal effort. The federal govern- since the creation of the ACA gets, and these decisions have left ment creates the ground rules for Medicaid expansion, and what more than 2 million low-income state participation in the program does the Covid-19 pandemic mean adults without any health care in exchange for large subsidies to in terms of the best way to struc- coverage.1,2 Meanwhile, earlier this the states. Before the ACA, states ture the program? year, the Trump administration received a “match rate” (formally In recent research, we analyzed invited states to submit proposals the federal medical assistance per- budget data from all 50 states to shift Medicaid to a block-grant centage, or FMAP), which varies by from 2010 through 2018 to assess or per-capita-allotment system states’ per capita income. Higher- the impact of the ACA Medicaid with a capped federal contribution. income states such as New York expansion. As expected, we found Most recently, the coronavirus epi- and California have a 50% match that expansion states experienced demic has hit like a thunderbolt, rate, while Mississippi — the poor- a substantial increase in Medicaid both federal and state policy- est state in the country — current- spending since implementation of makers are looking to Medicaid ly has an FMAP of nearly 78%. the expansion, with 24% higher as a central tool in their response The states are responsible for the growth than nonexpansion states to this national emergency. remainder of Medicaid program between 2013 and 2018. Criti- For all these areas, it has be- costs, which they finance through cally, when analyzing the source come increasingly important to sources including general state of funds, we found that this in- understand how Medicaid is paid revenues and taxes on health care crease in Medicaid spending was for. Medicaid’s basic financing providers. The ACA sweetened the subsidized entirely by increased
ly covered by many state Medicaid 50 programs.4 Thus, Medicaid ex- pansion appears to be a win–win 40 from the states’ perspective — giv- Relative Change per Year (%)
ing health insurance to millions of
30 low-income adults and offering 20 financial support to safety-net hos- pitals, without any adverse effects 10 on state budgets. Yet proposals for dramatic 0 changes to the way Medicaid is −10 paid for are in the works, with the Centers for Medicare and Medicaid −20 2010 2011 2012 2013 2014 2015 2016 2017 2018 Services (CMS) recently launching the “Healthy Adult Opportunity” B Spending from State Funds initiative that encourages states 50 to leave behind this system and shift to a fixed federal contribu- 40 tion. The basic economics of the Relative Change per Year (%)
30 Medicaid match-rate system clar-
ify the potential advantages and 20 drawbacks of such an overhaul. The large subsidy from the fed- 10 eral government leads states to 0 cover far more people under Med- icaid than if they were spending −10 only their own funds. Such broad- ened coverage was the intent of −20 2010 2011 2012 2013 2014 2015 2016 2017 2018 Congress in the 1960s and with the passage of the ACA in 2010. Changes in State Spending Associated with Medicaid Expansion, Using Federal Critics, however, contend that the Funds as Compared with State Funds, 2010–2018. subsidy leads states to be ineffi- The graphs show changes per year for states that have expanded Medicaid as of 2018, cient in running their programs, as compared with nonexpansion states, with 2013 as the reference year. Bars show 95% confidence intervals, using robust state-clustered standard errors. Models adjust producing high costs with what for state-year unemployment rates and per capita income. Adapted from Gruber and some have called “mediocre” out- Sommers.3 comes — despite a large body of evidence showing wide-ranging federal funding to expansion state spending, when they were health benefits of Medicaid expan- states, with no significant chang- on the hook for 5% of the expan- sion.5 Viewing the program as es in spending from state revenues sion costs in 2017 and 6% in overly expensive and ineffective, associated with Medicaid expan- 2018? Our findings in this regard the administration has pushed sion (see graphs).3 We also found are consistent with case studies states to accept caps on federal no evidence that Medicaid expan- from several states showing that support for Medicaid in exchange sion forced states to cut back on they have used federal dollars for flexibility to refashion the spending on other priorities, such from the Medicaid expansion to program in terms of who is eli- as education, transportation, or offset other areas of state spend- gible, what services are covered, public assistance, despite frequent ing, such as direct subsidies to and how care is delivered. assertions by opponents of expan- public hospitals and mental health The current Covid-19 crisis sion that the policy would inevita- centers, health care costs for peo- highlights a major flaw in this bly have such harmful effects. ple involved with the justice sys- proposal when it comes to unex- How could states have expand- tem, and a more generous match pected public health shocks. The ed Medicaid without increasing rate for optional groups previous- open-ended nature of the match-
