As 2017 ended, Medicaid's entitlement structure—and the funding base on which this structure rests—remained intact. During the Trump administration's first year, this largest and most unique form of US health insurance faced an existential threat under three distinct waves of legislative attack as part of the efforts to repeal the Affordable Care Act (ACA). The first two waves would have preserved the ACA's transformative change by maintaining the eligibility category created by health reform for working-age low-income adults. But the price exacted for this concession was enormous: loss of the very federal funding on which the ability to provide coverage rests. The third wave would have gone further, not only destroying Medicaid's open-ended funding structure but also stripping states of the option to extend federally-financed assistance to all poor working-age adults. The effort to effectively end Medicaid as we know it failed for both political and policy reasons, but the existential threat continues in Congress, the executive branch, and the courts.
This essay began its life early in the Trump administration, when despite a seeming alignment of the stars—a Republican president, a Republican Congress, thirty-four Republican governors, and complete Republican control of twenty-six state governments—Medicaid at least initially survived perhaps its gravest existential threat in fifty-two years. By the time 2017 ended, Medicaid had survived the onslaught of three distinct and powerful waves of attack. With each wave, Medicaid's enduring importance grew more visible to the public, as did evidence of what ending Medicaid would mean in the real world.
Whether Medicaid is able to prevail in the battles yet to come is an open question, and its future remains uncertain. By the time the sun set on 2017, Congress had enacted tax legislation that is projected to increase the budget deficit by some $1.5 trillion over the next decade, thereby setting the stage—if not in 2018 then down the road—for Medicaid's radical transformation as part of a far-reaching effort long dreamed about by opponents of public entitlement spending on the poor (Cassidy 2017). A further avenue of attack also lies with the Trump administration itself, which has made its intention clear to use special demonstration authority conferred under section 1115 of the Social Security Act to cull millions from the Medicaid rolls (Rosenbaum 2017c). Finally, an increasingly conservative judiciary could fundamentally destabilize Medicaid by overturning decades of decisions that have recognized Medicaid as conferring legally enforceable rights on the part of the children and adults on whose behalf the program is designed to function. Were Medicaid to cease to function as what the courts term a “rights-conferring” benefit, this seminal shift in judicial doctrine in and of itself would open the door to the arbitrary cessation of coverage by the government, with no meaningful legal recourse (Rosenbaum 2017a).
Both size and philosophy lie at the heart of the perpetual cloud under which Medicaid functions. In 1966 the program enrolled 4 million people and spent a total of less than $1 billion. Federal projections show that by 2025, total federal and state Medicaid spending will approach $1 trillion, and enrollment will surpass 80 million people (US Department of Health and Human Services 2016). The Congressional Budget Office (CBO) estimates that between 2018 and 2027 alone, the federal government will spend $4 trillion solely on beneficiaries under age sixty-five (CBO 2017a).
Medicaid's size is a function of whom the program serves and the services it covers; on a per capita cost basis, the program is a veritable bargain compared to the cost of buying comparable services with private insurance (Rosenbaum et al. 2016). This basic fact drastically limits the utility of cost containment strategies aimed at reducing the price that Medicaid pays for covered services, although important opportunities do appear to exist in the area of pharmaceutical pricing (MACPAC 2018). The ability to more aggressively negotiate the cost of drugs is important, as are states’ continuing efforts to achieve greater efficiencies in how health care is organized and delivered, particularly to beneficiaries who use long-term services and supports. But while these strategies are important, they will only slow program growth. Short of disinsuring millions through arbitrary spending caps or moving millions to alternative sources of coverage—which are bound to offer less coverage and to be more costly on a per capita basis than Medicaid—there are no easy fixes.
Size is not the only problem. By far the largest of all means-tested entitlements, Medicaid is a constant mark for those driven by powerful ideological opposition to government entitlement spending on the poor and medically vulnerable. Moreover, to those who have worked on Medicaid for decades, it feels as though the script never changes. The welfare tropes that for centuries—dating back to the English Poor Laws—have fueled attacks on the poor and the programs that serve them were on full display during 2017 and are driving current events as well (Rosenbaum 2018). Rather than investing time in developing policies for interventions that might make Medicaid stronger and more efficient, those who revere the program (including this author) find themselves stuck in a defensive crouch, unable to do more than fight to maintain Medicaid's basic existence.
This article begins by placing the debate over Medicaid's future within the broader context of its far-reaching role in the American health care system. It then turns to the 2017 legislation, which unfolded in three waves over the March–September time period. In the first wave, the American Health Care Act (AHCA; H.R. 1628) passed the House of Representatives on May 4, after initially failing on March 24. It then failed in the Senate, when that body's version of the bill, the Better Care Reconciliation Act of 2017: An Amendment in the Nature of a Substitute (BCRA), collapsed in spectacular fashion early in the morning of July 28 following Senator John McCain's epic televised thumbs-down vote in the immediate wake of his diagnosis of terminal cancer (Hulse 2017). The third wave—a last-minute legislative blitzkrieg mounted by Senators Lindsay Graham and Bill Cassidy—ended on September 24 when the Senate majority leader announced that votes were lacking (Haberkorn, Everett, and Kim 2017).
Following a chronicling of these distinct periods, the article turns to a discussion of what might lie ahead in Congress, the administration, and the courts.
The Indispensable Program
In a nation that prides itself on its reliance on a modestly regulated commercial health insurance market to finance health care for working-age Americans and their families, Medicaid emerges as the social welfare rock on which such reliance necessarily rests. Insurers have long known how crucial Medicaid is to their survival and their ability to avoid the types of financial risks that inevitably would bring a commercial insurance model to its knees; indeed, over many decades, the industry has been an advocate for Medicaid, including the Affordable Care Act (ACA) provisions extending Medicaid to all low-income working-age adults (America's Health Insurance Plans, n.d.).
In keeping with a long line of judicial decisions dating back decades, Medicaid, whose specific legislative provisions are actually silent on the matter, is recognized as a program that establishes a legal right to a defined set of benefits for those who are eligible (Rosenbaum et al. 2012: chap. 11). Under the terms of its entitlement, Medicaid is unique in whom and what it covers. Even as reconceptualized by the ACA, private health insurance—in particular, the individual market—exists under what only can be thought of as hothouse conditions. The individual insurance market necessarily rests on the existence of a stable risk pool containing enough healthy individuals to make the pool viable. To avoid adverse selection, enrollment must be limited to designated, brief open-enrollment periods and a handful of special enrollment periods unrelated to the need for health care.1 Benefits, even if relatively generous, must be carefully circumscribed, limited to what is offered under a “typical employer plan” (ACA, section 1302). This restricts coverage to preventive and acute-care needs and excludes the long-term services and supports essential to children and adults with significant disabilities. To avert perceived overutilization, insurers must impose significant cost sharing, a technique increasingly used in both the employer and individual markets, which employ high point-of-service out-of-pocket costs as the principal means for deterring utilization of ostensibly covered benefits (Kaiser Family Foundation 2017a).
By contrast, Medicaid is a hardy perennial, a program built to embrace risk. Most fundamental, perhaps, people who are eligible can enroll at any time; indeed, those who do enroll are entitled to retroactive coverage that can begin up to three months prior to the date of application date, to protect against catastrophic bills already incurred. There are no restrictions on access to Medicaid, precisely because Medicaid is designed to be available when health care is needed. It is Medicaid's fundamental accessibility feature for those who are eligible—as much as whom and what it covers—that gives it true safety-net statute.
Medicaid is able to operate this way because it rests on a general federal revenue base rather than on specifically designated federal revenue sources (such as Medicare Part A's payroll tax) or individual payment of actuarially sound premiums. Assuming that the states and the federal government are willing to spend the money, the financing needed to administer Medicaid can grow without regard to the direct impact of revenue growth on either special taxing sources or the size of the monthly enrollment fees paid by individuals.
Medicaid's potential for funding elasticity gives the program its uncommon strength. This potential also has led to repeated political crises; decades of reliance on Medicaid as the means for addressing generations of public health imperatives have led to a program of vast size. The program has been the subject of repeated efforts to curb program costs through strategies ranging from eligibility and benefit rollbacks and tools such as managed care, whose purpose is to improve operational efficiencies, frequently by tightening access to care (Smith and Moore 2015). Despite its size, Medicaid remains the low-cost alternative to private insurance coverage, with annual per capita spending 25 percent below the cost of comparable private insurance, chiefly because provider payment rates are lower (Sparer 2017).
In sum, Medicaid's size is a reflection of the load it is expected to bear. Medicaid has been the solution of choice for a vast array of public health challenges; simply put, no health care program has been assigned more tasks. The program provides a pathway to affordable health insurance for over 50 million of the nation's poorest working-age adults and children who—by design—were excluded from the ACA's subsidized private health insurance market, which offers premium subsidies and cost-sharing assistance only once people's income reaches the poverty threshold (ACA, section 1401). Medicaid represents the nation's long-term care strategy for children and adults with physical and mental disabilities. Medicaid funds nearly half of all US births (Markus et al. 2013), many involving women enrolled at the point of pregnancy, an explicit Medicaid eligibility category. Diagnosed breast and cervical cancer is an explicit basis of Medicaid eligibility (Paradise 2017), as is disability that prevents work, enabling Medicaid to finance health care for one in four people with severe mental illness (Zur, Musumeci, and Garfield 2017). To overcome barriers to health care, Medicaid can supplement both Medicare and employer-sponsored health plans, providing people with extensive health care needs with additional protection against the cost of long-term services and supports, as well as mitigating cost-sharing obligations that strain modest family budgets.
Medicaid is the nation's largest health care first responder, a source of health care financing that can deploy additional resources during times of public health crisis, both naturally occurring and man-made. The importance of this dimension of Medicaid can be seen in the catastrophic 2017 hurricane season that prompted both Congress and the Trump administration to turn to Medicaid for an infusion of funding to help weather massive damage to community health services (CMS 2017). In the most recent example of Medicaid's importance to public health, legislation to extend funding for the Children's Health Insurance Program (CHIP) pending in the House as of October 2017 would provide $1 billion in additional Medicaid assistance to Puerto Rico (Healthy Kids Act, H.R. 3921, 115 Cong. 1st sess.).
Medicaid's unique qualities don't stop at eligibility. Its coverage design, expressed through its definition of “medical assistance,” is immense, eclipsing the concept of what should be covered that is found in other forms of public and private health insurance. Its cost-sharing principles keep out-of-pocket costs nominal in order to allay the effects of health care impoverishment. In the case of children, Medicaid's early and periodic screening diagnosis and treatment (EPSDT) benefit is uniquely comprehensive, employing a concept of medical necessity that focuses on the value of health care to children's growth and development, not the conventional and restrictive “illness and injury” norms found in commercial insurance policies, that in turn can exclude health care for children with developmental disabilities as covered treatment (Rosenbaum and Wise, 2007). Unlike private insurance, Medicaid does not exclude payment when coverage is furnished in other than clinical care settings, enabling Medicaid to help support comprehensive health and social service interventions in schools, adult day health programs, Head Start centers, women's shelters, and other nontraditional locations. Indeed, during the 2017 debate, the impact of major Medicaid reductions on special education programs emerged as a matter of specific focus. (Green 2017)
In sum, despite its massive size—covering nearly 75 million people at a total spending of $575 billion in fiscal year 2016 (Kaiser Family Foundation 2017b)—Medicaid has emerged as the most nimble response to population health needs, and its growth has unfurled over five decades. With each change—whether for poor children and pregnant women, children and adults with severe disabilities in need of long-term services and supports, or working-age adults too poor to qualify for the private insurance market—Medicaid has cemented its place in the health care system. Indeed, Medicaid's importance is a fact so well recognized that its existence as a building block of US health care policy has been recognized even by the US Supreme Court. In National Federation of Independent Business v. Sebelius (567 U.S. 519 ), which effectively rendered the ACA's adult Medicaid eligibility expansion optional, Chief Justice John Roberts famously characterized the expansion as “a shift in kind, not degree.” Congress, he wrote, had acted unconstitutionally in repurposing Medicaid from welfare into one of the pillars on which a system of universal coverage now rests, without the voluntary consent of states: “Medicaid is transformed; [i]t is no longer a program to care for the neediest among us, but rather an element of a comprehensive national plan to provide universal health insurance coverage.” Although the chief justice's characterization was correct, those who know Medicaid also know that Medicaid had become a basic feature of the American health care system long before its most recent eligibility reconfiguration. Indeed, this reconfiguration came only after decades of experimentation by states, all aimed at making Medicaid work as insurance for the poor (Rosenbaum 2017c).
Because of its size and scope, Medicaid has faced existential crises before. Because the program consumes so much federal and state spending, the problem of cost is an eternal one. Within two years of its 1965 enactment, Congress already had moved to blunt program growth through the introduction of strict financial eligibility limits for those who qualified for medical assistance as a result of high medical bills (Stevens and Stevens, 1975). Medicaid would be further transformed over the decades, particularly in terms of the tools Congress has given states to manage program costs through eligibility and benefit targeting, reductions in provider payment rates, increased use of cost sharing, and requiring enrollment in tightly controlled managed care arrangements as a condition of coverage for most populations (Smith and Moore 2015). But the program continued to grow for all of the reasons that make its existence an indispensable fact of life.
The changes introduced into Medicaid essentially curbed the rate of program growth but did not reverse it. And it is fair to say that efforts to control Medicaid's size have been counterbalanced by five decades of eligibility and coverage expansions—including the transformational ACA reforms. Each of these expansion milestones represented a practical and political response to the question of how to finance health care for the poor and those impoverished by high medical costs. It would take nearly twenty years for Congress to begin to align Medicaid eligibility with federal poverty standards, first for pregnant women and children, later for parents of low-income children, and ultimately for all adults. Along the way, revolutionary restructuring gains were made for children and adults in need of long-term care and other high-need high-cost populations, some with explicit diagnoses such as diagnosed breast and cervical cancer (Smith and Moore 2015).
Interspersed with these gains, however, have come profound threats as part of periodic congressional efforts to curtail means-tested entitlement programs. Cash welfare assistance for deeply impoverished parents and children suffered a mortal blow in 1996 with the repeal of the Aid to Families with the Dependent Children program and its replacement with the Temporary Aid to Needy Families block grant. But Medicaid survived, given the far more potent political health care base on which it rested (Smith and Moore 2015). To be sure, in 1981, and again in 1995, Medicaid came close to being block granted (Lambrew 2005). But even when Medicaid was a fraction of its current size, ultimately it was saved by the same political and practical considerations at work in 2017 (Lambrew 2017).
The First Wave of Attack in 2017: The American Health Care Act
With the election of Donald Trump and Republican assumption of power in both houses of Congress, a broad assault on Medicaid was inevitable. As chair of the House Budget Committee, House Speaker Paul Ryan had included Medicaid block grants in earlier budget blueprints (Park, 2013). Furthermore, in its dry run on “repeal and replace,” Congress already had passed a rollback of the ACA Medicaid expansion along with a provision that would bar federal Medicaid funding for Planned Parenthood. The bill was vetoed by President Obama (DeBonis 2016), but the repeal measure, coupled with the earlier block grant, provided the general contours of the Republicans’ Medicaid strategy once repeal and replace was in full swing.
The Republican Medicaid legislative strategy was bolstered by a wave of public accusations that began shortly after the Court's Sebelius decision, aimed at building the arguments against voluntary state decisions to adopt the expansion. Opponents of the law generally, and Medicaid specifically, characterized Medicaid as causing “harm” and sought to establish Medicaid in the public's mind as a shameful descent into welfare for “able-bodied” adults rather than as a form of health insurance. The fact that Medicaid represented the least costly means of insuring the poor was never acknowledged. Governors in red states voiced these views (Sommers and Epstein 2013). A virtual deluge of blogs and articles hostile to Medicaid framed the program in denigrating terms. A May 2017 Google search by the author found that searches of Medicaid as “broken” produced 5,190,000 results; as “worse than having no insurance,” 6,290,000 results; and as “ineffective,” “riddled with fraud,” or “a giveaway to the able-bodied,” hundreds of thousands of additional results. Fifty Vetoes, a tract produced by Michael Cannon of the Cato Institute designed to disrupt state ACA implementation, exhorted states to reject the Medicaid expansion and to refuse to set up state-based health insurance marketplaces (Cannon 2013).2
This drumbeat of degradation was reflected in House Speaker Paul Ryan's tract “A Better Way,” published in February 2017, which devoted outsized space to the argument for dismantling Medicaid. The Speaker's arguments were framed as “empowering states and increasing flexibility,” and he characterized the program as one that generated “inconsistent quality and high fraud” that would “hurt patient safety.” In other words, he essentially asserted, Congress would improve the public's health and improve fiscal integrity by repealing the expansion and block granting the traditional program.
Others would seek to bolster his claims of fraud and ineffectiveness by misciting evidence or exaggerating its importance. One constant claim was that Medicaid caused the opioid epidemic. Although the ACA expansion coincided with the rise in opioid addiction rates, there is no evidence to suggest that the ACA expansion caused the crisis to grow (Goodman-Bacon and Sandoe 2017). Indeed, one would expect addiction to be more severe among Medicaid beneficiaries given their higher rates of poverty and Medicaid's positive impact on access to prescription drugs (Baicker et al. 2017). Another claim was that by eliminating use of an asset test, the ACA enabled lottery winners to hold onto their winnings and still retain Medicaid.3
Neither the Speaker nor other opponents offered an explanation of how states could continue to insure the poor given the exponential funding reductions he proposed. In their hands, the ACA's funding enhancements to assure that states had adequate resources to expand coverage became a mechanism for diverting resources away from traditional beneficiary groups, particularly people in need of long-term services and supports.4 The fact that some states had expanded Medicaid while others had not became an argument that the expansion was diverting taxpayer resources to certain privileged states, even though all states remained free to adopt the expansion and claim their funding at any time. Viewed in this light, the introduction of arbitrary federal spending limits became a means for restoring federal spending integrity while unleashing state flexibility and innovation.
This effort on the part of the House Speaker to frame a Medicaid block grant and repeal of expansion funding as in the public interest was effectively echoed in a March 2017 letter to the nation's governors, sent by US Department of Health and Human Services (HHS) Secretary Tom Price and Centers for Medicare and Medicaid Services (CMS) Administrator Seema Verma (Price and Verma 2017). In one of the more remarkable statements from public officials charged with enforcing the law, Price and Verma asserted that the Medicaid expansion was “a clear departure from the core, historical mission of the program.” In other words, they argued, an Act of Congress was contrary to public policy. Based on this argument, Price and Verma exhorted states to propose federal demonstrations—in expansions and nonexpansion states alike—that would bar coverage to those unable to fulfill welfare work requirements and inject private insurance principles into Medicaid, such as barring enrollment when health care is needed, to prepare beneficiaries for the private sector. In effect, administration officials sought to move Medicaid off its basic public health foundation and to fundamentally reshape its core purpose.5
With the stage set, the House of Representatives brought forth AHCA (H.R. 1628) in the winter; the measure ultimately passed on May 4 in slightly modified form, with no votes to spare. Because of the special fast-track legislative rules under which repeal-replace would happen—a process that requires only a simple Senate majority rather than a sixty-vote plurality required for legislation considered under regular order—the bill could not make changes in the statute that did not have a direct impact on federal outlays. As a result, the measure did not contain the comprehensive statutory revisions required to explicitly eliminate provisions of law on which rested long-standing judicial interpretations of the statute as creating a legal entitlement to assistance among those who are eligible. To be sure, the measure did include some legislative restructuring, including the elimination of retroactive eligibility (a feature of the program dating to 1965), a one-year bar against federal funding for covered services furnished by Planned Parenthood clinics, a state option to impose work requirements paralleling those used in federal welfare law, eligibility restrictions for beneficiaries who win the lottery, and a state option to block grant coverage of certain populations.6 But ironically, even as the bill withdrew nearly $1 trillion in federal funding, it maintained Medicaid's essential legal structure.
Crucially, in a concession to Republican expansion states, the AHCA preserved the ACA adult expansion population as an optional Medicaid eligibility category, thereby eliminating once and for all any notion that Medicaid should not serve as the central means for insuring the poorest working-age adults. At the same time, however, and along with slashing federal insurance subsidies for low- and moderate-income people (CBO 2017b), the House bill dramatically reduced federal Medicaid funding. Thus, it was this funding reduction, rather than elimination of the eligibility category itself, that according to the CBO drove the steep decline in beneficiaries and the overall number of people with health insurance (CBO 2017c). According to CBO estimates, by 2026, 24 million fewer people would be insured, and Medicaid reductions would account for 14 million of those no longer covered. Furthermore, Medicaid spending declines alone would have accounted for nearly $900 billion of the measure's more than $1 trillion in federal outlay reductions.
In terms of how such a steep spending reduction would be achieved, the House bill would have hit its mark in two ways. First, the bill reduced federal funding for the expansion population from the scheduled 90 percent level (beginning in 2020) to the normal federal contribution rate of between 50 percent and 75 percent. Second, the measure placed a spending cap (termed a per capita block grant) on the overall federal contribution to state programs—not only for the expansion population but also for tens of millions of traditional beneficiaries, including children, parents, people with disabilities, and the elderly. Under the bill's complex cap formula, states would receive a set amount of federal funding on a per-person basis, with an aggregate cap tied to each state's total Medicaid-enrolled population. Essentially, the bill created a crude rate-setting system, using dated expenditure history to set the base year and trending spending forward at a rate below actual Medicaid growth, which already, on a per capita basis, was low in relation to other insurers. The cap would hold Medicaid growth to the medical component of the Consumer Price Index (CPI) for low-income populations, with a slightly higher permissible growth rate (medical CPI [CPI-M] plus 1) for elderly and disabled populations permitted under the version of the AHCA that passed the House in May.
The system for calculating what each state would be paid was crude. In terms of recognizing which populations drive program spending, the bill specified only a handful of rate cells. Furthermore, it lacked any means for adjusting for changes over the years in demographic patient characteristics, volume and intensity, or technology innovation. While the measure would have exempted certain limited expenditure categories (e.g., payments to certain narrow types of hospitals) from the per capita block grant, the vast majority of institutional spending would be under the cap, as would spending on medical care and prescription drugs, the major sources of program cost. The HHS secretary would have the power to award additional funds during periods of declared emergency, but the secretary would be entirely in control over if and to what extent the federal government would be willing to increase program spending in response to emergency conditions. Furthermore, the measure gave the secretary immense powers to withhold or adjust federal payments as he or she determined appropriate, with virtually no ability on states’ part to challenge downward adjustments in advance of their imposition.
In the near term, it was the withdrawal of the ACA adult expansion funding that caused federal Medicaid outlays to fall steeply over the 2017–26 time period (CBO 2017c), with more limited effects produced by the per capita block grant. This, of course, is a testament to the fact that per capita Medicaid spending already is low, meaning that in the short term a fixed annual growth rate would yield little in the way of savings. But under a separate CBO estimate of the longer-term consequences of the Senate's version of the legislation, described below, it became clear that the cap itself would cause states to experience catastrophic losses. In assessing the twenty-year impact of the Senate's version of the legislation, the CBO concluded that by 2036 the cap would result in federal funding levels 35 percent below current spending levels (CBO 2017e).
Although the elimination of the ACA's enhancement funding for the expansion population tended to get the most ink, it was the federal spending cap on the traditional program that was so radical. It was the fixed cap that materially altered the long-standing federal-state financial relationship regarding division of responsibility for funding health care for medically indigent populations. One could argue that—viewed through the lens of the Sebelius decision—such a profound alteration of Medicaid's historic federal-state financial arrangement itself represented precisely the type of “shift in kind, not degree” that undergirded the Court's decision to declare the ACA Medicaid expansion unconstitutional if implemented on a mandatory, take-it-or-get-out-of-the-program basis. Indeed, from a constitutional perspective, one might characterize the 1965 Act as an accord between two sovereigns regarding how to account for public costs attributable to the large number of Americans who, by virtue of age, disability, poverty, or all of the above, are unable to afford the cost of necessary care. To be sure, over many decades, the federal government has repeatedly reenvisioned Medicaid's many missions, but in each reenvisioning the basic shared financing accord remained intact. States have relied on the funding that has accompanied this evolution in purpose—often unhappily so, but also in the recognition of Medicaid's essential position in their health care systems and their economies. States may have at times bitterly opposed Medicaid reforms, but under Medicaid's open-ended funding construct, even the reddest states have regularly broadened and deepened their efforts, extending their programs to new populations, building health system capacity, and strengthening their programs’ ability to undertake such massive national goals as the deinstitutionalization of people with disabilities able to live in the community. One can assume that, in the wake of Sebelius, the imposition of an arbitrary growth cap on a take-it-or-leave-it basis ultimately might have raised the same sort of constitutional challenge raised by the ACA expansion itself.
With House passage, the Senate was set to act.
The Second Wave of Attack: The Better Care Reconciliation Act
Despite its assertions that it would set aside the House bill and craft its own measure, the Senate built on the House chassis. BCRA (H.R. 1628, July 20, 2017) was essentially the same measure where Medicaid was concerned, although its block grant provisions actually were harsher, as the Senate looked to Medicaid for the additional funds it would need to offset the impact of reducing subsidies for private insurance. Like the AHCA, BCRA preserved ACA adults as an optional expansion group while eliminating the enhancement payments, although on a somewhat slower glide path. Ultimately, however, the enhanced ACA funding disappeared from the Senate bill as well.
Like AHCA, BCRA imposed a one-year moratorium on federal Medicaid funding for covered family planning and related services furnished by Planned Parenthood and also gave states the option to impose work requirements on low-income adults. As with the House measure, the Senate bill created a block grant option for states (in the case of the Senate, limited to adults) that would have stripped those included in the block grant of their legal entitlement to defined benefits if determined eligible. BCRA's per capita federal spending limit was tighter, capping the federal per capita payment for all populations, including the highest-cost, highest-need populations, at the CPI, without adjustment for medical prices, beginning in 2025. CBO's ten-year BCRA estimates were strikingly similar to those for AHCA. Medicaid continued to account for the bulk of the federal spending reductions and loss of coverage. According to the CBO, by 2026 federal spending on the Medicaid expansion population would decline by 87 percent, since most states would simply drop coverage for this group (CBO 2017f).
By the time BCRA was debated, a series of supplemental analyses performed by outside experts also had emerged. These analyses focused on illustrating state-specific levels of funding losses over time. The Kaiser Family Foundation, calculating federal Medicaid funding reductions to states over 2019–29, concluded that in 2029—following full implementation of the expansion funding rollback and the per capita cap—federal payments to states would be $114 billion below payments due under current law, a 16 percent decline (Kaiser Family Foundation 2017c). California alone would lose $25 billion that year, and New York, $13 billion. A separate expert analysis predicted a range of loss for each state over the 2020–26 time period that would flow from the per capita cap alone; the study concluded that even a cap with growth rate pegged to the CPI-M—above the Senate's post-2025 cap tied to the CPI alone—would result in immense losses to states (Mann et al. 2017). Whatever Medicaid flexibility governors and legislatures might want, no amount of federal deregulation could offset the losses of the magnitude reflected in the House and Senate bills. The demise of BCRA in the Senate was as much a function of its Medicaid provisions as the impact of its private insurance provisions on subsidized coverage.
The Third Wave of Attack: Graham-Cassidy in the Senate
Those who thought that BCRA's demise spelled the end of the line were mistaken. The threat came roaring back in September via a proposal developed by Senator Bill Cassidy (R-LA; a long-time proponent of Medicaid caps), who was joined by Senator Lindsay Graham (R-SC), one of the Senate's most senior leaders but a novice on health policy. Their bill arguably was the most radical of the three. Whereas both AHCA and BCRA would have preserved the ACA's Medicaid adult population as an optional eligibility group, thereby permanently altering Medicaid's fundamental statutory reach, Graham-Cassidy entirely expunged the group from the statute, instead offering limited subsidies for covering the poor in the private insurance market. Whereas its two predecessors would have continued to provide federal financing (although at a lower federal contribution rate) for the expansion population, Graham-Cassidy, by eliminating the expansion group, eliminated the federal funding that went with it. Instead, the bill essentially proposed to divert federal payments being made to expansion states into a single capped federal pool that combined payments to states made to date (and excluding payments that would have been made had all states expanded Medicaid) with an amount representing some portion of the federal payments for premium tax credits and cost-sharing reductions estimated to flow to states. This combined funding pool then would be allocated via an expanded block grant embedded within the CHIP statute (itself a block grant). According to sponsors, this approach offered the added benefit of barring use of federal allotments for abortion, a prohibition already part of the CHIP authorization itself rather than simply a feature of every annual appropriations bill since the first Hyde Amendment abortion exclusion was added in 1976.7 Most astounding perhaps, Graham-Cassidy would have ended all federal funding—even at dramatically reduced levels—in 2026.
Beyond its radical proposal to allocate a fixed amount of federal funds over all states via a time-limited block grant, Graham-Cassidy followed the AHCA/BCRA playbook where the structure of the Medicaid program was concerned: elimination of open-ended entitlement funding in favor of a per capita block grant coupled with other provisions found in the previous bills—the Planned Parenthood one-year federal funding ban, a state option to impose work requirements, elimination of retroactive coverage, and an express block grant option that would enable states to place caps on enrollment by poor adults.8
Despite the fact that Senator Cassidy had sketched out ideas along the lines of his measure for some time, the bill seemingly appeared out of left field. But as the September 30 deadline for legislation tied to FY 2017 approached,9 the proposal began to move at warp speed—so fast, in fact, that the CBO had no time to prepare cost and impact estimates, particularly since the sponsors were revising the provisions of the bill up to the eve of the vote in order to garner support. When the abbreviated CBO analysis finally appeared, it pegged the Medicaid funding losses at $1 trillion over ten years—the highest of the three bills. The cost estimate did not contain actual estimates of coverage losses but instead pegged them simply in the millions, given the radical funding reductions and the resulting limited financing states would receive in relation to the size of the low-income population and the cost of health care (CBO 2017g).
As with BCRA, state-by-state estimates prepared by outside experts also appeared. And like BCRA, the loss of funding and beneficiaries by state was enormous: over the 2020–26 time period, states that had expanded Medicaid would lose an estimated $105 billion as their claimed funds were redirected across all states (Blumberg et al. 2017).—in accordance with the peculiar theory that the expansion states were somehow stealing money that did not rightfully belong to them.
BCRA never came to a vote; the Senate majority leader simply pulled the measure on September 24.
Where Does Medicaid Go from Here?
It is never wise, at least not in the volatile times in which we live, to predict what will happen next. However, a few observations are in order. By the end of 2017, legislation costing the nation $1.5 trillion in lost revenue over the coming decade had been signed into law; any notion that the measure will pay for itself through dynamic growth was belied by virtually all reputable economists and congressional tax and budget experts (Joint Committee on Taxation 2017). In order to start digging out from under the phenomenal losses the measure would bring—not to mention the escalation of private premium costs as a result of repealing the individual mandate (Collins 2017)—the Trump administration naturally included massive federal Medicaid spending reductions in its fiscal year 2019 budget proposal, which formally endorsed the Graham-Cassidy strategy of replacing most Medicaid spending as well as federal insurance premium and cost-sharing subsidies with a much-reduced block grant (President's Budget, 2018).
One might imagine that were the administration willing to make only modest Medicaid spending reductions, it might yet strike a deal with the nation's Republican governors. But the governors made clear, even early in the 2017 legislative process, that while they were interested in more flexibility, they rejected a significant loss of federal funding. Their unwillingness to trade money for statutory relaxation is reflected in a remarkable letter sent to House and Senate leadership in March 2017 by four influential Republican governors (Kasich 2017), although the governors sought broad flexibility to control caseload and coverage, they flatly dismissed the federal funding reductions that congressional leadership sought as the quid pro quo. Don't expect this stalemate to change any time soon in light of Medicaid's importance to state economies. The administration's proposal is expected to go nowhere, although in the guise of welfare reform, it is impossible to say for sure (Rosenbaum 2018a). Damaging legislation is especially risky in a mid-term election year that appears increasingly perilous for Republicans, but the magic of casting aspersions on the poor tends to work well in an era in which the politics of resentment are ascendant.
A more realistic White House strategy involves using its administrative powers to reduce Medicaid by removing as many low-income working-age adults as possible. In this regard, the administration has turned to an approach that involves packaging severe reductions in Medicaid coverage of low income adults as Social Security Act – sanctioned experiments whose aims—like those of the English Poor Laws—are to lift the poor, in this case, by threatening their Medicaid coverage unless they work, pay premiums, and comply with additional and complex reporting requirements. This strategy, according to Seema Verma, the Trump administration's Centers for Medicare and Medicaid administrator (who, in her previous life, packaged and sold the idea of work demonstrations to state Medicaid programs as a private consultant), is intended to transform Medicaid into a program that “helps all beneficiaries reach their highest potential” (Seema Verma 2017).
The legal authority on which this strategy rests is section 1115 of the Social Security Act, which authorizes demonstrations that the HHS secretary determines will further the objectives of the Social Security Act state grant-in-aid program that is the subject of the demonstration. In the case of Medicaid, by law the program's objective is to furnish medical assistance to those who qualify for it (42 U.S.C., sec. 1396a-1), not, as the administration seeks to repurpose it, to put the poor to work. Despite the obvious gap between Medicaid's purpose and the administration's vision of Medicaid as a program for promoting employment, on January 11, 2018, HHS released a program solicitation for state Medicaid work demonstrations (CMS 2018). In fact, states had begun to apply for such demonstration authority well before the January 11 solicitation appeared (Rudowitz, Musumeci, and Hinton 2017). On January 12—one day following its invitation to states to experiment on the poor by threatening them with the loss of insurance unless they worked a certain number of hours, paid premiums, and complied with new paperwork requirements—CMS approved Kentucky's proposal to impose work requirements on all adults younger than 65 years, require premiums even on people with zero income, and add extensive new reporting obligations related to hours worked, premiums paid, changes in family circumstances, and eligibility renewal. As of March 2018, some seventeen states were either in the process of developing work demonstrations, undergoing public review and comment, or awaiting final action by HHS officials.
These Medicaid work demonstration proposals have proliferated even though overwhelming evidence underscores that virtually all working-age poor people work, are looking for work, are too sick to work, or are caring for family (Ku and Brantley 2017). That a policy concept so at odds with reality could move like lightening underscores the enduring power of the trope of the lazy poor person. This image hung over the Medicaid repeal debate, just as it has been used effectively in the past to drive other debates over federal spending on poor adults. Indeed, AHCA, BCRA, and Graham-Cassidy all would have allowed programs to condition Medicaid on work as a simple state option rather than having to meet the far more rigorous legal experimental standard imposed under section 1115.10
Of course, it is beyond argument that many 1115 demonstrations in the past have presented both risks and benefits. For example, the Obama administration section 1115 was used as a strategy to encourage states to adopt the ACA Medicaid expansion, permitting eligibility restrictions greater than those that would be permissible had states simply adopted the expansion as a state plan amendment. Even so, the Obama-era demonstrations produced a net gain in insuring the poor, thereby promoting Medicaid's statutory objective as laid out in law. By contrast the Trump era demonstrations are being pursued explicitly to reduce enrollment, not to generate a net coverage expansion. Whether such a use of 1115—as a tool for undermining rather than achieving the purposes of a program—is not merely unsound policy but also is unlawful is a question that ultimately will be answered in the courts.
Another strategy for undermining Medicaid is to wait for an increasingly conservative judiciary to act. As noted, Medicaid's existence as a recognized legal entitlement rests not on the express language of the law but on how judges have interpreted it. Medicare, for example, states explicitly that those eligible for assistance are legally entitled to their benefits; furthermore, the statute creates a private right of action in program beneficiaries that enables them to protect their legal interests in court when they believe that the federal government has improperly denied benefits to which they are entitled. The judicial process involved in protecting Medicare benefits may be slow and complex, but its very existence acts as a safeguard against arbitrary governmental action. The existence of a statutory legal right to coverage and a separate and explicit right to turn to the courts when benefits are threatened are the indispensable elements of entitlement laws (Jost 2003).
Medicaid is different. The statute lacks express legal entitlement language, a seemingly fatal omission in an age when an increasingly aggressively conservative judiciary is focused on constraining the existence of privately enforceable right (Rosenbaum, 2015). Furthermore, the Medicaid statute is silent on whether those whose benefits are threatened by potentially unlawful state conduct can turn to the courts in advance of HHS review of the legality of state actions. In Medicaid's case, the existence of a right to benefits is the function of a line of judicial rulings spanning a half century that have recognized the language of the statute as containing rights-creating language enforceable through the use of section 1983, a post–Civil War statute authorizing private suits to enjoin state actions that may violate federally secured rights. This separate civil rights statute has been the basis of actions against state governments to halt policies and practices that unlawfully interfere with access to coverage (Rosenbaum et al. 2012: chap. 11). In sum, Medicaid's very existence as a legal entitlement in individuals, hinges on the courts’ continued allegiance to decades of judicial interpretation regarding the interaction between Medicaid's statutory text and a separate civil rights private enforcement law.11
For years now there have been signs that this judicial recognition of Medicaid as conferring legally enforceable rights is on the verge of ending in the face of a succession of Supreme Court rulings that increasingly have tightened judicial standards regarding when a right to assistance can be said to exist (Rosenbaum et al. 2012: chap. 11). In 2015, in Armstrong v. Exceptional Child Center, Inc. (135 S.Ct. 1378) the Court pushed matters further, introducing new limits on access to the courts by private individuals in cases involving potentially unlawful state actions that threaten access to care. Armstrong focused on a provision of the Medicaid statute, 42 U.S.C., sec. 1396a(a)(30), whose purpose is to ensure that payment rates are sufficient to create access to care equal to that enjoyed by the population generally. Armstrong considered the question of whether this specific provision contains the type of rights-creating language necessary to create enforceability. However, speaking for as yet a minority of the justices, Justice Antonin Scalia would have gone further, declaring Medicaid as being akin to a contract between the federal government and the states rather than one that confers rights on individuals, thereby placing it entirely outside the purview of 1983.
Whether a newly configured Court will take the ultimate step and effectively overturn decades of prior rulings affecting the right to Medicaid coverage itself remains to be seen. But in a world in which conservative activist judges are willing to entertain far-reaching cases that overturn liberal precedent, one can expect further efforts by states to maneuver Medicaid cases into court in order to test whether the more liberal philosophy about protecting the interest of individuals who are the intended beneficiaries of a program remains a feature of Medicaid. Indeed, this effort to broaden the reach of Armstrong to encompass other provisions of the statute, such as the statutory language that guarantees individuals free choice of qualified providers, is under way; these cases focus on the power of states to exclude Planned Parenthood from Medicaid on grounds unrelated to its qualifications as a family planning provider (Rosenbaum 2017a). Indeed, in the wake of an appellate decision denying affected beneficiaries the ability to challenge exclusion of Planned Parenthood from the Arkansas Medicaid program, the Trump administration has wasted no time trying to maneuver into court even more state actions to bar these vital clinics from furnishing covered services they are eminently qualified to provide. The obvious hope on the administration's part is that the Supreme Court ultimately will extend Armstrong to Medicaid's freedom of choice protections, thereby shielding unlawful state actions against Planned Parenthood clinics from early judicial review, before the damage is done (Rosenbaum 2018b).
To the extent that the courts withdraw the protective shield they have built to stop the arbitrary and unlawful withdrawal of Medicaid benefits, one could foresee widespread state efforts to reduce their rolls, seemingly confident in the knowledge that the HHS secretary will do nothing to stop them. Inaction by federal officials in the face of ongoing state violations of federal Medicaid law is nothing new, but under the Trump administration federal refusal to intervene is likely to reach new heights.
In the end, what may save Medicaid is the level of public support for the program. One of the more remarkable aspects of the 2017 legislative battle to end Medicaid as we know it was its surge in popularity, as evidenced by public opinion polls showing extensive support for the program across the political spectrum (Kaiser Family Foundation 2017d). The fact of the matter is that, once a program grows to 75 million children and adults, everyone knows someone who depends on it—a child or adult with a disability, an elderly parent, a pregnant woman, a woman with breast cancer. At this point, it becomes difficult, to put it mildly, to simply unravel the program. But we can and must assume advocates for an unraveling will keep trying. In this regard, preparation must be far more than holding on in a defensive crouch. It necessarily involves an unceasing effort to make Medicaid work well for the populations it serves. Extensive evidence suggests that, despite monumental challenges, Medicaid does so (Paradise 2017). But over the coming years, the need to show that even a mammoth program can perform efficiently and effectively will, if anything, become more urgent, as will the need for thoughtful research that can continually document Medicaid's worth.
The author thanks the Commonwealth Fund for its ongoing support of the Medicaid analytic studies that made this essay possible.
The Trump administration, in what can only be viewed as an effort to destabilize this most basic of ACA achievements, virtually eliminated outreach, further slashed the annual open enrollment period and tightened special enrollment rules (Shafer and Dusetzina 2017).
Cannon's exhortation regarding rejection of state exchanges ultimately would help fuel the legal challenge to the Obama administration's payment of premium subsidies to eligible people living in states that relied on the federal exchange, a challenge that ultimately was rejected as baseless by the Supreme Court in King v. Burwell (135 S.Ct. 2480 ).
Among its amendments, the House bill would have restricted eligibility for lottery winners, saving about $610 million over a decade according to the Congressional Budget Office in separate estimates prepared for the extension of funding for the Children's Health Insurance Program (CBO 2017). This budgetary savings represented less than 10 percent of the ten-year Medicaid funding cuts that the House bill ultimately would seek to impose through federal spending caps.
The evidence offered for this diversion-of-resources argument was state waiting lists for people in need of community-supported services, a charge that turned out to be patently false; indeed, waiting lists tended to be found in nonexpansion states, which also tend to be states that devote fewer resources to Medicaid (Musumeci 2017).
As discussed below, these early exhortations would assume a more formal shape in November 2017, when the administration formally unveiled a broad initiative to use its special Social Security Act demonstration authority to encourage states to pursue experiments reducing medical assistance to the poor (Rosenbaum 2017b).
By permitting states to set an arbitrary limit on enrollment, the block grant option, like its 1981 and 1995 predecessors, likely would have effectively eliminated any claim of legal entitlement to coverage on behalf of those whose coverage was linked to block grant funding.
Moving antiabortion riders from annual appropriations legislation into the statutes that authorize the very existence of federal spending programs is, of course, the Holy Grail on the part of those who favor such restrictions.
As with prior bills, the two sponsors offered no explanation regarding what would happen to adults excluded from eligibility for any assistance.
Fashioned as a measure tied to the annual budget, the ACA repeal-replace strategy was effectively tethered to this arbitrary time period.
Within ten days of the Kentucky approval, advocates for the poor had sued to prevent the HHS secretary from using his 1115 authority to implement Medicaid work demonstrations. The complaint alleges that the secretary lacks the power to approve such demonstrations under 1115, which limits the use of experimental authority to pilots and demonstrations that promote the objectives of programs that are the subject of a demonstration. By statute, Medicaid's objective is to insure the poor, not put people to work. Stewart v. Azar (D.D.C. Civil Action No. 1:18-cv-152 Filed January 24, 2018).
States’ entitlement to federal funding they are owed, by contrast, is expressly stated in law.