The political processes surrounding the Affordable Care Act (ACA) offer valuable lessons about race and politics in the United States. In particular, the ACA underscores a critical tension between politics and policy in a racialized polity: even when policies are intended to target and address racial disparities, politics can undermine the steps necessary to do so. Close scrutiny of the ACA during its first decade reveals how race intersects with politics to render public policy less equitable and more vulnerable to erosion. Ultimately, this analysis points to the ways that racialized political processes are formidable barriers to equitable material outcomes. By examining such processes and making them visible, this article elucidates the possibilities, limits, and contours of public policy as a mechanism for achieving racial justice.
Analyzing the political processes surrounding the Affordable Care Act (ACA) can teach us valuable lessons about race and politics in the United States. In particular, the ACA underscores a critical tension between politics and policy in a racialized polity:1 even when policies are intended to winnow racial disparities, politics can undermine the steps necessary to do so. Close attention to the implementation of the ACA reveals how race intersects with politics to render public policy less equitable. Only by scrutinizing such processes can we discern how policies and politics might be wielded to achieve racial justice in health care.
The ACA and Racial Inequality
The ACA was designed to reduce health inequities based on race and ethnicity (Ossei-Owusu 2016). The text of the original bill (Pub. L. No. 111–148. 3–23–2010) contained 34 references to “disparities,” 28 references to either “discrimination” or “non-discrimination,” 33 instances using either the word racial or race, and 35 instances using either the word ethnicity or ethnic. Though Barack Obama's approach to advancing his policy goals was often deracialized, the explicit emphasis on race in the ACA reveals a pronounced goal of diminishing racial disparities (Gillion 2016; Lewis, Dowe, and Franklin 2013). Such intentions notwithstanding, the ACA reflects an incongruity between politics and policy. On the one hand, health care politics became more deeply racialized during the presidency of Barack Obama and has remained so (Banks 2014; Fiscella 2016; Knowles, Lowery, and Schaumberg 2010; Maxwell and Shields 2014; McCabe 2019; Mitchell and Dowe 2019; Morone 2018; Tesler 2012). On the other hand, ACA policy was a harbinger of racial promise. Even in the face of antagonistic racial politics, with white Americans disproportionately opposing Obamacare, the policies of the ACA had “the potential to truly alter the landscape of racial and ethnic health disparities in the United States” (Mitchell 2015: e-66). Indeed, from the vantage point of those concerned with the legacy of racism in the United States, the ACA was viewed as “a stealthy civil-rights achievement of the Obama presidency, promising to make health care less of a financial burden, end disparities in health-care coverage, ease barriers to access for people of color, and subsidize preventative health-care services that proved especially lacking in black neighborhoods” (Newkirk 2016).
In retrospect, some of these expectations proved true. The ACA reduced racial/ethnic disparities in health insurance coverage, access to care, and health care utilization (Buchmueller et al. 2016; Chaudry, Jackson, and Glied 2019; Chen at al. 2016; Gutierrez 2018; Lipton, Decker, and Sommers 2019; McMorrow et al. 2016; Park et al. 2018). The reduction of insurance coverage gaps was one of the most salient ways that the ACA had a salutary effect on racial inequity. Between 2013 and 2017, the coverage gap between black and white Americans declined from 11.0 to 5.3 percentage points (Chaudry, Jackson, and Glied 2019). Similarly, during the same period, the coverage gap between Hispanics and non-Hispanic whites dropped from 25.4 to 16.6 percentage points.
Despite such good news, the story of race and the ACA is not a straightforward narrative of success. Racial imbalances in health care access and quality persist in the post-ACA era (Artiga, Orgera, and Damico 2019; Buchmueller et al. 2016; Yue, Rasmussen, and Ponce 2018). Moreover, the progress of the ACA in lessening racial disparities has begun to plateau or reverse (Artiga, Orgera, and Damico 2019). Insurance coverage is again a good example. After the above-mentioned improvements in coverage rates between 2013 and 2017, the overall uninsured rate rose from 7.9% in 2017 to 8.5% in 2018 (Berchick, Barnett, and Upton 2019). Hispanic Americans were most affected, with a 1.6–percentage-point increase in their uninsured rate.
To better understand this and other shortcomings of the ACA with respect to racial equity, we must look to politics. Doing so uncovers distinctive patterns that have stunted the ACA's ability to properly function as a “civil rights achievement.” In this vein, I make two observations. First, even the most salient inequality-reducing feature of the ACA—Medicaid expansion—has endured politically induced variation, attenuating its effectiveness in diminishing racial disparities. Second, beyond Medicaid expansion, many of the numerous features of the ACA that explicitly target racial disparities have proven unstable or limited because their implementation has been contingent on political conditions.
The common thread uniting both of these points is that racialized politics constrains American public policy as a tool for equity. To detail this claim more precisely, I delineate complex and consequential connections between race, policy, and politics in American health care.
The Racialized Politics of Medicaid Expansion
As it was originally designed, one of the ACA's boldest and most promising mechanisms for reducing racial inequities was the expansion of Medicaid. Per the initial formulation of the ACA, Medicaid expansion would have offered public health insurance to all Americans with incomes at or below 138% of the federal poverty line. To secure the participation of every state, the federal government brandished both a carrot and a stick. The carrot consisted of generous federal funding that would cover 100% of the costs of expansion for nearly 3 years (from the beginning of 2014 through the end of 2016) and then gradually phase down to 90% by 2020. The stick meant that states refusing expansion would forfeit all of their federal Medicaid funding (not just the extra expansion resources). This combination of incentives and sanctions was intended to ensure the geographic consistency of Medicaid expansion, an outcome that would have been a major departure from the norm. Prior to the enactment of the ACA, access to Medicaid was limited and highly unequal (Michener 2018). Variable categorical eligibility criteria at both the national and state levels meant that program benefits were heterogeneous across groups (with children, the elderly, pregnant women, and other specific groups often receiving more generous benefits) and across states (with some locales offering a wider scope of benefits and broader eligibility criteria than others). If Medicaid expansion had proceeded as originally planned, this patchwork policy design would have been augmented with a more standardized national approach applied to all Americans at or below 138% of the federal poverty line. Though the planned expansionary tack was not explicitly race based, the outsized presence of blacks and Latinos among the population of Americans living in or near poverty (e.g., 20% of Medicaid beneficiaries are black and 30% are Latino; KFF n.d.) meant that uniform national expansion of Medicaid would have had inequality-reducing racial effects.
Despite the initial objectives of the ACA, political processes fundamentally altered its course. Just over 2 years after the passage of the law (and before its full implementation), the Supreme Court issued a decision declaring Medicaid expansion partially unconstitutional. In National Federation of Independent Business v. Sebelius, 567 U.S. 519 (2012), the court held that by threatening noncompliant states with the loss of all Medicaid funds, the 2010 expansion was coercive. The Sebelius decision transformed the trajectory of the ACA, empowering states to eschew the expansion if they saw fit to do so. Many states did. Decisions about whether to expand largely (though not entirely) fell along partisan lines (Barrilleaux and Rainey 2014; Callaghan and Jacobs 2016; Jacobs and Callaghan 2013). States with Democratic legislative majorities and Democratic executives adopted the expansion most swiftly, while states with divided governments or Republican legislative majorities were less likely to do so, particularly in the South.
The intercession of the Supreme Court in the Sebelius case was consistent with the enduring role of federalism in the American political system, specifically with regard to health care, and especially concerning racialized resources (Michener 2018). People of color have long been disproportionately disadvantaged by federalism (Brown 2003; Lieberman and Lapinski 2001; Lowndes, Novkov, and Warren 2008; Miller 2008; Riker 1964; Soss, Fording, and Schram 2011; Tani 2016). Indeed, federalism “has been one of the chief bulwarks of racial domination in the United States” (Brown 2003: 56). Particularly (but not exclusively) with respect to health care policies, racial disparities have been the frequent outcome of enhancing state discretion (Michener 2018). Given this larger national historical context, the court's decision in Sebelius had clear negative implications for the racial equitability of health resources.
The most concrete upshot of Sebelius is that many southern states have been able to evade Medicaid expansion (see fig. 1). The resulting racial distributional patterns have been stark. A 2015 Kaiser Family Foundation report (Artiga, Damico, and Garfield 2015) found that more than 60% of uninsured poor black adults excluded from Medicaid due to states' refusal to expand (i.e. those in the coverage gap) lived in just four southern states: Georgia (19%), Texas (16%), Florida (14%), and Louisiana (11%). Among Latinos, the patterns were even more striking. Nearly 8 in 10 Latinos in the coverage gap resided in just two states: Texas (52%) and Florida (27%). More generally, many of the southern states that declined to adopt were places with large shares of blacks (Mississippi, 38%; Louisiana, 32%; Georgia, 31%; Alabama, 27%; South Carolina, 27%) or Latinos (Texas, 40%; Florida, 26%). Ultimately, the racial demographics of the South have meant that the concentration of nonexpansion states in the region is a source of significant racial inequality in health care access.
Crucially, this inequity was induced by racialized political decisions. Numerous studies have demonstrated this. Lanford and Quadagno (2015) found that racial resentment was closely linked to Medicaid expansion, with lower racial sympathy and higher racial resentment (on the state level) correlated with stronger resistance to Medicaid expansion. Grogan and Park (2017) found that Medicaid expansion was racialized in terms of public support (with whites much less likely to support expansion) and policy adoption (with state expansion decisions positively correlated with white opinion but uncorrelated with nonwhite support). Grogan and Park also found that when the size of the black population increased and white support was relatively low, states were significantly less likely to expand Medicaid. This helps us to make sense of nonexpansion in southern states with significant health care needs but large black populations. Racial representational disparities combined with racial differences in policy preferences have been barriers to Medicaid expansion. As a result, there are significantly higher proportions of uninsured Americans in nonexpansion states, with the largest and most evident disadvantages among people of color (see fig. 2).
The Racialized Politics of Medicaid Demonstration Waivers
State Medicaid demonstration (section 1115) waivers are also crucial elements of the ACA with implications for race and politics. Waivers provide states with the flexibility to implement new policies that (ostensibly) benefit state residents. In the post-ACA era, waivers have offered states an avenue for shaping Medicaid to suit their political prerogatives. In particular, section 1115 waivers have allowed Republicans who may otherwise be loath to adopt Medicaid expansion to do so, thus providing insurance to thousands of low-income state residents while signaling a distance from the Democrats who passed the ACA. The unprecedented uses of 1115 waivers in the post-ACA period are racialized as a result of racial disproportionalities in Medicaid, broader economic and social inequities, and racial biases at the root of policies such as work requirements.
Arkansas is a prime illustration. In March 2018 the state proposed the Arkansas Works program via an amendment to a prior section 1115 waiver. The previous waiver (2013) had already placed Arkansas at the vanguard of innovation through a Medicaid demonstration that created what came to be known as the “private option.”2 By initiating Arkansas Works, the 2018 amendment to the initial waiver went in a very different direction, heavily focusing on work-reporting requirements. The Centers for Medicare and Medicaid Services approved the proposal, ushering in a key change: a “work and community engagement” requirement for members of the expansion group under age 50. This necessitated that during any given month beneficiaries must either meet an exemption (e.g., medially frail, pregnant, etc.) or complete 80 hours of qualifying activities, including employment, education, and community service.
The most prominent result of this policy was massive coverage loss. In just 7 months more than 18,000 people were disenrolled (Wagner and Schubel 2019). Early research surveying low-income Arkansans has confirmed that work-reporting requirements are associated with a substantial loss of Medicaid coverage, a rise in the percentage of uninsured persons, and no significant changes in employment (Sommers et al. 2019). Most germane, the case of Arkansas epitomizes the significance of race in understanding Medicaid waivers. Racial disproportionalities in the state's Medicaid program mean that black Americans are particularly vulnerable to the negative effects of Arkansas Works (Michener 2019; Sommers et al. 2019). Though black people account for less than 13% of the US population and only 15% of Arkansas residents, they make up 26% of Medicaid beneficiaries in the state. This means that 47% of black Arkansans rely on Medicaid (compared to 25% of white Arkansans and 31% of Latino Arkansans). Any policy that results in the large-scale loss of Medicaid coverage is likely to have lopsided racial consequences. Indeed, consistent with this expectation, recent analyses of the effects of work requirements in the Supplemental Nutrition Assistance Program and Temporary Assistance to Needy Families show that for both programs black Americans were unevenly affected by work requirements (Brantley, Pillai, and Ku 2019; Hall and Burrowes 2019). These studies suggest that, even beyond Arkansas, work requirements do not bode well for racial equity.3
Over and above racial disproportionalities in Medicaid, broader social and economic racial biases also shape the effects of work requirements. For example, research demonstrates how detrimental a criminal record is to labor market outcomes, especially for black people (Pager 2008). In a state like Arkansas, the incarceration rate for blacks is very high relative to whites, making a criminal record a barrier to employment for significant subsets of black Arkansans (see fig. 3). By necessitating employment as a qualification for receiving Medicaid, work requirements can thus intersect with disparities in other institutional venues (labor market, criminal justice system) to spread disadvantage across arenas.
Even beyond the disadvantage of a criminal record, compelling evidence points to wide-ranging racial discrimination in low-wage labor markets (Bertrand and Mullainathan 2004; Pager, Bonikowski, and Western 2009). The black unemployment rate is regularly twice that of white unemployment (Wilson 2019). Conditioning Medicaid benefits on seeking and finding work is more burdensome to racially marginalized populations facing significant structural obstacles to employment.
Race is also an imperative aspect of the politics of Medicaid demonstration waivers because the assumptions buoying support for policies like work requirements are themselves racialized. Many Americans overestimate the presence of black people on welfare rolls (Delaney and Edwards-Levy 2018). Racial stereotypes about undeserving black people taking advantage of government largesse underlie widespread opposition to public benefits, motivating support for policies that make those benefits harder to get (Brown-Iannuzzi et al. 2019; Gilens 1999).
Race and the ACA: Beyond Expansion
Many ACA provisions directly targeted racial disparities. These policies cover a gamut of issues, but their overarching goal is to explicitly marshal the resources of the federal government to reduce health disparities. Three illustrative initiatives in the ACA include:
Consistent and systematized health data collection by race, ethnicity, and language
Increased racial and ethnic diversity in the health care workforce
Nondiscrimination in health programs and activities
All three policy emphases reflect potential mechanisms for advancing health equity (Andrulis et al. 2013; Bristow, Butler, and Smedley, 2004; Dovidio and Fiske 2012; Ulmer, McFadden, and Nerenz 2009).
As some of the ACA's most targeted efforts to confront racial inequality, the implementation and politics of these policies are instructive. Table 1 summarizes some specific statutes, identifies the entities responsible for implementing them, and details the funding associated with each.4 The key takeaway is that implementation is highly contingent on political conditions (e.g., the positioning of federal bureaucrats, and recurring appropriations). Given this reality, ACA statutes directly confronting race (those with the most overt “civil rights” implications) have been acutely vulnerable to “inconsistent and fluctuating levels of federal engagement” (King 2017: 357). This has been especially true in the hyperpolarized partisan political environment that shapes ACA implementation (Béland, Rocco, and Waddan 2015). The consequence has been “the slowing down or outright death of federal civil rights activism” because the “enforcement of policy is weak. Many of the institutional reforms and national standards needed for amelioration are often given insufficient resources for effective implementation” (King 2017: 357–58). A closer look at the trajectory of the data collection, workforce diversification, and nondiscrimination initiatives underscores this.
Race/Ethnicity Data Collection
Section 4302(a) of the ACA stipulates that the Department of Health and Humans Services (HHS) secretary “shall ensure that, by not later than 2 years after the date of enactment of this title, any federally conducted or supported health care or public health program, activity or survey (including Current Population Surveys and American Community Surveys conducted by the Bureau of Labor Statistics and the Bureau of the Census) collects and reports, to the extent practicable—(A) data on race, ethnicity, sex, primary language, and disability status for applicants, recipients, or participants.” The objective of this statute was to “support a more focused national strategy to eliminate health and health care disparities among Medicaid and CHIP enrollees” (Burwell 2014: 2).
Particularly crucial is that the main implementing authority is the HHS secretary—a salient and high-profile political appointee. Between 2010 and 2016, when HHS was run by appointees of President Barack Obama, there was significant federal activity related to enforcing section 4302. The Federal Register recorded 15 rules that cited section 4302, most between 2012 and 2016 (National Archives n.d.). In the 3–year period following that (between the time the Trump administration took control of HHS in January 2017 and early January 2020) only three rules have cited section 4302. While race/ethnicity data have “remained largely incomplete,” there is little indication that data collection remains a federal priority (Ng et al. 2017). Near exclusive reliance on the HHS, a politically salient bureaucracy run by a political appointee, has limited the effectiveness of section 4302.
The ACA contains numerous provisions to enhance racial diversity within the American health care workforce. Many have proven difficult to implement. Even during the Obama administration, a polarized Congress refused to appropriate $3 million to establish the National Health Care Workforce Commission as stipulated by ACA section 5101. Securing appropriations for other aspects of the law related to workforce diversity has been similarly challenging. Statutes related to workforce diversity remain either unfunded or underfunded.5 Such programs are supported “in intent” by the ACA but have perennially faced severe resource constraints (Andrulis et al. 2013). The shift in federal priorities since the election of President Trump has further imperiled funding. The fiscal year 2020 budget proposal for HHS proposed eliminating $88 million for diversity training and $151 million for nursing workforce development (DHHS n.d.). Given continued partisan polarization, it is not at all clear “whether the resources and political will to support a broad spectrum of critical programs and actions will be sufficient to meet service goals and people's need” (Andrulis et al. 2013).
Section 1557, the nondiscrimination provision of the ACA, prohibits discrimination in health care programs on the basis of race. Building on existing federal civil rights laws, section 1557 extends nondiscrimination protections to individuals participating in “any health program or activity, any part of which is receiving Federal financial assistance.” Though 1557 did not mandate follow-up regulatory activity, the Obama administration reinforced the law by proposing a “final rule” for implementation (81 C.F.R. 31375). Among other things, the final rule required that covered entities post notices of nondiscrimination in the top 15 languages spoken statewide. These requirements were most directly relevant to people with limited English proficiency, a population that is 88% nonwhite (Batalova and Zong 2016).
Section 1557 has been directly undermined by shifting political winds. In May 2019, the Office for Civil Rights and HHS proposed to revise the final rule issued by the Obama administration in 2016 (National Archives 2019). The 2019 iteration of the rule repealed requirements “to mail beneficiaries, enrollees, and others, notices concerning non-discrimination and the availability of language assistance services.” To justify this change, the Office for Civil Right and HHS point to Executive Order 13765, “Minimizing the Economic Burden of the Patient Protection and Affordable Care Act Pending Repeal” (2017). This order asserts the federal government's goal of curtailing the financial burdens of the ACA and directs executive branch agency heads in charge of ACA enforcement to “exercise all authority and discretion available to them to waive, defer, grant exemptions from, or delay the implementation of any provision or requirement of the Act that would impose a fiscal burden.”
There is not yet any accounting of the precise effects of such regulatory change on racial health disparities. The larger point, however, is that the politics of the ACA regulatory process unraveled antidiscrimination policies that many people saw as a step forward for racial equity. Indeed, HHS received nearly 156,000 comments responding to its 2019 regulatory change. A (unsystematic) review of these comments indicates that many of them were in opposition to the rule. Yet, the scope that the ACA left for regulatory and bureaucratic maneuvering enabled important policies to be tightly tethered to political conditions.
In a polity where race is a political flashpoint, ACA policies meant to (indirectly or directly) address racial disparities were politically significant steps. Yet, those steps were deeply precarious. In this way, the ACA epitomizes a difficult problem in American politics: the distance between policy intentions and policy outcomes cannot be bridged without attending to the constraints of profoundly racialized social, economic, and political systems.
Following Gotham (2000) and Bonilla-Silva (1997), I define racialization as economic, social, and political processes by which “people are sorted into racial categories, resources are distributed along racial lines, and state policy shapes and is shaped by the racial contours of society” (Gotham 2000: 293).
Arkansas's initial 2013 Medicaid expansion (under Democratic governor Mike Beebe) occurred through a 1115 waiver that allowed the use of state Medicaid funds to provide premium assistance to eligible beneficiaries, enabling them to purchase private health insurance via the state health insurance marketplace. This allowed Arkansas to insure roughly 220,000 Medicaid beneficiaries via commercial provider networks.
Medicaid work requirements have been approved in ten states: Arizona, Arkansas, Indiana, Kentucky, Michigan, New Hampshire, Ohio, Utah, South Carolina, and Wisconsin. In four of these, work requirements have been set aside by state courts (Kentucky, Arkansas, Michigan, and New Hampshire). In several others, they have not yet been implemented (Arizona, Wisconsin, Utah, and Ohio). Numerous other states have pending waivers that await approval, including several states with substantial black populations (Mississippi, Georgia, and Tennessee).
Table 1 does not provide an exhaustive accounting of all relevant provisions. The ACA contains a vast array of policies, and even those highlighted here are more detailed and nuanced than there is space to elaborate.
One important exception worth studying further is the Health Profession Opportunity Grant, authorized via section 5507 of the ACA. This grant project has received consistent funding from Congress.