How the Two-Tier Welfare State Undermines Political Participation
At the end of September, millions of Americans once again breathed a collective sigh of relief as the Graham-Cassidy bill, the worst proposal yet aimed at dismantling the Affordable Care Act (ACA), fizzled in Congress. Like previous proposals, Graham-Cassidy called for vast retrenchment across the healthcare system and aimed to essentially block grant Medicaid, thus stripping the program of its entitlement status and giving states only a limited amount of funding for its administration each year, regardless of need. The Graham-Cassidy proposal came just months after the American Health Care Act (AHCA), the “repeal and replace” bill that called for many of the same reforms, died on the Senate floor (with substantial credit due in both cases to direct action by disability rights activists).
Yet although Graham-Cassidy is dead, the ideology behind it is alive and well, as illustrated by recent state-level efforts to further eviscerate the existing system of healthcare for low-income families. In Maine, for example, which this November will become the first state to decide whether to expand Medicaid eligibility as permitted under the ACA via popular vote, Gov. Paul LePage is making every effort to further stigmatize the program to undermine the expansion effort. In August, LePage threatened to sue the Secretary of State if the program was referred to as “insurance” rather than “welfare” on the ballot initiative, shamelessly seeking to capitalize on welfare stigma and the trope of the “undeserving poor” in an effort to deprive 70,000 low-income Maine residents of access to fundamental care.
These attacks on Medicaid sharply diverge from the other key healthcare proposal that has emerged in 2017: Medicare for All. Last month, California Sen. Kamala Harris made headlines when she announced she would co-sponsor Sen. Bernie Sanders’ “Medicare for All” bill, becoming the first of 15 senators (and counting) to publicly declare that “healthcare should be a right, not a privilege.” For the many activists on the left who have long been advocating for single payer healthcare despite a recalcitrant Congress, Sen. Harris’ endorsement, and the cascade of support that followed, served as a vindicating example of the effectiveness of putting steady pressure on center-left politicians. Likewise, in Sen. Harris’ home state, SB 562, a bill to establish a single-payer system that was backed by National Nurses United, the California Nurses Association, and 350 other community and labor organizations passed out of the Senate and gained substantial momentum throughout the spring before it stalled in committee. If passed, the bill would make California the first state in the U.S. to provide universal health coverage.
It goes without saying that these two approaches—drastically cutting or restricting access to Medicaid and “Medicare for All”—would have vastly different implications for health. It’s estimated that the Graham-Cassidy bill would have resulted in 32 million people losing health coverage, whereas SB 562 would newly extend coverage to nearly 3 million in California alone. But they would also have sharply contrasting implications for democracy—and we need to care about both.
Over the past two decades, a growing body of research on “policy feedback” has found that the design of government programs can either cultivate or inhibit political participation by beneficiaries. Generally, experience with means-tested programs is associated with lower voter turnout and engagement relative to participation in universal programs. Stigmatizing and punitive programs like Temporary Assistance for Needy Families (TANF, or colloquially, “welfare”) have been shown to have the most negative effects on measures of political participation. As explained by Bruch, Ferree and Soss (2010),
“Policy designs do more than just distribute resources; they convey potent messages about political identities, possibilities, and realities...Personal experiences with public policy have the power to teach lessons about group status, government responsiveness, and the efficacy and wisdom of exercising voice as a citizen.”
While prior studies have already confirmed that lower-income people are less likely to vote, due in no small part to a wide range of legal and structural barriers, policy feedback research goes a step further by examining how the design of public assistance programs influences these socioeconomic disparities and tacitly shapes understandings of political agency.
For example, in a longitudinal study of over 1000 young adults in St. Paul, Minnesota, Swartz et al. (2008) found that those who received “welfare” were significantly less likely to vote than both those who received no government benefits and those who received “first-tier” government benefits, such as unemployment insurance, worker’s compensation, or disability benefits. In the 2000 election, just 29% of welfare recipients voted, compared to 52% of those receiving other government assistance and 43% of those receiving no assistance. In addition, welfare recipients reported significantly lower views of their political self-efficacy than the other two groups. The authors hypothesized that these disparities were likely “due, in part, to the discretionary and punitive administration of welfare programs” and the stigma associated with them.
Likewise, in a study looking at individuals’ cumulative lifetime experience with 18 different social programs, Mettler and Stonecash (2008) found that the more universal programs people had accessed, the more likely they were to vote, while the converse was true for means-tested programs, controlling for income, age, education, and other demographic factors. At the same time, respondents with direct experience with either type of program were significantly more likely to describe those programs as effective in fulfilling their stated purpose. For example, 61% of those who had received welfare described it as “effective,” compared to just 28% of non-recipients. The authors summarized the implications as follows:
“[T]he same individuals whose experiences of means-tested programs appears to have given them more confidence in government programs are also less likely to vote. Thus, the political voices of those who have a positive message to convey about government are less likely to be heard. This may leave the programs from which such individuals benefit less well represented in the political process, such that they are less likely to have their benefits increased to maintain their value in light of inflation and more likely to become subject, over time, to more punitive and restrictive rules and procedures.”
Further, other research finds that even among means-tested programs, different policy design choices produce different results for civic and political participation. Those programs with the most punitive and paternalistic authority structures, such as TANF, have a much more negative impact than programs with more democratic or participatory structures. And even within TANF, there is a significant variation depending on the policy choices made at the state level about features like work requirements, sanctions, and time limits. According to Bruch et al., state TANF programs with the most paternalistic and punitive policy designs have much stronger negative effects on political and civic participation than those with less paternalistic policies. Notably, recent research from the Urban Institute found that states with larger Black populations have stricter TANF policies, including harsher sanctions—suggesting that all of these deleterious effects of welfare on political participation are likely to disproportionately impact low-income communities of color.
By vastly cutting its funding and giving states the discretion to put new conditions on eligibility, the Graham-Cassidy bill, like the AHCA before it, sought to lay the groundwork for making Medicaid more like TANF. Yet even with national reforms in limbo, at the state level, Republican governors are taking steps to emulate the deep policy retrenchment put in motion by President Clinton’s 1996 welfare reforms by transforming Medicaid into a more stigmatizing and paternalistic program as soon as possible.
For example, in Wisconsin, Gov. Scott Walker has been seeking to make his state the first to require drug tests for Medicaid beneficiaries, as at least 10 states have done for TANF. In Maine, Gov. LePage is aiming to impose work requirements for Medicaid and bring back the asset test, despite evidence that many Medicaid recipients have health barriers to work and the vast majority have hardly any assets to speak of. And in perhaps the most creative approach to restricting access to fundamental care, Kentucky Gov. Matt Bevin recently proposed a new “rewards” program through which “people can accumulate points for activities such as passing a GED exam, completing job training or completing wellness activities such as stop-smoking classes, points that go toward the purchase of services such as dental or vision care.”
As documented by political scientist Jamila Michener, who has a forthcoming book specifically examining the policy feedback effects of states’ Medicaid policies, “When Medicaid is administered via processes that are belittling and degrading, beneficiaries come to believe that it’s hopeless to try to change the system, or influence the world.” It makes sense; if your primary experience with the state is as a target of suspicion, surveillance, and stigma, your faith in the capacity of government institutions to do good—and your ability to affect those institutions’ actions—will naturally erode.
None of this is not to say that people participating in Medicaid, TANF, and other means-tested programs have not been politically engaged – some of the most powerful social movements have had their origins in organizing and activism by welfare recipients. And welfare recipients have also long been calling attention to these programs’ flaws, as illustrated by the National Welfare Rights Organization’s calls for a universal basic income nearly 50 years before it was being touted as a “financial innovation” by Silicon Valley executives. But the existing systems aren’t set up to support this kind of solidarity or group identity; they’re set up to stigmatize and disempower.
Importantly, making different choices can produce different results. Just as policy feedback research has found that stigmatizing, “second tier” programs like TANF can negatively affect political engagement, universal, “first-tier” benefits can have the opposite effect. Social Security, for example, the strongest universal program we currently have in the U.S., has been found to increase political participation among low-income retirees. Likewise, as Mettler has found, recipients of the G.I. Bill’s education subsidies, which enabled 7.8 million Americans returning from World War II to attend college for free, “experienced a universal program characterized by generosity, fairness, and inclusivity, one that incorporated them more fully as citizens; their subsequent rates of political participation increased as a result” (though it’s critical to note that Black veterans were largely excluded from the G.I. Bill’s housing loans, with long-term implications for the racial wealth gap). Finally, Head Start, despite being a means-tested program, has been found to positively affect political and civic participation, which is likely explained by its active engagement of parents in decision-making, which contrasts with TANF’s top-down, punitive orientation.
We should all care about structure of our healthcare and public assistance programs because of their direct effects on health and households’ finances. Yet we do ourselves a disservice if we divorce these decisions about the social safety net from broader conversations about political power and voice. Lower voter turnout rates by lower socioeconomic status households are not inevitable, which is made plain by the considerably higher levels of participation by low-income Americans just a few decades ago. Addressing the barriers to broader participation will require action on many fronts, from expanding automatic registration to ending felony disenfranchisement and voter ID laws. But the overall structure and even detailed policy features of social programs also matter, and ensuring that the way we deliver healthcare and government benefits is as inclusive and empowering as possible must be a priority.
Bruch, Sarah K., Myra Marx Ferree, and Joe Soss. "From policy to polity democracy, paternalism, and the incorporation of disadvantaged citizens." American Sociological Review 75.2 (2010): 205-226.
Campbell, Andrea Louise. How policies make citizens: Senior political activism and the American welfare state. Princeton University Press, 2003.
Lawless, Jennifer L., and Richard L. Fox. "Political participation of the urban poor." Social Problems 48.3 (2001): 362-385.
Michener, Jamila D. "People, Places, Power: Medicaid Concentration and Local Political Participation." Journal of Health Politics, Policy and Law(2017): 3940468.
Mettler, Suzanne, and Jeffrey M. Stonecash. "Government program usage and political voice." Social Science Quarterly 89.2 (2008): 273-293.
Mettler, Suzanne. "Bringing the state back in to civic engagement: Policy feedback effects of the GI Bill for World War II veterans." American Political Science Review 96.02 (2002): 351-365.
Mettler, Suzanne, and Joe Soss. "The consequences of public policy for democratic citizenship: Bridging policy studies and mass politics." Perspectives on politics 2.01 (2004): 55-73.
Soss, Joe. "Lessons of welfare: Policy design, political learning, and political action." American Political Science Review 93.02 (1999): 363-380.
Swartz, Teresa Toguchi, et al. "Welfare and citizenship: The effects of government assistance on young adults' civic participation." The Sociological Quarterly 50.4 (2009): 633-665