International comparisons of US health care are common but mostly focus on comparing its performance to peers or asking why the United States remains so far from universal coverage. Here the authors ask how other comparative research could shed light on the unusual politics and structure of US health care and how the US experience could bring more to international conversations about health care and the welfare state. After introducing the concept of casing—asking what the Affordable Care Act (ACA) might be a case of—the authors discuss different “casings” of the ACA: complex legislation, path dependency, demos-constraining institutions, deep social cleavages, segmentalism, or the persistence of the welfare state. Each of these pictures of the ACA has strong support in the US-focused literature. Each also cases the ACA as part of a different experience shared with other countries, with different implications for how to analyze it and what we can learn from it. The final section discusses the implications for selecting cases that might shed light on the US experience and that make the United States look less exceptional and more tractable as an object of research.
During debates on the Affordable Care Act (ACA), reforms in the legislation were frequently compared to health systems in other countries. On the pro-reform side, comparative data from health systems around the world were used in political debate to demonstrate how comparatively expensive yet mediocre US health care had become and to argue the case for reform (Peterson 2011). Among the ACA's opponents, images of “socialist” or authoritarian health care systems were used to invoke fear about the ACA and what it would mean for the future role of the federal government (Ehlke 2011). Commentators continued to talk about Canadian-, Singaporean-, French-, or Dutch-style health care or “single payer” or “universal health care,” without much context or analysis (Brown 2012; Marmor, Freeman, and Okma 2005). This tendency lives on in discussions about the ACA and ideas like “Medicare for all.” Phrases like single payer are still incorrectly used by the Left as shorthand for “good health care system” (Greer, Jarman, and Donnelly 2019; Sparer, Brown, and Jacobs, 2009).
It might be a good time, therefore, to revisit the question of how best to compare the US health care system to those in other countries. Ten years after the passage of the ACA, how should we make sense of the reforms from a comparative perspective? Can we make such comparisons truly meaningful when the United States seems to be, at least superficially, an “exceptional case”?
Understanding something as complex as the ACA, or US health care policy, in a comparative context is a significant endeavor that few have undertaken (Marmor and Klein 2012; Tuohy 1999, 2018; White 2013). But we can apply some thinking from comparative social inquiry to make the process easier. The following sections draw from literature on the comparative method in order to think about how we might define the ACA as a phenomenon and how we might compare it to other cases. A vast, cross-disciplinary body of literature in the social sciences examines different ways to define, select, and compare cases. Although this body of work is far too voluminous to examine extensively here, we provide some useful starting points for generating comparative insights that can help us to understand the ACA in a new light.
Casing as an Act of Research
One of the first things that comparative scholars will often do when working on a project is to try to construct a “case” by figuring out the boundaries of the phenomenon to be studied. Ragin (1992) describes the process of constructing a case, or “casing,” as a fundamental act of research. Researchers move back and forth between considering their case in the light of relevant theoretical ideas and empirical evidence in a process known as retroduction (Ragin 2011). This is far closer to the actual practice of research than much of the standard discussion of inductive and deductive research. In reality, most findings are not driven by the testing of a prior induction or the atheoretical induction of facts. They come from researchers who realize what is truly interesting about their topic as they move more deeply into both relevant theory and their case. Casing permits what we most want: the identification of portable causal arguments that can be convincingly used in some cases, as well as the scope of conditions under which they cannot be used (Falleti and Lynch 2009). When investigating complex social phenomena, the process of casing will almost certainly continue throughout the study, with researchers refining the boundaries of what their case is as their understanding grows.
With these insights in mind, how should we begin to “case” the ACA? After a decade of complex implementation, the ACA is much more than the law itself or the surrounding political debate. Put more specifically, what is the ACA a case of? What kinds of theory illuminate it, and what kinds of theory does it build? In the context of current debates in comparative politics, some promising ways to case the ACA include defining it as complex legislation, path dependency, demos-constraining institutions, deep social cleavages, segmentalism, persistence of the welfare state, or mandatory private insurance. These case definitions are by no means exhaustive. In a real study, researchers would change their conceptual understanding of the case throughout the project through the process of gathering and synthesizing information. Nevertheless, running through some examples is perhaps the best way to show how casing the ACA might work in practice. Each of these different casings highlights a dimension of US health care politics with international equivalents.
Uniquely Complex Legislation
For example, is the ACA a case of an American style of uniquely complex legislation? An increasing number of scholars have written about how public policy in the United States (especially health policy) is needlessly complex and builds in layers of interests at the expense of accountability, traceability, effectiveness, or efficiency. Whether we call it the delegated welfare state (Morgan and Campbell 2011), the submerged welfare state (Mettler 2011), the hidden welfare state (Howard 1999), or kludgeocracy (Teles 2012), the complexity of US social policy and its pernicious effects are a focus of much scholarship today (Greer 2018). The ACA is certainly a case of a very complex and indirect piece of legislation. Even politically and administratively simpler coverage mechanisms, such as extension of state insurance regulation, were complex and largely disguised from voters who might give politicians credit (Béland, Rocco, and Waddan 2016).
This intuition and extensive exploration of policy dynamics in the United States have not, however, been matched with efforts to see if US policy is indeed such an outlier in complexity, and why. US policy making is more elaborate than that of the Westminster systems, such as those in the United Kingdom or Canada. But are US policy processes really more complex and riddled with rent seekers than public policy making in France, or more obscure and ineffective than public policy processes in Italy? For that matter, is the relative simplicity of Westminster policymaking matched by relatively simpler policies?
One reason these authors give for the complexity of public policy is the fragmentation of the US political system, combined with its enormously dense interest group infrastructure, which means that interest groups can defend themselves against efforts to rationalize the system or effectively contain costs. Much of the complexity of the ACA comes from Democrats' decision to sidestep conflict by leaving most of the previous system in place (Oberlander 2016), which helped secure passage and make implementation possible at the price of extending the complexity of health policy. In a variant of this argument, Drutman (2015) persuasively argues that lobbyists, eager to demonstrate their return on investment, might also be making all policy more complex than it needs to be. This might help explain the lobby-filled US health care world, but it might also explain EU policy making, where legislation is just as complex, with just as many lobbyists eager to demonstrate the need for lobbying.
Most of the authors in the literature on complex public policy point to interest group activity as a key driver of the complexity they identify. But if the ACA, and US politics and public policy in general, are more complex than other countries with more waste, then maybe it is too simple to blame only interest groups. A powerful explanation for the complexity and partiality of US coverage politics can be found in path dependency, a concept with a particularly strong lineage in comparative politics research. The case is perhaps best put in the case of US health policy by Mayes (2004). Path dependency arguments emphasize the extent to which decisions at one time lock in costs and benefits and trigger positive- and negative-sum dynamics. Even if another approach would be better, the costs of transitioning to it are now so big that it is impossible, and we need not look to further explanations such as partisanship or interest groups in understanding why. In other words, once you have taken the wrong turn, turning around and going back hours later is so costly that you don't. Thus, for example, one might argue that once Medicare or employer-sponsored health care were established in policy, they created their own constituencies among elites and the public, became harder to reform, and jointly became an obstacle to a coverage reform. The evidence for such a thesis is not hard to find (e.g., Campbell 2003; Gottschalk 1999; Klein 2003; Mettler and Soss 2004; Oberlander 2003; Quadagno 2005; Tuohy 1999).
Path dependency arguments are powerful in explaining that divergence sometimes just is. At the same time, path dependency arguments also have some weaknesses that comparative politics scholars have been trying to address (Mahoney and Thelen 2015). First, while they produce an interesting search for critical junctures in history, they do less explanatory work now—each country ends up on its own path. Second, this search for critical junctures can turn into a swamp (Brown 2010): At what point, exactly, did the United States embark irrevocably on the path leading to today? Was any particular trajectory a necessary result of a given event? Third, there is no guarantee that any given policy triggers a self-reinforcing feedback loop (Oberlander and Weaver 2015). Those opposed to the policy and its effects might be able to mobilize against it and make countervailing policies that just lead to stalemate or to policy cycling (Tuohy 2018).
If the ACA is more, or differently, kludgeocratic, delegated, and submerged than the policy of other countries—it is certainly less universal—and if the difference is not just due to history, then is the outcome because of the number and strength of demos-constraining institutions, which explicitly constrain democracy, in the US political system (Stepan and Linz 2011)? Demos-constraining institutions can work in two ways. One is through federal-level veto points (e.g., the Supreme Court, the independence of two legislative branches and the executive, or the filibuster). The other is through fragmentation (e.g., dividing the people into multiple jurisdictions, such as 50 states and about 90,000 local governments), which promotes intergovernmental competition and raises the costs of enacting policy by demanding enactment in more jurisdictions. Medicaid expansion in the ACA showed them working together: one veto player, the Supreme Court, rewrote the law to make expansion optional for states, which created 50 new venues in which to contest the policy. American demos-constraining institutions are both strikingly numerous by international standards and, in many cases, particularly strong, as with the absurdly malapportioned Senate, its filibuster, or the Electoral College (Stepan and Linz 2011). It thus should be no surprise that the United States has such a problem of political, social, and economic inequality: its institutions impede the most effective action against such problems (Stepan and Linz 2011).
The United States is certainly poor at delivering policy outcomes that reflect the overall preferences of its population. Page and Gilens (2017) contentiously argued that there was no relationship between the percentage of the American public who support a policy and its likelihood of passage. There is no clear comparative literature that tells us whether the United States is more or less responsive than other countries. The ACA in this perspective, though, is not particularly ambitious legislation compared to universal health care, but it is something like the best that could be got through American political institutions as they currently stand (Tuohy 2018). This account is highly plausible and fits with studies of other complex and demos-constraining political environments, such as Brazil (Arretche 2013; Segatto and Béland 2019) and the European Union (Héritier 1999; Obinger, Leibfried, and Castles 2005). In those polities, any kind of policy success depends on creativity and clever bypasses of demos-constraining institutions—a phenomenon that Americanists recognize as “unorthodox lawmaking” (Sinclair 2016).
Deeper Social Cleavages
Of course, these demos-constraining institutions, from the beginning, have been sustained by those who do not benefit from more democracy, including slaveholders and the beneficiaries of Jim Crow, as well as other interests that would not benefit from the passage of popular redistributive policies. So perhaps both the institutions and the policies reflect deeper social cleavages, notably those of race and perhaps also of class. This perspective flips the preceding one, as when Riker (1964: 155) said of state's rights that “if one disapproves of racism, one should disapprove of federalism.” If American institutions developed as a long series of compromises with antidemocratic or undemocratic forces that are still at work today, should we instead focus on those forces and view institutions as endogenous to that explanation?
Marx (1998), for example, shed considerable light on the United States by comparing it to two other countries with similar legacies of white supremacy: Brazil and South Africa. Lieberman (2009) likewise demonstrated the negative effects of racial hierarchies and politics on AIDS policies with an argument that could well apply to the US. There certainly is persuasive evidence that the ACA was “racialized” in the eyes of many whites (Grogan and Park 2017; Tesler 2011) and that this is part of a larger rise of white identity and nationalism (Maxwell and Shields 2019; Morone 2018). From this perspective, the United States is in a catch-22: racism that manifests in inherited institutions, in public opinion, and in organized political forces impedes redistribution even when it would be in most whites' interests, but at the same time it is a redistributive welfare state that enables progress toward racial equity (Katznelson 2005; Lieberman 1998).
Through that lens, the ACA, while it extends coverage and promotes equity in a variety of ways (Grogan 2017), is still only a partial coverage extension by the standards of other rich countries, and it came with a furious and racialized backlash. Perhaps it could be read as just another episode in intertwined battles over race and redistribution, or another case of the extension of universal social programs that follows on a democratic advance. The backlash against the ACA, which cost Democrats congressional seats in 2010 (Nyhan et al. 2012; Saldin 2011), could in this casing be read as another one of the backlashes that follow on civil rights victories. Pursuing the implications of all the work identifying race as a key variable in America health politics drives the casing process, and comparisons, in new comparative directions—toward the other societies marked by industrial chattel slavery and racial hierarchy, which are found principally in the New World, rather than long established states in a European continent marked by centuries of state and nationalist efforts to create homogeneity. It equally, though, raises complex questions of causality, since institutions are not perfect reflections of today's politics, and in public opinion beliefs about race work through complex ideas of “deservingness” (Lynch and Gollust 2010).
Another, related way to view the extension of health care access in the United States is as part of a push, led by the Left, to reduce segmentalism (sometimes also known as stratification) in health care policy. In this case, Latin American or some East Asian countries, rather than European ones, might be the most useful comparators (Haggard and Kaufman 2008). Latin America and southern Europe have histories of “segmentalist” systems in which particular groups have different health care access, with some groups, such as the workers of the most favored companies, enjoying good benefits and some others frozen out entirely or dependent on a thin public safety net (Martínez Franzoni and Sánchez-Ancochea 2018). By such a definition, the United States is, like Chile, Brazil, or Mexico, a country whose politics are shaped by a segmentalist legacy and where the debate is about whether and how to universalize its health services (Greer, Jarman, and Donnelly 2019; Greer and Méndez 2015).
Persistence of the Welfare State
But despite being flawed and hard to claim credit for, and being caught in a backlash of tremendous force, the ACA persists. Is the ACA best viewed as a case of the persistence of the welfare state? Since the 1980s, scholars of comparative welfare states found not so much a rollback as a slowing of the extension of the welfare state in what is known as the “new politics” of welfare (Pierson 1996). The basic logic is that in any system it is hard to take away benefits that people currently enjoy (Brooks and Manza 2007). Veto points in the institutions will make it harder to shrink the welfare by empowering both interests that oppose cuts and interests that block change or extension. If the program does lose its effectiveness, it probably will not be through big legislative changes. Rather, it will suffer “policy drift,” in which failure to update a policy means that over time it ceases to fulfill, and perhaps even undermines, its original goals (Hacker 2004). Of course, the argument that politicians will avoid overt retrenchment is a poor fit with the actual experience of the ACA, which was almost repealed despite obvious political and policy reasons not to (Hacker and Pierson 2018). America's numerous veto points just barely saved the ACA, but from a partisan assault whose magnitude most theorists of new politics would hardly have expected.
Comparing the ACA to Other Cases
The next step in applying the comparative method is to select further cases for comparison. Once researchers have at least some preliminary understanding of what the ACA is a case of, they can begin to think about it not just as a case in isolation but as part of a constellation of cases that can be grouped into sets or categorized in different ways. In other words, how does the ACA, as a case of X, relate to other cases of X? What dimensions of the US experience allow us to create meaningful comparisons that contribute to our international as well as US expertise?
Again, selecting cases is an act of research. Case selection is a preoccupation of literature in comparative politics because it is central to the credibility of the findings, shapes the kinds of lessons that might be learned, and above all, determines the scope conditions for any generalization. Done badly, comparative analysis can degenerate into simple impressionistic analogy. Efforts to develop an approach to case selection based on frequentist statistical methods (King, Keohane, and Verba 1994) also tend to fail (Brady and Collier 2010) because treating qualitative comparison simply as a way to address “small n problems” does not exploit the richness of a case as a set of interacting variables. More productive efforts to develop case selection have focused on just that richness: identifying subtler ways that variables interact within cases over time in order to tease out configurations of variables and their effects.
Attention to case selection equally allows us to identify less valid inferences. For example, there are probably too many UK-US comparisons, since a shared language and an English tendency to borrow fashionable American phrases like population health and accountable care organizations can distract from the fact that in important respects they are very different polities. The highly centralized, executive-focused United Kingdom, the size of the public National Health Service systems, and the strong concept of positive social citizenship among UK citizens all make it more of a contrast than a suitable comparison with the United States. It is highly unlikely, for example, that endless debates about putative “Americanization” of the British health system have brought much clarity to UK policy conversations (Powell, Béland, and Waddan 2018).
Each of the six “casings” of the ACA we identify above points to a different set of comparisons that would shed light on the US experience and future, while contributing to an international conversation. If the question is just how distinctively complex US legislation is and why, then cases from the Netherlands, Switzerland, and Italy can teach us about the politics of complexity. If the question is about how path dependency works, then identifying similar cases at key junctures can be illuminating, for example, Canada in the 1970s (Maioni 1998). If the question is how demos-constraining institutions shape the United States and how they may be worked around, then the veto-ridden European Union is a case, as well as other wealthy states such as Switzerland, where policy making is built around anticipation and management of vetoes. If the question is how to understand deep social cleavages, particularly the changing racial politics of health in divided societies, then a more delicate operation is needed to seek lessons from quite distant countries, such as Brazil or South Africa, where racial inequalities and politics are important to any distributive question. If the question is how to understand segmentalist legacies, adding an international dimension to the extensive literature on path-dependent and interest-group-reinforced segmentalism in the United States, then relevant cases would be countries that substantially overcame segmentalism (e.g., Spain and Brazil; Linos 2013) and ones that failed to do so, as well as the hybrids that have evolved (Wong 2004).
Once we “case” the ACA, then we can see it in a new light and see what lessons can be exchanged with the rest of the world. It might not be comfortable to compare the United States to Brazil or South Africa, or an easy research project to figure out Japanese health care financing and politics in order to shed light on the experience of others (Schoppa 2006; White 2013), but those projects can be revealing if done with rigorous attention to case selection and inference.
Comparisons of the US health system are often made with an overt agenda, typically of showing its extraordinary expense, inequity, and mediocre results (Cohn 2009; Reid 2010; Schneider et al. 2017). While understanding these differences may be important, viewing the United States as an exceptional case, or just a poor performer, constrains our thinking about both health problems and potential solutions.
To the extent that the United States is a case of a system with too many veto players, we can expect partial and fragmentary legislation, but equally, the experiences of other countries show that complexity and indirection of the ACA and much US policy might not be inevitable. The United States stands out relative to European countries for the depth and historical importance of its domestic racial cleavage and its segmentalism, but those phenomena are also ones that many Latin American and southern European countries face and are addressing in their politics. In other words, if the United States is cased carefully, there is scope to introduce new thinking about the ACA and the rest of the world into our health policy research.