Laboratories of What?

American Federalism and the Politics of Democratic Subversion

— Based on a paper given at a conference on Democratic Resilience, Cornell University, Nov. 2019

People lined up to vote outside Riverside High School in Milwaukee on Tuesday. Credit: Milwaukee Journal-Sentinel, via Reuters

From time to time public attention is focused on scandalous situations in state government…These moments pass; state affairs recover their wonted obscurity and it is assumed that the wrongdoers have been exposed and punished.

Grant McConnell (1966)

The 2016 presidential election brought on a blizzard of foreboding announcements about American democracy. Yet as political scientists and pundits alike turned their gaze towards the spectacle of Trump’s Washington, fewer seemed as concerned about what was happening in places like Raleigh or Jefferson City. In fact, scholars and commentators troubled by abuses of power in the executive branch pointed to federalism as — in Corey Brettschneider’s words––“the most effective tool” for protecting democracy, “especially if other constitutional checks fail.” States, it was argued, provided crucial venues for dissent and the formation of alternative governing coalitions. And while new analyses of democratic backsliding mentioned gerrymandering in state legislatures and state-level episodes of ‘constitutional hardball’, they tended to focus their attention on the national level. If crises of democracy are like diseases, one might have concluded that the states were — with a small number of exceptions — immune.

Such benign images of the states — what V.O. Key once called the “pleasant fictions of American political mythology” — are nothing new. That the most commonly applied sobriquet for the states is “the laboratories of democracy” is evidence enough. Tellingly, mass-level trust in state and local governments has remained remarkably resilient, even as trust in national institutions has cratered. To some extent, empirical research confirms the states’ role as democratic laboratories. Structurally speaking, state constitutions contain numerous positive rights that are not to be found at the national level. Nearly half allow for some form of direct democracy, enhancing opportunities for mass influence. There is even a venerable line of research suggesting that state governments are, on balance, responsive to mass opinion.

Yet the notion that state governments are a source of democratic resilience in the US belies a good deal of historical and contemporary evidence. Consider the following contrast. In cross-national indices of democracy, the US has improved markedly over the last sixty years. Figure 1 compares the US to seven older democracies with federal systems on the Varieties of Democracy (V-Dem) Institute’s Electoral Democracy Index. Between 1958 and 2018, the US shifts from being a low outlier on the index to achieving near parity with older federal democracies. Yet when it comes to subnational democracy, the same indices reveal another pattern altogether (Figure 2). Indeed, while democracy in the states has no doubt improved over the last sixty years, it remains substantially more uneven than in peer countries. Not only do cross-sectional analyses reveal significant differences among states in opinion-policy congruence, the character of state electoral and representative institutions diverges in meaningful ways. States that make it easier for voters to participate in direct democracy tend to produce a greater amount of collective goods and tend to feature greater levels of political equality. In other states, however, the manipulation of electoral and legislative rules curtails basic democratic rights.

Figure 1. Electoral Democracy Index Scores for US and Older Federal Democracies, 1958 and 2018. [Source: Author’s analysis based on data from Coppedge, M., Gerring, J., Knutsen, C.H., et al. (2019). V-Dem Country-Year Dataset v9, Varieties of Democracy (V-Dem) Project.; Scores are based on a weighted average of five component indices measuring freedom of association, clean elections, freedom of expression, elected officials, and suffrage and the five-way multiplicative interaction between those indices.]
Figure 2. Subnational Election Evenness in US and Older Federal Democracies, 1958 and 2018 [Source: see Figure 1. Scores are based on expert ratings of countries using an ordinal measure ranging from 0–2.]

There is an urgent need to re-evaluate the relationship between federalism and democratic resilience in the United States. Inequality in democratic rights and liberties was itself baked into the Constitution. These inequalities have been dismissed, however, because scholars tend to conceptualize states as relatively isolated laboratories of democracy. Yet the Constitution makes the states into the infrastructure of democracy: state laws shape the exercise of rights and liberties, structure national elections and legislative districts, and affect the development of civil-society institutions in ways that have national reach. Indeed, episodes of democratic collapse at the state level have had profound reverberations for national politics.

To understand the endurance of uneven subnational democracy, it is worth remembering––as Rob Mickey’s work has shown––that subnational authoritarianism persisted in the US South until the 1960s, yielding only after the nationalization of political conflict and protracted interventions by central state authorities. Yet, territorial democratization was not self-enforcing. Racial apartheid and localized economic oligarchies have had enduring effects on state politics. More importantly, soon after the democratic transition, conservative political elites countermobilized, and began to chip away at the new rights regime. State parties motivated by counter-majoritarian ideological agendas have, among other things, violated informal governing norms, actively limited ballot-box access through enacting new administrative burdens and have employed increasingly sophisticated approaches to gerrymandering to lock in seat shares.

The erosion of subnational democracy after the consolidation of the 1960s has depended upon efforts to localize and privatize conflict — what Adam Przeworski calls “subversion by stealth.” By contrast, strengthening the states as part of the ‘infrastructure of democracy’ will require the re-nationalization of contests over the boundaries of political incorporation. In doing so, reformers will have to think carefully about how to enhance the legibility of democratic deficits in a highly polarized context. They will also be forced to deal with a legislative process prone to gridlock and a judiciary that is increasingly hostile to democracy-enhancing reforms.

The language scholars use to describe American state governments may contribute to the shock we experience when democratic norm violations occur at the ballot box or on the legislative floor. One source of confusion here is the phrase “laboratories of democracy,” famously coined by Louis Brandeis in 1937, but which did not gain wide traction until the 1980s. (Ngram data are illustrative). This metaphor refers to states’ capacity to learn from one another’s policy successes (and, less frequently, their mistakes). It thereby emphasizes experimentation itself as a democratic good. The tacit assumption here is that subnational governments have the capacity to resist undemocratic maneuvers and to adopt and promote democratizing reforms when Congress is resistant to them. Further, it is assumed that reforms which retrench democracy in a single state can be costlessly observed by others, rejected, and thereby contained at worst contained to a smaller territorial area. Those subject to these reforms can also hypothetically exit the regime by “voting with their feet.”

The laboratory metaphor, as Susan Rose-Ackerman and David Super have noted, fails to account for structural limitations’ on states’ capacity as innovators and sites of representation. Yet the problem goes deeper: conceptualizing the states as relatively isolated laboratories fundamentally ignores the multiple roles they play in co-constituting the national polity. First, and perhaps most importantly, states construct the national electorate through their administration of federal elections. Notwithstanding the nationalization of voting rights in the 1960s and increasing judicial scrutiny of state election practices, state election practices have continued to vary; following the Supreme Court’s elimination of the Voting Rights Act preclearance formula in Shelby County v. Holder, states’ discretion over election laws expanded yet further. The effects of election laws do not stop at the state line, however. Racial apartheid in the South constructed a “Jim Crow Congress”; insulated from electoral competition, Southern committee chairs became the fulcrum of national policymaking — foreclosing the New Deal’s social democratic aspirations. Even after the dismantling of Jim Crow laws, Southern states continue to exhibit lower levels of voter turnout in national elections, and lower levels of responsiveness to — reforms intended to reduce the cost of voting. Regardless of geography, enduring unevenness in electoral democracy continues to shape national politics. The disfranchisement of felons significantly shaped the outcome of the 2000 presidential election as well as several Senate elections in the latter half of the twentieth century. While the effect of photo ID laws on turnout is much disputed, Francesco Esposito and his colleagues have found that they clearly depress voter registration in presidential election years by as much as 7.6 percentage points. Setting aside any concerns about voter suppression, state voter registration requirements, when implemented in municipalities with weak administrative capacity, can — as Barry Burden and Lauren Neiheisel’s work shows––lower turnout in national elections by as much as two percentage points. In short, an assessment of US electoral democracy would be incomplete without attending to unevenness in state election laws and practices.

Just as state election laws re-constitute the national electorate, states decennially re-constitute Congress through redistricting. States’ discretion over this process has diminished since the so-called judicial “reapportionment revolution.” Even so, state legislative redistricting continues to vary in ways that are consequential for national governance. Partisan gerrymandering skews the ideological distribution of state legislatures and the policy outputs of state government. Republican gerrymanders, as Devin Cauhgey and his collaborators have shown––advantage conservative politicians and yield higher levels of conservative policy output. As such, gerrymandering invariably affects federal policy, especially given the substantial role states play in administering federal grants (which make up over 30 percent of states’ annual budgets) and implementing federal regulations. While it was not the only factor impeding states’ expansion of Medicaid, aggressive gerrymanders in states like North Carolina and Wisconsin helped to undermine state legislators’ incentives to do so, even though expansion is supported by majorities of voters in both states. To be sure, partisan control of redistricting has a smaller effect on seat shares in Congress than it did before the Supreme Court’s decisions in Baker v. Carr (1962) and Wesberry v. Sanders (1964). Nevertheless, partisan gerrymandering has significant downstream effects on the integrity of political parties. In districts where gerrymandering disadvantages their party, prospective congressional candidates are far less likely to challenge incumbents. As Nick Stephanapoulos and Chris Warshaw have suggested, those who do run in congressional districts biased against their party also have significantly weaker resumes and receive significantly fewer campaign contributions than their peers in unbiased districts.

Finally, governance in the fifty states plays a formative role in structuring both the public and private spheres. Prior to the judicial construction of what William Leuchtenberg calls the “Second Bill of Rights” in the 1960s, states’ used their police powers in ways that limited the exercise of individual rights. Perhaps most infamously, states exerted the power — with the Supreme Court’s blessing — to sterilize tens of thousands of “undesirable” people, including people of color, immigrants, unwed mothers, the poor, the disabled, and the mentally ill. On the other hand, as Emily Zackin suggests, positive rights contained in state constitutions played a pivotal role in creating civil society through the guarantee of free primary and secondary education. Even as the federal government gained greater control over other areas of public provision, states have retained substantial leverage in the sphere of education. State choices on education policy, in turn, construct citizens’ participatory behaviors and views of government (I find the work of Sarah Bruch and Joe Soss particularly persuasive on this point). Education is not the only such ‘formative’ institution shaped by the states. As Jamila Michener’s research shows, citizens’ experience with state Medicaid programs has a sizable impact on their propensity to participate in politics. Citizens in states making the largest reductions in Medicaid benefits were between four and nine percentage points less likely to vote, register, or participate when compared to those in states that did not reduce their benefits.

To sum up, we cannot treat democratic deficits in the states as mere variation in a dataset where N equals 50. Rather, the states compose the national polity through the administration of national elections, the construction of Congress, the implementation of federal policies, and the structuring of public and private spheres. Assessing democracy in the states thus helps to clarify the analysis of democratic resilience in the United States as a single case.

While the quality of subnational democracy has ramifications for the national polity, few studies treat democracy in the states as a subject of study in its own right, as Carol Weissert notes in a probing review of the literature. Rather than comparative assessments of regime type in, say, Wisconsin and Minnesota, political scientists have produced a great mass of empirical studies comparing party competitiveness, election administration, the effects of state policy on civic participation, and opinion-policy congruence. While voluminous, the fragmentation of this literature impedes our understanding of subnational democracy in the United States. This is especially important if we want to assess the endurance of the democratic transition ushered in by the Second Reconstruction.

We can begin to evaluate the unevenness in subnational democracy by considering how states vary on three basic criteria. First, do states extend equal voting rights to all adult citizens and do their votes have roughly equal weight? Second, do the vast majority of citizens in a state participate in elections? Third, is there effective competition among organized political parties? According to Kim Quaile Hill, when analyzed with these criteria, only fifteen states qualified as at least modestly “polyarchic” in the middle of the twentieth century. By contrast, eleven states — exclusively in the South — constitute closed or relatively closed party oligarchies. While state-level democracy persisted in the South even after the end of Reconstruction, it eroded as the result of racial violence, election fraud, and ostensibly race-neutral laws which disenfranchised the black population. In the years that followed, state governments in the South existed as one-party authoritarian enclaves — denying the right to vote to African Americans and poor whites, brutally repressing dissent, and monopolizing control over the political arena. Even when we look outside the South, the quality of democracy is mixed. Prior to the reapportionment revolution, incumbent parties in many state legislatures also strategically leveraged malapportionment to constrain their political opponents, making the state legislature — as V.O. Key put it––“ill equipped” to serve the “basic function of dissent.”

By the 1980s, the passage and reauthorization of the Voting Rights Act had virtually eliminated the de jure forms of vote suppression that had existed under Jim Crow. Yet pockets of resistance remained, and systematic efforts to erect “second generation” barriers to voting rights emerged in a number of states. Registration systems in all but one state continued to pose barriers to voting. State scores on an index measuring the cost of voting, developed by Quan Li and colleagues, increased on average by 20 percent between 1996 and 2012. Malapportionment disappeared, yet new technologies permitted more sophisticated forms of partisan gerrymandering. Following the 2010 Census, fourteen states severely gerrymandered their congressional districts and six did the same to their state assemblies.

Second, while the Voting Rights Act helped to correct major racial disparities in voter registration and turnout, substantial variation remained. Between 1980 and 1986, the range in state turnout rates compressed, but it did not improve across the fifty states. Moreover, four states (Georgia, South Carolina, Texas, and Virginia), had average voter-turnout rates below one third of the voting-age population. By 2018, the floor for turnout had risen from 28 to 36 percent, yet the ceiling fell slightly from 65 to 63 percent. In other words, the greatest improvement in turnout rates came from those that had historically performed poorly. As David Bateman has found, however, turnout did not markedly improve among “middle of the pack” states, many of which began implementing tighter voter restrictions during this period.

Third, the democratization of the South brought with it the end of closed party oligarchies in the states. Even so, party competitiveness — as measured by the Ranney Index — remains limited in many states. Whereas Democrats dominated the largest share of states in the 1980s, between 2012 and 2018, nearly two-thirds of the states were dominated by Republicans. By contrast, only 17 states could be classified as competitive two-party systems. Indeed, while the end of party oligarchies opened up the possibility of two-party competition in the South, one-party Republican dominance soon emerged there due to several factors, including the increasing ideological alignment between the Republican Party and conservative Southern voters, the fact that loyal Southern Democratic voters have gradually aged out of the electorate, and incumbency advantages enjoyed by Republicans in majority-white jurisdictions.

Broadening our conceptualization of democracy beyond these political characteristics reveals additional evidence of unevenness. Three examples will suffice. First, the organizational capacity of nonelite actors to make demands on government is highly inconsistent across states. In 2018, the percent of wage and salary workers with union representation varied from 3.6 percent in South Carolina to 24 percent in Hawaii. As Laura Bucci has shown, the rise of right-to-work legislation in the states over the latter half of the twentieth century has significantly weakened the working class’s capacity to effectively represent the interests of nonelite Americans.

Second, states vary considerably in their protection of basic civil rights. North Dakota, which has recently seen a rash of recent hate crimes, provides only limited recourse for victims. In recent years, the Kansas legislature has violated the state constitution by failing to provide “reasonably equal access to substantially similar equal educational opportunity” (Gannon v. State, 2014). Perhaps most importantly, as Amy Lerman and Vesla Weaver show, state policies across the country have injured democratic citizenship by constructing an uneven, but ever-widening system of mass incarceration. In 2017, the state-prison incarceration rate (per 100,000 people) ranged from 134 in Massachusetts to 720 in Louisiana. This maps onto broader variation in the punitiveness of state-level policies on sentencing, parole, and juvenile justice.

Third, as argued by Suzanne Mettler in her classic Dividing Citizens, affecting social rights have also been historically uneven. Variation among states in per-person Medicaid spending, for example, reflects stark differences in states’ philosophies about toward income-transfer programs. Thin Medicaid benefit packages, in turn, have a negative effect on low-income citizens’ political participation. Because states cannot engage in countercyclical spending, economic downturns can also yield retrenchment in state commitments to social rights. Since the Great Recession, spending on higher education has fallen in nearly every state. In eight states, for example, including Alabama, Arizona, Pennsylvania, and South Carolina, funding for higher education fell by more than 30 percent between 2008 and 2017. Over the same period, tuition at public colleges increased by 35 percent.

The evidence here points to persistent inequalities in state-level democracy, even in the wake of a significant democratic transition. To be sure, the closed party oligarchies of the past have disappeared. Yet the foundations of democratic rule — free and fair elections, competitive parties, and institutional support for civil and social rights — are not even across the fifty states. Such territorial variation is perhaps to be expected in a large, diverse polity. If anything, however, this suggests that the quality of state-level democracy is understudied. This is especially troublesome, given both the historical roots of the variation as well as present trends that may undermine subnational democracy.

How should we understand the endurance of unevenness in democratic rule following the belated consolidation of American democracy? Several factors seem important. First, the legal instruments of that consolidation have been subject to gradual institutional erosion. Consider the Voting Rights Act (VRA), arguably the keystone of what Edward Gibson calls the ‘territorial regime’ supporting subnational democracy. Its powerful preclearance provisions, routinely reauthorized and expanded by wide bipartisan majorities gave it the reputation for being the most effective civil rights law ever enacted. Yet as Jesse Rhodes notes, the VRA’s solid majorities in Congress concealed a sustained antipathy to the law on the political right, and enduring partisan polarization on the question of racial justice. Between 1982 and 2006, the Department of Justice blocked over 700 attempts by VRA-covered jurisdictions to enact discriminatory election rules. Core coalition partners within the Republican Party — including representatives of organized business who feared the potential redistributive implications of minority political incorporation — turned to administrative and legal venues to contest the law soon after the VRA’s enactment. By the 1980s, the Reagan administration helped to slow action on preclearance enforcement at the Department of Justice. Conservative judicial nominees advocated for strong limitations on majority-minority redistricting and application of preclearance provisions. By the early 2000s, these efforts had grown more aggressive. At the same time, after a decade of movement towards less restrictive voter-registration procedures, states began to experiment with new forms of voter restriction, including limits on early voting, tightening registration requirements, voter-identification rules, and greater barriers to felon re-enfranchisement. The Supreme Court’s 5–4 decision to strike the VRA’s preclearance formula in Shelby County v. Holder thus represented a culmination of a decades-long series of maneuvers to dismantle the territorial regime that emerged following the Second Reconstruction. After it came a deluge. Twenty-eight states adopted new restrictions on voter registration, including identification requirements, purges of voter rolls, cuts to early voting, and the elimination of polling locations.

De-consolidating democracy required the displacement of debates over voting rights from high-salience venues like Congress to arenas where public attention would be minimal: the bureaucracy and the courts. It also demanded a legal means to weaken federal agencies’ leverage over the conduct of local elections — further parochializing conflict. Indeed, the recipe for what Przeworski calls “subversion by stealth” need not be unlawful. Rather, subversive maneuvers can include legal (or legally ambiguous) manipulations of rules and norms that diminish access to elections, rights, or accountability.

Yet if subversion by stealth is the means of deconstructing subnational democracy, where is the motive? To find the answer, we have to consider the shifting incentives of state-level politicians. Not all elected officials pursue strategies of boundary control. Rather, what matters is whether politicians can achieve their desired ends through electoral politics. After they are elected, officials intensely committed to an unpopular ideological goal face a stark choice. They must decide whether or not to use the power of the state to entrench themselves in office and weaken their opponents’ opportunities for electoral and policy success.

It bears mentioning, then, that the period following the democratic transition witnessed a major shift in the ideological landscape of American politics. With the decline of the one-party South, the centralization of government, and the decline of local media landscapes, state governments became increasingly ideologically aligned with national, polarized parties .The recent rise of undemocratic maneuvers at the state level is at least in part attributable to asymmetric polarization in the states. Consider restrictions on voting. Whereas many states expanded access to the ballot in the 1990s and early 2000s via voting by mail and early voting, a growing number of states have passed laws increasing the cost of voting through tightening voter registration deadlines, rolling back early voting, limiting the number of polling stations, and reducing polling hours. Voter restrictions are the subject of intensely polarized battles. Between 2006 and 2011, states with an unencumbered Republican majority in the state legislature were significantly more likely to enact voter restrictions than states where Democrats had control. Yet when it comes to voting rights, partisanship intersects with race: legislatures were significantly more likely to restrict voting in states with larger black populations or where minority turnout had increased sharply in the 2008 presidential election. In general, new restrictions on voting have a depressing effect on voter turnout. When North Carolina enacted a strong photo ID requirement for voting, it resulted in a 2.6 percent decrease in turnout for voters without an ID in the general election. As Justin Grimmer and Jesse Yoder show, this deterrent effect persisted even after the law was suspended, likely because the law created confusion and because voters lacked information about changing requirements for voting. Countermobilization may help to minimize the impact of these laws on turnout. Even so, administrative barriers to voting are, in Ben Highton’s words, undeniably “real, nontrivial, and unequal in impact.”

Next, consider the rise of partisan gerrymandering in the 1990s and 2000s. Here, it is not merely the asymmetric polarization of the parties that matters, but also Republicans’ increasingly sophisticated efforts to coordinate redistricting plans across state legislatures. In 2010 the Republican State Leadership Committee launched REDMAP (short for Redistricting Majority Project) to invest in swing-state races with the explicit aim of controlling redistricting in swing states like Wisconsin and North Carolina. The availability of sophisticated mapping software like Maptitude enabled Republicans to redraw districts to lock in their electoral gains in the 2010 midterms. By 2012, Wisconsin’s gerrymander had ensured that, while Democrats won 53% of the votes for State Assembly, they won only 39% of assembly seats. This case is emblematic of a broader asymmetric pattern in gerrymandering. Since 1972, unified Democratic control of state government has had a consistent effect on one prominent measure of gerrymandering, the efficiency gap. By contrast, as Nick Stephanapoulos has found, the treatment effect of unified Republican control on the efficiency gap in state legislatures nearly doubled between the 1972–1990 period and the years between 1992 and 2014. In other words, both parties gerrymander, but in recent years Republicans have engaged in a far more intense form of gerrymandering than Democrats.

Party polarization, as Jake Grumbach has pointed out, also intersects with another trend that threatens democracy in the states: the infiltration of ideological networks into party organizations. Over the latter half of the twentieth century, successive waves of reform and organizational change gradually hollowed out local party organizations, disconnecting party actors from the work of candidate selection and policy development. Taking their place has been a looser assemblage of intense policy demanders, who leverage party apparatuses to pursue their interests. Within the Republican Party, conservative network organizations such as the American Legislative Exchange Council (ALEC) and the State Policy Network (SPN) organized to advance an agenda of upwardly redistributive policies that lack a mass base of support, as Alex Hertel-Fernandez’s work has shown. Understandably, then, these organizations have also promoted policies that weaken mass participation and civil-society infrastructure. Among other model legislation promoted by ALEC are bills that restrict ballot-box access through imposing administrative burdens and limit the use of ballot initiatives on issues like the minimum wage. ALEC has also promoted right-to-work laws, which weaken the organizational capacity of labor unions, as well as preemption laws which bar local governments from crafting worker protections. Other model bills have the capacity to undermine civic engagement through privatizing public education systems.

When ideologically extreme parties experience costly electoral threats, “constitutional hardball” is never far behind. 48 hours after Roy Cooper (D)’s victory in the 2016 North Carolina gubernatorial election, Republicans introduced a package of legislation that, among other things, restricted Cooper’s ability to make cabinet appointments, cut the size of the executive branch by 275 percent, and gave Republicans control over the state Board of Elections during election years. The legislation was soon signed by outgoing governor Pat McCrory (R). Two years later, Republicans in Wisconsin introduced a similar package of legislation, which cabined the authority of governor-elect Tony Evers (D) over key state commissions, the state’s Medicaid and nutritional assistance programs. Signed into law by outgoing governor Scott Walker (R), the package also restricted the number of days allotted for early voting. Both McCrory and Walker’s ideological commitments mattered here. In Michigan, outgoing governor Rick Snyder (R) — who had frequently disagreed with hard-right members of his own party — vetoed similar legislation, introduced after the election of Gretchen Whitmer (D) to the office in 2018.

As the Michigan example shows, institutional veto points are a conditional bulwark against democratic erosion. The courts provide another illustration. Despite numerous district court rulings overturning partisan gerrymanders, the Supreme Court’s 5–4 decision in Rucho v. Common Cause (2019) effectively deemed the subject non-justiciable, closing the window of opportunity offered by Anthony Kennedy’s concurrence in Vieth v. Jubelirer (2006). A similar pattern has played out on the issue of partisan power grabs. In an ideologically divided 4–3 decision, Wisconsin’s Supreme Court upheld lame-duck legislation passed in 2018, ruling that the state’s Constitution “affords the Legislature absolute discretion to determine the rules of its own proceedings” (League of Women Voters v. Evers, 2019). Less than a year later, a federal district-court ruling effectively punted on the Wisconsin legislation, suggesting that it lacked jurisdiction to “police the boundaries” between the branches “in the absence of a concrete and particularized harm and the violation of a federal constitutional right” (Democratic Party of Wisconsin v. Vos, 2019).

Judicial deference on issues of subnational democracy highlights a broader epistemic challenge in strengthening subnational democracy. If “subversion by stealth” is the most common form of democratic erosion, as Adam Przeworski suggests, policing it requires a great deal of societal consensus on when so-called “bright line” rules have been violated. Yet given that partisanship strongly colors public perceptions of such violations, the normative path towards democratic resilience seems to me to be a fraught one.

Subnational democratization is best described as a slow-moving institutional process. It is slow moving, in part, because the U.S. Constitution imposes rigid barriers to changing the territorial regime. These structural barriers also enable local elites who wish to preserve countermajoritarian arrangements to engage more easily in strategies of boundary control. This includes designing electoral and legislative structures to preempt competition and monopolizing links to the national polity to limit outside influence in territorial affairs. In a nationally democratic polity, efforts at boundary control are often designed to fly under the radar, making their identification and eradication all the more difficult.

When democracy in the states is threatened, there are no quick fixes or work-arounds, no substitutes for the hard work of political organizing and coalition building. Indeed, piecemeal reformism has created problems of its own. Because subnational incumbents often work to parochialize conflict, addressing democratic deficits in the states demands the nationalization of conflict. This can occur, as Ed Gibson notes, through either party-led transitions, in which national leaders invest state-level resources to defeat incumbents, or center-led transitions, in which national authorities intervene to transform the rules of the game. Of course, nationalizing conflict comes with challenges of its own. If the federal judiciary is unwilling to wade into the “political thicket”, legislative or executive action will necessary. The Shelby majority, for example, framed its decision as an invitation for Congress to rewrite Section 4(b) of the Voting Rights Act (VRA). Yet in part because political conflict over the rights of citizenship is already so nationalized, efforts to restore the Voting Rights Act — let alone enhance it — have faced stiff political opposition.

Members of every Congress since Shelby have introduced standalone legislation to restore the preclearance formula, updating it to reflect new forms of voter suppression, including photo ID laws. Yet, with only a modicum of Republican support — gathered by its sponsor, Rep. James Sensenbrenner (R–WI) — the legislation has never received a vote in either chamber. Rather than reintroducing this legislation when they retook the House in 2019, Democrats broadened the scope of democracy reform legislation to include, among other things, measures to require automatic voter registration, nationwide early voting, a ban on partisan gerrymandering, and voter restoration for citizens with past criminal convictions. The legislation passed on a strict party-line vote in the House and was dead on arrival in the Republican-controlled Senate.

In addition to elite polarization on the expansion of voting rights, attitudes at the mass level are mixed. In advance of the 2016 election, a third of the country expected voter fraud to be a “major problem.” Such fears have not abated, especially given President Trump’s circulation of the myth that there was widespread voter fraud in 2016. Self-identified Democrats overwhelmingly support doing “everything possible” to make it easier to vote. Yet self-identified Republicans appear evenly divided on the question. By contrast, 85 percent of self-identified Republicans and 93 percent of Democrats — at least in the abstract — support measures to limit partisan gerrymandering.

Effectively nationalizing conflict will thus require redoubled efforts to enhance the visibility of democratic deficits and their material consequences for the lives of ordinary voters. It is worth remembering that the legislative record supporting the 2006 reauthorization of the VRA came to roughly 15,000 pages. Despite Congress’s failure to revise the preclearance formula, field hearings on voting rights have continued. Because state politics are obscure and the process of democratic erosion is slow and intentionally ambiguous, more such efforts to create a greater social consciousness about democratic erosion will likely be necessary. In the past, spatially distributed observatories enhanced public understanding and governance of major social and natural phenomena — from climate change to global poverty. A similar observatory of state-level democracy might do the same. It could, for example, produce information about local compliance with democratic norms, state-level institutional obstacles to democratization, opportunities for coalition building, and a better sense of the means necessary to enforce compliance with democracy-enhancing reforms.

Expanding knowledge about the conditions of democracy is no substitute for political action, however. If partisan polarization on the question of subnational governance persists at its current level, and if the courts are not willing partners in ‘center-led intervention,’ popular mobilization by nonelites — in the form of protests, boycotts, and strikes — may ultimately be necessary to defend democratic values in the states. As Adaner Usmani has argued, democratization depends in no small part on the ability of nonelites to engage in disruptive action. If this is so, strengthening democracy in the states may depend on restoring — and indeed deepening — the capacity of ordinary Americans to engage in organized struggle. The raw materials of this capacity are evident in recent social movements that offer decentralized challenges to inequality, as Eric Blanc’s documentation of the “red-state revolt” suggests. In the midst of tremendous political setbacks for organized labor, more American workers went on strike in 2018 than at any time since 1986. Whether these movements can be stitched together, expanded, and targeted at institutional change, remains to be seen. If anything, however, this uncertainty should remind us that the American states are not so much ‘laboratories of democracy’ as sites of an ongoing democratization project.

Acknowledgements: Thanks to Rick Valelly and Rob Lieberman for comments on an earlier draft of this text.

Assistant professor of Political Science // Marquette University // Coauthor, Obamacare Wars (2016) // Coeditor, APD and the Trump Presidency (2020)

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store