How the states went nuts: Democratic backsliding in state capitals — and how to defeat it

Every week seems to bring a new stress fracture in American democracy to light. While the drama of the Jan. 6 hearings focuses attention on former Donald Trump’s top-down predation, the Supreme Court’s overturning of Roe v. Wade has unleashed a torrent of state and local conflicts that only grow more intense. This is happening even as many states now enforcing or enacting abortion bans laws actually have pro-choice majorities, as noted in this Monkey Cage analysis by political scientists Jacob Grumbach and Christopher Warshaw.

What comes next is very much up for grabs, but Grumbach’s new book “Laboratories Against Democracy: How National Parties Transformed State Politics,” casts a dramatically different light on how we got here, which in turn says a lot about the task of setting things right. 

While “Trump has been characterized as an aberrant wrecking ball that disrupted American politics, Grumbach writes, “it was the states that were the wrecking ball, clearing a path for Trumpism throughout the American political system.” His book refutes received wisdom about wisdom of American federalism: Rather than stabilizing and strengthening American democracy, the relative autonomy of states has played a significant role in undermining it, and the disconnect between abortion laws and public opinion disconnect is just one example of a broader democratic breakdown. 

State politics is not so much the culprit as the conduit, as Grumbach’s subtitle indicates. As the two national parties became more homogeneous with the breakdown of the New Deal coalition, the nature of state-level politics changed significantly. In the period Grumbach’s data covers, from 2000 to 2018, there was significant change in public opinion regarding marijuana legalization and LGBTQ rights that was reflected in state legislation. But these two high-salience issues were the exception, not the rule. “For the rest of the issue areas, state policies have changed profoundly, but state opinion has been mostly static,” he writes. “And you can’t explain change with a constant.”  

This antidemocratic lawmaking was just one example of a nationally-driven agenda that voters weren’t asking for, either on the state or national level.  That’s the good news, as well as the bad. To make sense of this contradictory state of affairs, I recently spoke with Grumbach about his new book. This interview has been edited for length and clarity.

In your preface you use the metaphor that the states served as a “wrecking ball clearing a path for Trumpism.” What are some salient developments you’d cite as examples of this?

There’s many angles to take on this. One is that since the 2010s there is a huge shift in state-level democratic institutions and some serious democratic backsliding in a handful of states, including key Midwestern and Atlantic seaboard swing states. States like Wisconsin and North Carolina, which Barack Obama won in 2008, proved hugely consequential, as an example, in the 2016 election.

In the 2010 redistricting cycle, you saw states like Wisconsin set records for partisan bias in legislative gerrymandering for both their U.S. House seats as well as their state legislative seats. This allowed a minority of voters to set state legislative majorities, and also to affect who controls the U.S. House. So, that’s one thing. Later, after the Supreme Court ruling in Shelby County v. Holder, those state legislative majorities were able to restrict voting rights and make it more costly and burdensome to vote.

By contrast, other states, like where I am in Washington state or Colorado, were expanding access to the ballot over that same time through things like automatic voter registration, same-day voter registration and expansive mail balloting with drop-off locations. That divergence in democratic performance proved really consequential, and that’s happening at the state level. And if the Supreme Court rules on the “independent state legislature” doctrine, which may be incoming, it will be even more so.

You also write that “the three monumental crises in 2020 [meaning the pandemic, policing and democracy] revealed an American political system that lacked the capacity to solve fundamental challenges.” Keeping in mind that you don’t blame federalism alone, what would you highlight to make your point?

Those three crises are different, but they all strongly point us to the role of decentralization and state governments in not solving these problems and allowing them to worsen over time. First would be COVID-19. Early on state governments had trouble coordinating on getting PPE equipment to essential workers, and lack of coordination during the economic crisis precipitated by COVID-19 was hugely important. Also, states don’t have a unified unemployment insurance system. They have various underfunded and decentralized forms of welfare state provision, like unemployment insurance or Medicaid. The fact that states were allowed to reject Medicaid expansion, even though the federal government is paying for it through the Affordable Care Act, means that people who are working poor, but may not have children do not have access to Medicaid in these states, which worsens COVID relief, and COVID-based health care. 

Finally, with Trump in the White House and governors of different parties in the states, decentralized authority means decentralized accountability. So that means that no one political office or legislature was really responsible for the outcomes of COVID-19. Decentralized authority means that politicians at different levels of government — mayors, governors, presidents — can reasonably point to the other and say, “That crisis going on in your area, it’s this other level of government’s fault!” This makes it extremely difficult for voters — especially in the context of sensationalist national media and the decline of state and local journalism — to hold their politicians accountable. If you’re a Republican, you can say, “It was my dumb Democratic governor’s fault.” If you’re a Democrat, you can say, “It’s Donald Trump’s fault.” It’s really hard to draw lines of accountability there. 

What about the second crisis: policing?

Policing is constitutionally a state level authority, and states then delegate authority to local governments. So mayors are the commanders in chief of their police departments and states then have ultimate authority over them. Yet what we see pretty consistently is that police departments tend to be the de facto government and they’re extremely insulated from democratic inputs. Police departments are essentially impervious to reform, so when reform-minded mayors and district attorneys come in at the local level what you see is not much change.

One finding I have in the book is that criminal justice policy, unlike other policy areas like abortion rights, the minimum wage, labor relations, health insurance or climate change — all of those areas are really polarized by party. But in terms of policing and incarceration, there’s not much difference between red and blue areas, they all tend to take the “tough on crime” approach, even when reform-minded state and local politicians take office. Unlike in other democracies that tend to have centralized authority over policing, in the U.S. this decentralized form of policing allows police department cartels to really run their towns and cities. 

Want a daily wrap-up of all the news and commentary Salon has to offer? Subscribe to our morning newsletter, Crash Course.

And what about the last crisis: democratic backsliding?

We see throughout American history that state-level authority — state legislatures, especially — have been the main forces of democratic backsliding. They’re often aided by a permissive Supreme Court that says, “States, you can do what you wish” with respect to elections or gerrymandering or, in the past, Jim Crow laws and before that slavery, giving states free rein to do that. Democratic backsliding right now is not as extreme as under Jim Crow or slavery, but it is meaningful. What we see is that Congress then decides to get its act together and stop the state-based backsliding, or not. And Congress has not organized to pass laws to ban gerrymandering or certain forms of voter suppression, which it certainly could.

You argue that today’s Democratic and Republican parties have nationalized in a way that has fundamentally changed how American federalism operates. So, two questions. First, how have they nationalized in a new way?

Throughout American history, the constitutional system has been really decentralized. But what’s new and recent is that the political parties are no longer decentralized. They’re very nationalized through a series of processes since the 1970s. One is the breakdown of the New Deal coalition, which meant the Democratic Party in the mid-20th century was extremely decentralized. Southern Dixiecrats were pro-Jim Crow, while Northern labor and civil rights Democrats were in favor of egalitarian racial democracy and economic justice. So the Democratic Party was very decentralized through racial politics and then, more recently, you see the Republican Party take on a “Southern strategy” since Nixon which made the party much more Southern.

So that’s part of it: racial realignment. But the other processes are things like the nationalization of fundraising since the ’70s, where activist organizations, new technology, changes in campaign finance, economic inequality which gave donors much more money to spend, all those things contribute to nationalizing how politicians can be successful. No longer do you tap into local elites to get elected, you can tap into national fundraising and also nationalized media. Since the 1990s, the advent of the internet and Craigslist really destroyed newspaper revenue, so we saw the decline of state and local political journalism, the rise of the internet and especially cable news.

It seems cliché at this point to say that Fox News had a big effect, but it really had an overwhelming effect on nationalizing politics and making the Republican Party much more competitive. That started in the 1990s, which coincided with the Gingrich revolution that made the Republican Party much more extreme and aggressive and nationalized in orientation. Great research by Greg Martin, for example, shows causally that Fox News has improved the electoral fortunes of Republicans over this time period. On the Democratic side, you’ve got groups like MoveOn, environmental groups and various other activist groups that have nationalized the Democratic Party. There are huge differences between the parties on all sorts of dimensions, but it is true that both became nationalized. 

So how did this connect with the main consequences that you cite? First, the resurgence of state-level policymaking?

As the parties nationalized you see that Democrats across the country become more similar to each other. There are no longer Southern conservative Democrats and northern liberal Democrats. There’s one Democratic Party and one Republican Party. States like Mississippi and Arkansas had Democratic state legislatures up to the 2000s. It’s really remarkable how long it took for them to go Republican. They had a lot of Joe Manchins and even more conservative Democratic state legislators in the South. 

Those states become Republican and start passing things for the national Republican agenda. That also happened in the Midwest after 2010. On the Democratic side you see states start to have similar agendas on things like climate and LGBTQ rights and so forth.  

Talk about Louis Brandeis’ phrase “laboratories of democracy,” and why it didn’t work out that way.

The idea was that states would serve as national laboratories making policy experiments and would learn best practices from each other. So we have a financial crisis, let’s see which states do best in tweaking economic policy, and we’ll copy them. In the modern period you don’t really see any evidence of that. Policies that do well are no more likely to be emulated by other states, unless that state is controlled by the same party. It’s much more about two national communities of state governments, when it was not all about party in the previous period. 

What about democratic backsliding in states controlled by the Republicans?

States have done really bad things for democracy in the past — slavery, Jim Crow and so forth. But the politics of those moments of backsliding were much more regional. They were barbaric, but they were really not about national goals. Whereas now you see through Jan. 6, for example, as well as issues around trans rights or critical race theory, that these are really national in orientation. If you listen to the Republican voting base, they say, “Our country is slipping away,” and their opposition to the direction of the country is in terms of multiracial democracy rather than highly localized and regional conflicts like Jim Crow legislation. 

You’re seeing a national orientation, where if the Supreme Court rules on this independent state legislature doctrine, state governments and state legislatures are really battling over who is going to control the country. So when they decide to gerrymandering House districts or suppress the vote, or shut down ballot drop-boxes or whatever — all these voter suppression policies are done with national ambitions to affect national elections, not just control of your state. And that is fundamentally different than before. So the collision of these national parties with these decentralized institutions is creating this explosive moment for American democracy that we may see in the 2024 presidential election.

Talk about the arguments in favor of federalism. We’ve already discussed the Brandeis tradition, but what about the “decentralist” argument?

Going back to the Federalist Papers and James Madison, the “decentralists” argue that in a large, diverse country like the U.S., it will be more harmonious if we can customize our local areas to the policies we like and not have to battle this out at the national level. So places that are very religious can have a more socially conservative life, while places can be more permissive and libertarian in spirit, they can be customized to local culture and conditions. 

Another argument Madison made is this idea of “double security”: Federalism, by decentralizing authority, meant that one autocrat or dictator couldn’t capture the whole system easily. I think there’s something to both of those arguments, in the best of times. If you have a would-be dictator with national power, certainly that’s not the time to centralize all sorts of institutions. At the same time, state-based authority and decentralization helped propel Trump to office in the first place, right? So there’s a flipside to the trade-off. 

What about the “new federalist” argument? 

So the new federalist arguments are a set of more political economy-based theoretical arguments in the latter half of the 20th century as the social sciences were getting more sophisticated. They believe a lot of the arguments of the decentralists, but use more sophisticated tools. One of their big arguments is that governments will be more efficient and effective because individuals in a decentralized system can move to the place that they’d like to live under.  

There’s not much evidence that’s working very well. First, it’s hugely costly to move, even across states, and the people that are able to threaten to move and can influence governments by saying. “I don’t like the taxes in this state, I’m going to move,” tend to be much wealthier people. So this gives wealthy individuals and big business an advantage over ordinary workers and voters. You hear this all the time, whereas ordinary people don’t have that same ability to threaten to leave. 

One key observation you build on is that while policy has shifted dramatically at the state level, public opinion in the states has been mostly static over the past generation. 

A key example is abortion policy. I wish I had written this knowing all of this would happen now, but I think I called it pretty accurately. Legal abortion has like 61% support in the U.S. and has been at about that level for a generation or more. But right now we’re seeing some states banning abortion despite no real change in public opinion or, if anything, an increase in support for legal abortion over time. 

Note that the issues where you do see a lot of responsiveness are those that have seen big changes in public opinion, like on marijuana. States that saw huge support for legal marijuana legalized it. The same with LGBTQ rights. Before the Supreme Court ruling that legalized same-sex marriage in all states, more progressive states started implementing marriage equality. 

But on other issues where you see states take hugely important policy changes over the past generation — on taxes, or on labor relations, where in the Midwest they totally dismantled labor unions, for example — you actually don’t see people suddenly saying, “Oh, I really hate unions.” That’s just not happening. It’s happening because of these group coalitions when a party takes power in a state and has an ambitious policy agenda based on its coalition. Now the party that controls your state really, really matters for policy outcomes. Again in most areas, you don’t see much change in public opinion despite these huge policy changes. 

The “laboratories of democracy” idea wasn’t about isolated state-level experimentation, but the collective learning process that’s typical of science. So how did that model break down?

One trend I do find is that policy emulation or copying between states is much more within one party — parties only share policies within themselves now — than it was in the 1970s. Policies that political economists used to think were objective standards of quality — things like whether the economy grows or whether unemployment gets reduced by state economic policies — are not likely to be emulated by other states if they’re controlled by the other party. So it doesn’t matter how well the state next to you is doing: If it’s controlled by the other party, you will not copy their policies. 

Your most crucial finding has to do with the undermining of democracy itself. How do you go about measuring the state of democracy and democratic erosion? 

Democracy is a really tough, big concept. One thing that democratic theorists and philosophers do is to break it down into different components. So I mostly talk about electoral democracy. That’s things like: Are elections free and fair? Do people have a reasonably easy time getting to the polling place and casting a ballot, or are there really tough burdens to doing that? How fairly are districts drawn in legislative maps? How secure are elections and how much integrity do they have? Do states follow public opinion when passing policy?

I take a bunch of indicators or variables of measures of how fair districts are, or whether a state has certain policies around allowing people to register to vote on Election Day.  Or, you know, whether they allow absentee voters, or do they have to prove an onerous health burden, things like that. I put them into a statistical model which then tells me how to weigh those different indicators in measuring democracy. So it’s not me imposing my philosophy of things on it. It generates a state democracy score over the past couple of decades, and the big finding there is that some states have done some really serious democratic backsliding, and that has pretty major consequences for democratic performance in the U.S. as a whole, and for the stability of the American political system.

What were the competing causal theories that you tested to explain this democratic backsliding?

This has been a big question for a long time. What causes a society to be a democracy or an autocracy or an oligarchy or whatnot? There’s been hundreds of years of theorizing about this. With the focus on the U.S. and recent democracies, there’s been some focus on the idea that democracy is really about political competition between the parties, or that it’s about polarization between the parties: How distant are the two parties? If they’re very polarized, democracies may not work very well. 

Another argument is about immigration and racial threat. This is true around the world: As societies change through immigration or there’s increased economic or political power of a minority group within society, the majority group gets threatened and will reduce democracy. You see in European far-right parties that are anti-democracy that they are really mobilized against new immigration. 

Finally there’s one line of theory that’s more focused on parties of the wealthy and parties of racial hierarchy, Those tend to be anti-democratic parties and the Republican Party, kind of uniquely around the world, is a party whose elite constituency wants high-end tax cuts and things that help the very wealthy, and also wants to restrict multiracial democracy and maintain racial hierarchy at the electoral base level. 

Both of those things point towards not wanting to expand democracy and potentially wanting to backslide democracy. What I find is none of those other potential causes are really driving democratic changes in the states. It’s really the national Republican Party. When it controls a state it backslides democracy and I think this can be explained by the fact that the wealthy constituency of the Republican Party doesn’t want a robust democracy of working people redistributing their wealth, and the electoral base doesn’t want a party that has important political influence and political power for Black Americans especially, as well as recent immigrant groups.

You use the term “plutocratic populist partnership” in talking about the GOP. In contrast, you note that civil rights activists such as Martin Luther King Jr. and Bayard Rustin, as well as labor leaders such as Philip Randolph and Walter Reuther, emphasized linkages between race, class and democracy, arguing that powerful interests exploit racial division for political gain. This is precisely the argument that Ian Haney López and Anat Shenker-Osorio make, and I read this passage as a de facto endorsement of their approach.

Yeah, I really like those two. I would say they’re focused on the behavioral aspect of it and whether messaging in the race-class narrative, as they put it, is most effective. I don’t do that. I do the political economy and policy side rather than the psychological and behavioral side, so I can’t comment on whether that’s the most effective messaging. It seems pretty effective to me,  but that’s not my area of expertise. But it’s clear that one thing that’s driving changes in democratic institutions — and we see this in terms of political coalitions throughout American history — is this tragedy of the white working class rejecting coalitions with Americans of color that are also working class. They’ve rejected class-based multiracial coalitions, and as W.E.B. Du Bois has written, have been willing to accept a lower status that at least is not the lowest rung. 

The Republican Party at the elite level, which I study much more, is clearly about maintaining this coalition. You see wealthy individuals and big business titans, they realize they’re not going to be popular for their economic platforms. It’s really hard to win elections saying, “Let’s cut taxes for billionaires.” So you have to build a coalition with more popular items and somehow get members of the white working class to support the big-business party that supports tax cuts and anti-labor policy. One way to do that is to really focus on anti-immigration politics or social conservatism around gender and sexuality, to avoid elections being about your economic platform. Because it’s clear that the Democratic economic platform of more labor rights and health care spending tends to be more popular, but Republicans are very electorally competitive and it’s not for their economic platform. It’s for these other cultural areas of politics. 

So what do you suggest should be done to protect American democracy, in terms of the problems you identify?

A basic pointer for politicians at the national level is this: If you have the opportunity, it’s really important to pass national policy that prevents states from doing this type of democratic backsliding. I know that seems ridiculous to say, with getting things through the U.S. Senate and the Supreme Court the way it is. But it’s clear there have been missed opportunities to pass national policy that would prevent democratic backsliding. You could pass a gerrymander ban. You can reform the Electoral Count Act to prevent a stolen presidential election by state legislatures. There’s a lot of things like that. 

There’s also things that are just as important for protecting democracy, through national policy supporting the labor movement and labor unions. My research with Paul Framer shows that the decline or destruction of labor unions has really increased the power of resentment-based politics, and has helped the Republican Party in this era of the culture war. The decline of labor is not just a problem for wages and health benefits and working situations, it’s also a big deal for democracy. 

For ordinary activist types reading this book, I would say that state-level politics is highly important and hard to monitor, and that organizations really matter and it takes a long time to produce electoral gains. So long-term organizing, especially through the labor movement, is important. The labor movement is well situated for this, because locals within the unions are federated but have a national coordinating mechanism. It’s important to get involved with long-term robust organizations, not just sending a check to a national group in D.C., but getting involved in your community in an organizational capacity. Even though it’s extremely boring and thankless, it’s more effective over the long term than all the phone-banking and text-banking every presidential election.

One reason labor unions are really important is that we go to work all day. That’s like our main task in life. That’s one reason why the gun-rights community on the right has been very successful, and the religious right. Having something that brings you together through a social aspect makes you a really powerful force in politics. Groups on the left should think about that. The resurgence of young people organizing union campaigns at Amazon and Starbucks is, I think, one of the bright spots in American politics.