Debating Class Dealignment
Class dealignment posits that Democrats have been losing working-class voters in favor of middle- and upper-class voters. Is this actually happening? And to what extent is it a problem?
Neighboring houses in Northumberland, Pennsylvania display signs for opposing presidential candidates. (Paul Weaver / Pacific Press / LightRocket via Getty Images)
- Interview by
- Chris Maisano
A specter is haunting socialist politics — the specter of the “Brahmin left.” The renowned political economist Thomas Piketty, with Amory Gethin and Clara Martínez-Toledano, discussed this group in an influential 2022 research paper on political cleavages in Western democracies since the mid-twentieth century. This group is referring to highly educated voters, typically employed in professional occupations, whose allegiance to center-left and left-wing parties has grown in recent decades.
By calling them “Brahmins,” Piketty and his collaborators likened them to the religious figures and intellectuals who occupied an elite position in India’s traditional caste system. In their framework, these educational elites of the rich capitalist world face off against an economic elite they call the “Merchant right,” while the needs of the vast majority go unacknowledged and unmet. Faced with such a dismal choice, working-class voters are either dropping out of politics entirely or shifting their allegiances from the Left to the Right.
This process of “class dealignment” has become a staple of analysis and commentary across the political spectrum, including the socialist left. It’s clear that electoral patterns in rich capitalist democracies have changed, in some cases rather dramatically since the postwar years. But it’s not always clear what exactly is happening in electoral politics, nor why, or what socialists could do to shift political conflict onto a more favorable terrain.
Jared Abbott and Chris Maisano spoke recently to hash out their respective views on the question of class dealignment in the United States. Abbott is a political scientist and the director of the Center for Working Class Politics, and is a coauthor of “Trump’s Kryptonite: How Progressives Can Win Back the Working Class.” Maisano is a Jacobin contributing editor. Their conversation has been condensed and edited for clarity.
The idea of class dealignment has become a staple of both academic and journalistic discourse. A print issue of Jacobin was recently devoted to it. It’s one of the major perspectives on what’s happening in American politics today. So what is class dealignment, and what’s your own perspective on it?
Unfortunately, when people talk about dealignment, they often talk past each other, because they’re actually talking about different concepts but using the same terms to talk about them. This is particularly the case when it comes to definitions of the working class, for instance those based on educational attainment versus occupation or income-based definitions. There are potentially quite different implications depending on which one you use.
Essentially, class dealignment is the theory that the Democratic Party has been losing working-class voters (and in some versions also gaining middle- and upper-class voters). Some people say it goes all the way back to the post–civil rights backlash against Democrats, fueled by Richard Nixon’s “Southern strategy,” in which the Republican Party’s race-based appeals drew Southern whites away from their historic allegiance to the Democrats.
Others argue that the Democrats’ turn toward neoliberalism starting in the late 1970s and early 1980s generated a class-based resentment toward the Democratic Party, which in turn led a swath of working-class people, especially in deindustrialized places, away from the Democrats.
Yet another version of the dealignment is something we hear a lot of in the news today, which is the idea that the working class, especially working-class white voters, have jumped ship from the Democratic Party during the [Donald] Trump era. For some who advance this idea, it’s because Trump effectively tapped into working-class anger around mass layoffs and the loss of manufacturing jobs to foreign competition, or because of Trump’s stoking of racial resentment against immigrants and black people. For others, it’s because working-class white people are increasingly alienated by the cultural elitism of the “Brahmin left,” highly educated urban professionals upon whom center-left parties like the Democrats increasingly rely.
So there are various versions of the class dealignment argument out there. The version that stresses the Democrats’ neoliberal turn has strong support in studies showing the immediate effects of NAFTA on Democratic electoral support in the most hard-hit communities but gets trickier when thought of as a longer-term trend, because while working-class support for Democrats is lower today than it was in the 1990s, [Barack] Obama temporarily reversed dealignment, so there isn’t a clear and consistent downward trend in working-class support for Democrats in the electoral data until about 2012 — though we do see a steady decline in working-class party affiliation with the Democrats going back to the 1970s. Also, Biden actually did better among working-class white voters in 2020 than Hillary Clinton did in 2016. That said, we do see a pretty sharp decline in support for the Democratic Party since 2012.
Somewhat more controversially, we see some evidence from different surveys and voter file–based analyses of a modest move away from the Democratic Party among African Americans, and especially among Latinos, since 2016. So when it comes to the electoral data, this is a pretty recent phenomenon, partly because Bill Clinton and Barack Obama made attempts to stop the bleeding that actually were pretty successful in the short term.
At the same time, other analysts argue that there hasn’t been a class dealignment at all. This is partly based on the idea that using educational attainment as a proxy for class obscures the fact that we haven’t seen as much dealignment among low-income, high-education voters, and that educational dealignment is mainly capturing the grievances of comparatively well-off whites without a four-year college degree. But that’s not entirely true. Even when we look at definitions of the working class that account for both educational attainment as well as income, or occupation-based understandings of the working class, we still see a decline in support for Democrats since 2012.
That said, it’s true that different definitions of the working class do indeed yield important differences with respect to political attitudes. For example, if you look at trends in social attitudes and economic policy views since the 1980s or 1990s using most definitions of the working class (education and income, occupation, stock ownership, class self-identification, etc.), working-class voters in general have not become any more conservative relative to middle-class voters. But if educational attainment is your method of defining which class a person is in, then you see that voters with college degrees have become substantially more progressive than voters without a college degree, especially on social issues but even on some economic issues. So depending on which way you measure class, the implications for which working-class voters might comprise part of a broad progressive coalition look quite different.
But I think there are reasons to worry about educational dealignment regardless of whether we consider voters without a college degree members of the working class. For one, without winning more of these voters, Democrats’ medium-term prospects in key battleground states look pretty dim. But it is still important to remember that there are real differences that result from using educational attainment as an index for class location versus using occupation.
That’s definitely the case. I’ve pointed this out myself when responding critically to some class dealignment arguments. This all looks very different depending on how you’re defining the working class, whether you’re looking at occupation, income, educational attainment, or some combination of those things.
I have to say, my views on these questions have changed over time. Coming out of the 2016 election, I was fairly convinced that there was something like a process of class dealignment happening in the United States. But since more elections have happened and more data have come in, I think the situation is more complicated than that.
My views on these questions have become much closer to Michael Podhorzer, who’s a sharp critic of class dealignment arguments. Podhorzer was the AFL-CIO [American Federation of Labor and Congress of Industrial Organizations] political director for many years, and he introduces a dimension to the analysis that I think gets left out of many class dealignment arguments. He argues that there isn’t a single political ecosystem in the US, but rather three.
There are states where the Democratic Party is politically predominant, states where the Republicans run the show, and a smaller group of states that are competitive between them. He points out that every subgroup of voters is more Democratic in blue states than in purple states, and more Democratic in purple states than in red states.
Whether you break down voters by educational attainment, or income, or whatever, you see that Democratic candidates tend to do well among many different groups of voters in blue states, Republican candidates do the same thing in red states, and there’s competition within each group in purple states. So in that sense, what’s happening isn’t class dealignment so much as a geographic or regional cleavage that runs through classes and subgroups in the electorate.
To make this more concrete, take my apartment building in Flatbush, in the middle of Brooklyn. My neighbors run the gamut from native-born white people with advanced degrees to families of Haitian or Hispanic immigrants where nobody has a college degree, maybe except for the kids. Some people are religious, others are not. For some people, English is their first language, for others it’s not. Household income is probably all over the place, and so on.
But when I was campaigning for a fellow Democratic Socialists of America [DSA] member for state senate, everyone who’s registered to vote in the building was on my list. Everyone who votes in the building votes Democratic, whether they have a PhD, a big salary, are immigrants who speak English as their second language, didn’t go to college, whatever. No matter the occupation, household income, educational attainment, race, or national origin, because we all live in this particular political ecosystem, we’re all Democratic voters.
The flip side of that would be a neighborhood in a place where the GOP predominates and therefore wins big among everyone who votes regardless of what they do for a living or if they have a college degree. They’re all voting Republican. The paradox of political nationalization is that it is driving ever larger divergences between states and regions, and those divergences are running through every social group and class rather than sorting each one of them out into different parties.
That makes sense, and I am sure there’s a lot of truth to it. But I still think the realities of life for many working-class people — even in thriving metropolises like New York or Los Angeles — are drastically different from those of middle- and upper-class people who enjoy greater job security and future career prospects, and can afford to opt out of many of the day-to-day quality-of-life issues that working-class people cannot. Moreover, I just don’t think the existing data is consistent with the notion that what we’re really seeing are regional rather than class cleavages.
Podhorzer argues that the correlation between educational attainment and voting patterns is a spurious one that’s actually driven by the different economic, political, and social realities of different states. As I mentioned earlier, I find this totally plausible. But when you look at trends over time in class voting in blue, purple, and red states, you see a consistent pattern of class dealignment in all three, even when you use a definition of the working class that accounts for both education and income. Dealignment is not just a story of disaffected non-college whites in MAGA country, it is a cross-regional phenomenon.
Regardless of whatever else he may be or represent, isn’t the fact that Eric Adams’s primary challengers only did well in the most affluent areas of New York City an obvious example of how working-class voters in liberal states can still react very differently to Democratic politicians than their Brahmin counterparts? I still think we have strong reason to believe that working-class voters, regardless of the the region in which they live, share similar experiences in terms of their sense of economic dislocation, disenfranchisement, disaffection, frustration with the quality of public services, and their sense that the Democratic Party and progressives have become detached from what working people need.
It’s true that Adams did well among many working-class voters, particularly in largely black areas outside of Manhattan. But for all his posing as a “blue-collar candidate,” his austerity agenda is hurting working people the most, making him historically unpopular across the board.
I recently read a very good book that spoke to my sense of what I think is going on, particularly in deindustrialized areas, called Rust Belt Union Blues by Lainey Newman and Theda Skocpol.
I’ve been talking it up, it’s a great one.
They look at counties in Western Pennsylvania that have been deindustrialized and which for many decades voted solidly Democratic in presidential and other elections. Today, they’ve flipped to the Republicans up and down the ballot. One of the most interesting things about the book is that you can find evidence in it both for and against a class dealignment story.
The support for a class dealignment story seems straightforward. These are places that, within many people’s living memory, voted overwhelmingly for Democrats; now they don’t. The evidence against a class dealignment story is the fact that places like these have been completely transformed socially, organizationally, economically, and politically. They are just not the same places they were a few decades ago. The plants have been shut down or mechanized, and the people who used to work in them have, in a lot of cases, moved away or are dead.
At one point in the book, they note that there was a mass exodus of workers from the region, something like a Pittsburgh-area diaspora that spread around the country as employment in steel and related industries collapsed. There is still some industrial production in these places, and there are still prime-age working people living there too. But the population has declined, and the composition of the population is different. It’s much older, in many cases, than it was in the past.
So in that sense, there are some real dangers in taking a snapshot of a geographical area at one point in time, then another snapshot later in time, and comparing the election results. We’re not looking at the same group of people in both snapshots; the composition of the population and the electorate has changed. There’s an Obama-Trump county in the upper peninsula of Michigan, for example, that now has a median age of fifty, which likely means the median age of the people who actually vote there is even higher. If it were its own country, it would be the second-oldest country in the world. I’m not sure we can draw any big conclusions about class dealignment based on places like this, nor do I think talking to voters there about wages or unions or job guarantees would get you very far.
Another aspect of their book that complicates the class dealignment story is how they document the changing social and organizational environment in Western Pennsylvania. When industry was present and unions were strong, an overlapping and mutually reinforcing set of networks and organizations encouraged habitual voting for the Democratic Party. That’s your union local — the key anchoring institution in this situation — overlapping with your Catholic parish church down the street, with your Slovak fraternal organization and so on. Everyone living on your street has someone working in the same plant, they might walk to work together, and they drink together at the neighborhood bar after work every day.
These have all eroded or disappeared, and to the extent that they’ve been replaced, there’s a pretty different set of institutions now. It’s gun clubs, police union locals, and evangelical megachurches overlapping with and reinforcing each other to encourage habitual voting for Republican candidates. Election season messaging, no matter how compelling it sounds, seems pretty thin compared to this.
Their research was illuminating for me, and it shed new light on some of the election data I’ve been poking around in. For example, if you look at the election results for the 2022 Michigan governor’s race, [Democratic candidate] Gretchen Whitmer did well with every single income and educational attainment group, and she did better among the lowest-income voters than she did among middle- or high-income voters. One of the few groups with whom she did really badly was white evangelical Protestants, who voted en masse for the extremely MAGA Republican candidate, Tudor Dixon. In that sense, I think that class dealignment analyses sometimes lose sight of the broader social and organizational environments that people are embedded in.
When I was cutting my teeth politically at the tail end of Bill Clinton’s presidency and under George W. Bush, talk about the role of white evangelical Christianity in American politics was everywhere. I feel like this has somehow dropped out of many conversations about American politics, particularly on the socialist left, even though it’s still very important. A working-class white voter might be voting Republican, but are they doing so because of their dissatisfaction with the Democratic Party’s economic agenda? Or is it because of their religious views or their sense that the country is changing socially and culturally in ways they don’t like? It’s not always clear.
There’s also evidence suggesting a growing ideological gap between women and men across countries, with the former becoming more progressive and the latter more conservative. Here in the US, much of the shift toward Trump we observe among black or Hispanic voters is concentrated among the men in those racial groups. The genders are not equally distributed across occupations, either. Women, for example, are disproportionately concentrated in what the literature calls “sociocultural professions,” particularly in health care and educational services. Are nurses, teachers, or social workers in the working class, or not? If we locate them outside the working class, as the “Trump’s Kryptonite” report does, then this will affect our interpretation of the data.
All this is to say, what appears to be a class-based or economic cleavage might actually be evidence of something else. If that’s the case, drawing strategic political conclusions based on an analytical misinterpretation could be very counterproductive.
It’s true that economically based appeals to disaffected, especially white voters in the deindustrialized areas you’re talking about is a gamble. It may not work. It may also be true that the original cause of their hatred of or disaffection from the Democratic Party was economic but may have morphed into other guises in terms of religion, conservative cultural values, et cetera. So even if the original grievances were economic in nature, it may be impossible to appeal to them based on economic self-interest, because now they don’t see what’s happening through that economic prism.
In their book, Newman and Skocpol talk a lot about how workers in Western Pennsylvania used to think about the elites as corporate bosses pitted against workers. Now they tend to think about it through a “Rich Men North of Richmond” framework, with the elites sitting in Washington, DC, or the state capital.
That is a real challenge for people who think there are electoral and political implications to be drawn from class dealignment with respect not just to white working-class voters, but people in general, in the deindustrialized areas that have become very conservative. That is an important critique to think about. I think political appeals to them could not be based just on economics alone, that’s never going to work. The other part of it has to be appealing to people’s sense of being left behind or being taken for granted. Is there some kind of way a populist coalition-building strategy can build upon appeals that would find a hearing among the broad working class, not just among this mythologized deindustrialized white working class?
There’s reason to think it’s possible, but it is a complicated question. People have real grievances with the Democratic Party that are economically based, that cut across region and race and age and many other characteristics, in the sense that for a very long time, the Democratic Party really was not doing very well for working people. In fact, it’s only been in the last year and a half that we’ve seen historic, at least in comparison to the Democratic Party’s recent past, policy measures by the Biden administration aimed at benefiting working people, even though most of them aren’t feeling it and therefore don’t reward him or the Democrats generally for it. It’s the first time we’ve seen the neoliberal ice being chipped in a meaningful way, in policy and not just rhetoric, since the neoliberal era began.
Economic grievances do play a role in people’s electoral choices, even if it’s sometimes indirect. I think that there is reason to think that there are shared economic grievances that cut across the broad working class, including these workers in Western Pennsylvania that we’re talking about, that could form the basis of a broad coalition. Obviously, you’re not going to get a lot of these super dyed-in-the-wool MAGA folks or almost any of them, but you might reach some MAGA-curious swing voters or low-frequency voters in those areas.
But how we think about all of this really depends on what question we’re trying to answer. If the question has to do with long-term electoral strategy or the longer-term project of building a working-class coalition for the Left, I think the answer will be different to one we give for ensuring the MAGA crowd doesn’t take national power in the next few election cycles. For that latter question, it’s just the case that if Democrats and progressives can’t maintain and even increase their electoral support among Newman and Skocpol’s working-class voters in deindustrialized areas, then the electoral math just isn’t there for winning national power, as they can’t depend on reviled presidential candidates and horrendous Supreme Court decisions that galvanize the resistance forever.
On the question of longer-term class dealignment, there is a reasonable case to be made that this group of people I’m talking about is not worth the trouble because of an increasingly large group of voters of color and urban voters, who in many cases straddle the working class and the professions, and who are very progressive across the board, and this seems like a more natural fit for today’s left. And I think that there’s something to that, something we obviously lose if we don’t consider “sociocultural professionals” as a key part of the broader coalition we’d need to build.
Still, I am somewhat skeptical that it makes sense to build a coalition, as the Democrats have been doing, around pretty affluent professionals — not necessarily teachers or nurses — who sound progressive on surveys that don’t force them to make any real choices. Sam Zacher had a really interesting article in Perspectives on Politics last year where he showed how affluent professionals tended to be more conservative in real-world settings when they had to vote on referendums that forced them to put their money where their mouth is on redistributive policies. And anyway, surveys consistently show that working-class Americans remain more progressive on most economic issues than middle-class voters.
So I worry about the class composition of the Left becoming too professionalized because there are actually trade-offs in terms of the priorities of the different groups in the party that could have serious implications for progressives’ ability to deliver long-overdue transformative economic reforms for working people.
The other concern I have is that if we are not working in these communities where it’s really hard, many people are never going to meet a progressive or a Democrat. They’re just going to become so siloed that the long-term prospects of the progressive movement to extend outside of big cities and in very Democratic places is going to get smaller and smaller as the power of the right-wing and the MAGA faction completely overtakes all other political orientations. This would leave us with a high level of reactionary political intransigence in more than half the states of this country, which would have really significant political effects.
In their book, Newman and Skocpol talk about the need to be present in some of these communities and show that progressives are doing something for them. Democrats and progressives need to run campaigns where they’re going to only get 30 or 40 percent of the vote, or maybe even less, to promote a long-term process of rebuilding in these communities. To say that we don’t need them is something that would have long-term political implications that would be pretty scary.
But how do we account for the fact that Democratic candidates have, all things considered, actually been on a pretty good electoral run recently? We keep expecting Democrats to get wiped out, and it keeps not happening. Not only do they not get wiped out, but they often do better than we might expect, particularly in statewide races where every single vote counts the same.
Candidates like Whitmer are winning in Michigan. Janet Protasiewicz crushed her MAGA Republican opponent in last year’s Wisconsin Supreme Court election. John Fetterman and Josh Shapiro won in Pennsylvania. Katie Hobbs and Mark Kelly both won their races in Arizona. If you look at the data, they all did well among working-class voters, whether you’re defining this in terms of educational attainment or household income. When Democrats won trifectas in Michigan and Minnesota, they pursued a pretty progressive agenda on both economic and social policy. I’m definitely not saying that Democrats don’t have any problems with working-class voters, but I think this is sometimes overstated by both the Left and the Right.
I definitely agree with you about needing to run progressive people anywhere it’s feasible, even if you’re not going to win the election any time soon. What’s interesting about this, and this something else I’ve learned from engaging with the work of Skocpol and her associates, is the people that have been doing this are often “resistance libs,” a group I’ve definitely softened on.
She and her collaborators have looked at these counties in Pennsylvania, Wisconsin, and other places where working or retired teachers, librarians, or health care workers, often college-educated women, formed progressive organizations from scratch or revitalized moribund Democratic Party county committees. In some cases, they’ve fielded candidates to run against Republicans for the first time in a long time. I think of teachers and librarians and health care workers as part of the working class, and they’re doing exactly what we’re saying we want to see working-class people in nonmetropolitan areas to do.
How do we square this with the fact that these developments are often equated on the Left with a move away from progressive politics within the Democratic Party, because activists like this supposedly have different class interests than, say, a manual blue-collar worker or a non-college-educated service worker?
It’s not just professionals who are running for office in red states, and when they do, it’s often in the more urbanized parts of red states, not the rural or small-town areas. I’m not saying this is a bad thing, far from it. Indeed, our analysis at the Center for Working-Class Politics found that teacher candidates actually appealed more to rural voters than candidates with a manual-labor background. But it’s an important distinction, because you can run as a progressive in Charlestown [, Massachusetts] or Morgantown, West Virginia, in a way that you can’t run anywhere else in the state. So if you’re seeing a teacher run for city council in a college town in a red state with an approach that’s across the board progressive, it’s easier for them to do it in a place like that than it would be elsewhere. But this is more about the specific political orientation of candidates than it is about candidates’ occupation. That is, there is no reason to think that teachers couldn’t also run competitive races in rural/small-town areas.
At the same time, there are also working-class people besides teachers and health care workers that have been running for office in red states. They just don’t get talked about a lot. For instance, in 2022, there were around a dozen Democrats with service sector or manual-labor occupations who ran for Congress and won primaries in red or purple states like Texas and Arizona and a few other southern states. That’s not a lot, but there are more examples of working-class people in manual or service sector occupations running in red states than the narrative that you gave implies.
Though you’re right that the sorts of resistance liberals you mentioned above are playing an important role, and I don’t think that they should be in any way demeaned for doing that. There are progressive professionals who are just genuinely committed activists, and there’s a lot of them — I mean, shit, we are two of them!
But there’s a difference between recognizing the value of individual teachers and nurses as key progressive candidates and activists and the broader political implications of an increasingly affluent progressive electoral base. There are obviously many middle-class professionals and even corporate executives who are meaningfully progressive across the board. But I don’t think that changes the troubling political implications of having millions of more affluent folks within the Democratic electoral coalition generally.
Regarding your point about Democrats often not losing, well, you’re right. That’s a very reasonable argument, and I think that it is good to have debates about this. These are not actually clear-cut issues, and I’m happy to admit that. It’s not impossible to imagine that Democrats can keep doing what they’re doing and keep winning in places like Michigan or Wisconsin with the same sort of coalition that they have, although again, people like Whitmer and Fetterman won because they did a much better job among working-class white voters/non-college-educated white voters. It wasn’t just the professional-poor coalition we often see from Democratic candidates.
Democrats may continue to go down this path they’ve pursued over the last couple of election cycles, and they may continue to succeed. I just think that the combination of more Latino and African American voters abandoning the Democrats, the preponderance of white working-class voters in key swing states, and the extremely narrow margins that decide elections in those states suggest that without a course correction Democrats won’t be able to hold on for much longer.
We can’t just keep relying on having Republicans put up really crappy candidates. Or maybe we can! They’re really good at it and they may keep doing it, but I wouldn’t bet on that as a long-term strategy for Democrats.
But the broader issue is that the party’s current strategy is not a means of building a coalition for change. It’s a recipe for divided government, or for having really narrow majorities in the House and the Senate. That precludes more progressive policies from being made in Washington even if you have a Democratic president, as we saw with Joe Manchin and Kyrsten Sinema. When you have these narrow margins, you can’t actually deliver on any kind of robust, large-scale economic programs, or on social issues for that matter of the kind that you and I think are necessary for moving toward socialism or social democracy. If we don’t get those things, then working-class people are going to continue to be pissed off at the Democratic Party. They are going to continue to be more disillusioned by it.
We really need to take seriously this idea of building the margins up even in more conservative working-class areas. Otherwise, we’re simply not going to be able to build a durable coalition that can pass the legislation needed to convince working-class people that progressives can actually do anything for them. We could, of course, just ignore the national stage altogether and just focus on the state level. But if we are thinking in terms of national electoral politics and the potential for transformative economic or social reforms that might create a real realignment with the Democratic Party and working-class people, a la the New Deal or something like that, that’s going to require a bigger coalition than what the Democrats currently have.