Election Reflections: How Will Workers and Unions Impact The 2020 Election?
Historians can point to a number of times and places where a political party for workers showed some signs of possibkle life in the United States. My favorite instance is Henry George’s mayoral campaign in New York City.
“We are building a movement for the abolition of industrial slavery,” he told a packed house at his campaign launch event in the Great Hall of Cooper Union. “And what we do on this side of the water will send its impulse across the land and over the sea, and give courage to all men to think and act.”
George, a prominent socialist, lost to a Republican, a 28-year old named Theodore Roosevelt, who marshalled the Tammany machine to outmatch him. “The best opportunity to launch an American Labor Party,” wrote Michael Kazin, “probably died when the polls closed in Manhattan that November evening in 1886.”
The U.S. working class has never had a party to advance its interests; instead today we have two parties that oppose them. How do workers generally relate to those parties?
American workers had long been dependable Democratic voters, with strong affinities for New Deal-style reforms. Not anymore. Today, we see the inversion of our historic political geography, where states with high percentages of working class voters are solid red states. Slowly, mostly during presidential election cycles, we see evidence that the demographics and class politics of the working-class electorate has shifted. The political identity and behavior of U.S. workers is not what we thought, and they consistently confound the pundits and politicians who think they know how workers think and what they want.
Middle and working class Americans — manual laborers, hourly wage workers frontline service providers, etc — make up half of the electorate, but are woefully under-represented in politics. While African Americans and women have moved into roles as professional politicians, working class people remain stuck where they were in 1960. They simply have no voice in politics. As Martin Gilens of Princeton and Benjamin Page of Northwestern have shown, the interests and demands of average workers never win against the wishes of the tiny elite with all the influence. American democracy exists in our imaginations — we instead have a society in which one percent of the population controls nearly all the wealth.
The resulting marginalization in politics is often considered to be the main reason why so many workers seem disaffected and out of step with their historical political home in the Democratic Party, and even why we saw so many Obama -Trump voters in 2016. Thus commenced a national obsession with the psyche of the “white working class.” Ordinary workers still tend to vote blue and it is still the wealthy who are the most dependable right-wing voters. Class-based voting patterns linger, but these trends are faint echoes of the past. Some major shifts, however, have changed the working class composition to produce an historic political realignment.
Unions tend to make people more democratic and as unions have lost ground among whites, many turned more conservative. Also, voting patterns based on educational attainment have inverted in the post-war period across most rich capitalist countries. Educated elites used to vote conservative and now they’re the most committed democrats, a switch which the economist Thomas Piketty says delivers us the “Brahmin left.” This Left is increasingly disconnected from the working class, stands against it, and aghast at it, from its elite vantage point within universities and the heights of the knowledge economy. It’s not a new thesis, but the data he marshals is quite impressive.
Changing voting patterns by education has had important consequences for the outcome of national elections. This is because there are so many more voters without four year degrees than those who have them, and our electoral institutions offer them out-weighted advantages. Political scientist John Rodden points to the urban-rural divide as a causal reason for why the Left has such a hard time winning elections. There are about 14 million more registered democratic voters in the country than republicans, but the geographically-based winner-take-all system means that progressive votes that are captured in dense areas (cities) are routinely beaten by the disproportionate privileges given to the rural electorate.
For many years, polarized voting on the basis of class helped to overcome this structural bias. Unions were helpful here because they channeled workers’ votes toward the democrats. The decline of the unions, the rising salience of social issues and the decline of class-based voting mean that working class interests are often drowned out within each party, and are not as driven by a movement. Still, about ten percent of Americans are union members (about 6.3% in the private sector and 37 percent in the public sector). That’s 16.4 million people.
They matter a lot in elections and will factor heavily in this one.
Union members are about 12% more likely to vote than non-union members but much of that difference might be accounted for the socio-economic differences between members and non-members. It seems that union members on average vote at similar rates as do those who hold pro-union views, whether or not they’re card-carrying / dues-paying union members or not.
This graph shows how union households have voted in presidential elections from 1968–2016. A union household means at least one person who regularly resides in the home is a union member. This data is often used instead of union member data for a few reasons, but mainly because the sample is larger, the data is easier to find, and, according to my calculations anyway, the trends are nearly identical even though some research shows they aren’t. The political stats people at the AFL-CIO, who conduct regular polls on their members, claim their data is far superior than the measly information we get from other sources, in this case the General Social Survey. But their data is private. So it goes…
The most interesting trend now is from 2008 to 2016, where the union share voting democratic takes a significant dive, as unionists turn toward Republicans. Trump’s strategy was always to pull Obama union-voters away from Clinton and he largely succeeded. In fact, his margin of the union vote share was only eight points behind Clinton, the best marginfor a Republican since Reagan won his second term in 1984.
But it’s even worse. Trump didn’t just pull working class whites to his side. Changing demographics since when Reagan was in office means that the union electorate is much more racially diverse than it was in the eighties. In other words, Trump more union votes away despite the working and middle class base of the unions being far less dominated by whites.
Some of this change can be explained by rising numbers of women in the unions and rising racism and nativism within the U.S. working class. Both groups voted more for Hillary but at far lower levels than they voted for Obama. This shift was enough to tilt the election in Trump’s favor. Nate Silver provides data from the Cooperative Congressional Election Study that shows:
“Obama won union voters by 34.4 percentage points in 2012, but Clinton did so by only 16.7 points in 2016. That roughly 18-point swing was worth a net of 1.2 percentage points for Trump in Pennsylvania, 1.1 points in Wisconsin and 1.7 points in Michigan based on their rates of union membership — and those totals were larger than his margins of victory in those states.”
The union vote matters even more in swing states like Pennsylvania and Wisconsin. Biden kicked off his campaign in PA, where about 12% of the state is unionized, and where he’s hoping Trump’s failed promises to workers — keep plants open, saving coal industry — will encourage unionists to lean back to blue. PA voters have epitomized some of the other shifts regarding education mentioned above, with lifelong democrats shifting right in 2016 and higher-status Republican voters leaving Trump this year.
Just for kicks, in October I conducted a survey of about 700 essential workers — those who worked in person during the pandemic — focused heavily on Pennsylvania and Wisconsin. To do this I purchased ads on Facebook that targeted essential workers in retail, health services, education, manufacturing, and entertainment industries. In general, the demographics skewed liberal, with more self-identified democrats responding to the survey and a barrage hateful comments from Republican-identified people arguing that COVID-19 is a liberal hoax. Yet, as you can see below whether or not someone was a union member seemed inconsequential to how they planned to vote in 2020, suggesting that union membership actually has a de-polarizing impact of workers.
In general, most of my survey respondents reported that they’d become more liberal during 2020. They believed Biden was a better candidate to pull us out of the Covid recession and that he’d be better as a candidate to help workers. But those beliefs were strongly corelated with political partisanship. And there seems to be a lot of that.
The graphs above measure the views of each party from the perspective of their opponents. These days Democratic- and Republican-identified working class people hate each other more than either of them hate bosses. This simple fact can explain a large part of the transformation of the union vote over the last century. In general, the survey found that the more a respondent put themselves at the far end of their respective political spectrum, the more likely they were to say that they were voting against their opposed candidate, not that they really endorsed their own for his merits. It’s predictable except for an interesting finding among democrats. For democrats, there is an inflection point where “very liberal” people are less likely to view Democrats positively than moderates, perhaps showing a rift between the Left and establishment liberals. On the other hand, “very conservative” people rate the Republican party more highly than moderate conservatives. This could be because the GOP itself is “very conservative” whereas the mainstream Democratic party is moderate by contrast. This suggests not only a different ideological bent between the two but different campaigning strategies. Trump is relying on the same voters he did in 2016 — white industrial workers in the Midwest — and polls suggest he’s unlikely to get as many of them.
Take Ohio, which Trump did in 2016. Trump won whites without a college degree in Ohio by 24 points, and beat Clinton by 8 points in the general, but now he’s tied with Biden for those voters. Same in Pennsylvania, where Trump’s lead shrunk to nearly half among non-college whites according to a recent Fox News poll. This shift can largely be accounted for not by a stronger Democratic candidate but because Trump has performed so miserably in blue-collar America, that even his own base seems to have, at least partially abandoned him. Trump’s popularity among white working class voters has diminished, as those voters seem to be trending toward Biden. This trend still holds even given the reports of a rightward shift among Blacks and Latinos.
I spent the last few years interviewing workers around the country about their jobs to write a book on why we spend so many hours on the job. I met a few folks from a former GM plant who remember vividly the promises Trump made when he came to Ohio after that. “He told us not to sell our houses, that he’d rebuild the factories, get things back to work,” one told me. “It all looks like bullshit now.” There seems to be a small realignment taking place within Trump’s former stronghold. Will it be enough?
The polls suggest it will, and there’s reason to think the polls will be more accurate this time around. The union vote will be key and the situation looks slightly different. In 2016 unionists helped Clinton defeat Bernie in the primary and helped Trump defeat Clinton in the general. The small shifts perceptible in the union vote and the white working class vote are hopeful. But hope isn’t enough.
Unions across the country are currently mobilizing to challenge the outcome of a fraudulent election with a mass strike. As Trump’s chances look dim — though not pitch dark — in a reasonably fair contest, the GOP has embarked on the most aggressive campaign of voter suppression in decades, a hallmark of a coup. Should that actually happen unions across the country are gearing up to shut things down. Labor leaders in my state, Vermont, not typically a union stronghold, are sidestepping the AFL-CIO leadership and taking a strike voteits members that they are confident will pass. They’re not alone.
It shouldn’t come to this. But it is encouraging to see a level of preparedness from the only corner of civil society with the power to stop a coup. Workers definitely need their own party, but until then, they can still act like they already have one.
Jamie McCallum is professor of sociology at Middlebury College and the author of Worked Over: How Round-The-Clock-Work Is Killing The American Dream published in 2020 by Basic Books.
Gracias! I thank Jenna LaTour for helping research this little rundown, as well as for helping me and Caitlin Barr build the survey. Cooper Kelley helped me analyze it.