Sean O’Brien wasn’t supposed to be here. Three years earlier, he had been elected president of the International Brotherhood of Teamsters, a union known for its bare-knuckle politics. But in mid-July 2024, he stood in a packed hall about to address the national convention of one of the nation’s two major political parties. That in itself wasn’t terribly surprising; unions were enjoying a wave of popularity, and so could rightly have expected some opportunistic speaking invitations. The twist, however, was that the invitation didn’t come from the party you would expect. O’Brien had asked both parties for a primetime speaking slot, but only one had responded—the GOP. So here he was, one of the leaders of America’s biggest unions, about to address the assembled faithful of a party known for its historical suspicion of unions.
What followed was one of the political season’s most bracing speeches. For fifteen intense minutes, O’Brien thundered against what he described as a culture of corporate corruption. He accused corporate leaders of trading patriotism for profits—consolidating power while outsourcing jobs. He said that they had aggrandized themselves at the expense of their neighbors and their nation. And with the help of their paid assistants in Washington, they had sold out the country’s most important stakeholders—American workers. “Never forget,” he said, “American workers own this nation. We are not renters. We are not tenants. But the corporate elite treat us like squatters, and that is a crime.”
These themes were familiar to American labor. They would have sounded perfectly natural at any union rally. But even so, they drew stark reactions from all parts of the political spectrum. On the right, traditional GOP constituencies expressed distaste for O’Brien’s inflated rhetoric. Many in the room sat silent; others got up and left. And the welcome was no warmer outside the convention hall. Writers at the conservative National Review fretted about the Teamsters’ record of left-wing advocacy. And the Federalist Society was so alarmed that it staged a forum on the future of conservative attitudes toward big labor.
On the left, the angst was even more acute. Labor leaders questioned the wisdom of lending credibility to the GOP. The party had, after all, opposed nearly every pro-union reform for a generation. Less restrained, some pro-labor writers ridiculed O’Brien for falling into a trap. He had been duped, they said, into lending his name to a faux-populist play for working-class votes. And perhaps most disdainful of all were members of O’Brien’s own union. Some of them suggested that by appearing on stage, he had lent tacit support to a “racist” and “misogynistic” agenda.
Yet for all their vitriol, these criticisms were noteworthy for what they didn’t say. Almost none of them blamed O’Brien for lending his support to a political party as such. It didn’t seem to occur to anyone that he should have held himself aloof from electoral politics. Indeed, it would have seemed futile for him even to try. In 2024, there was no such thing as a nonpolitical union leader: politics and unionism had become synonymous.
Start your day with Public Discourse
Sign up and get our daily essays sent straight to your inbox.That change has consequences for American labor. O’Brien’s appearance was notable because it was unusual: most major labor unions still overwhelmingly support progressive candidates and causes. Their members, however, are increasingly conservative. They also increasingly say they want unions with no role in politics. That divide has contributed to declining membership and even constitutional tension. If not addressed, it could be the final note in labor’s long, slow swan song.
From Voluntarism to Partisanship
Politics and labor weren’t always linked. At the turn of the twentieth century, unions were avowedly apolitical. Under the leadership of Samuel Gompers, first president of the American Federation of Labor, unions adhered to the doctrine of “voluntarism,” a rejection of politics in favor of direct industrial action. Gompers believed that workers could never achieve lasting gains through politics. Politics, he thought, was the road to dependence; and dependence was antithetical to his vision for the labor movement. Instead, he thought, workers had to rely on their own strength. They had to win better conditions on their own. And they could do that only if they banded together under the aegis of a voluntaristic union.
Voluntarism wasn’t just an ideal; it was a practical strategy. Gompers thought that while unions could survive without state paternalism, they could never survive without industrial partnership. He knew they needed to secure jobs for their members, which meant they had to work with business. He therefore actively sought partnership and accommodation. Rather than regulate businesses into submission, he would cajole them into cooperation. He would find willing employers and strike mutually beneficial deals: industry and labor would reach their own accommodations.
But voluntarism has been dead for a long time now. Far from rejecting politics, modern unions now fully embrace it. They write policy platforms, endorse political candidates, and fund electoral campaigns. They have their own political action committees. In fact, many now even see themselves as primarily political actors. They think they can score more victories at the ballot box than they can at the bargaining table. And they put their money where their mouth is, spending six dollars on politics for every four dollars they spend on anything else.
That shift in focus has changed not only union tactics, but also the very meaning of unionism. Unionism now goes hand in hand with a suite of political views. It embraces positions on taxes, climate change, homelessness, and a swath of other social and economic issues. To take one recent example, in California, large healthcare unions are funding a statewide ballot initiative to tax billionaires. Similarly, multiple unions have sued the Trump administration over its efforts to shutter agencies like the Consumer Financial Protection Bureau. And a growing number of local unions now pursue “common good” bargaining—a strategy that promotes not only higher wages but a particular idea of social progress.
Whether these positions are right or wrong, they plainly stray far from unions’ traditional territory. They connect to the workplace only tangentially. Yes, a billionaire tax might help shore up a state’s budget, which might in turn protect unionized state jobs. But a billionaire tax isn’t just about unionized jobs; it’s about a specific view of good public policy. It’s about politics. And as unions’ political activity has taken center stage, it becomes harder and harder to see where unionism ends and politics begins. Increasingly, they are the same thing.
Politics, Expression, and the First Amendment
That change isn’t just an interesting sociological phenomenon; it also has constitutional significance. In Janus v. AFSCME, the U.S. Supreme Court held that the government cannot force public employees to fund union activities. It reasoned that public-sector bargaining necessarily requires bargaining with the government. Bargaining with the government implicates public policy, which makes it political. And the government cannot force people to associate with political positions. So likewise, it cannot force them to support public unions’ activities.
Yet today, federal law does something similar to private-sector employees. The law channels collective bargaining through a concept called “exclusive representation.” As the name implies, exclusive representation gives a union the exclusive right to bargain for wages, hours, and working conditions. It also denies an individual employee the right to bargain for herself. In effect, it forces her to delegate her bargaining authority to a union—an organization that, again, is increasingly associated with a discrete set of political viewpoints. The result is like requiring people to register their pets with PETA or their firearms with the NRA. It forces them to associate with a politically active group with a distinctive political agenda.
Janus would seem to condemn that arrangement. But so far, lower courts have refused to take Janus’s logic that far. They’ve rejected challenges to exclusive representation by relying on older, public-sector precedents. They say that whatever Janus might imply about exclusive representation, the older precedents continue to allow exclusivity in the public sector. And until those precedents are overruled, exclusivity remains locked in—both for public and private workers.
As unions’ political activity has taken center stage, it becomes harder and harder to see where unionism ends and politics begins. Increasingly, they are the same thing.
Divide and Exit
The losers in this fight are union members. It is no secret that union members have long been more conservative than union leaders. While union leaders historically aligned themselves with the Democratic Party, a substantial proportion of members have always identified as Republicans. And that proportion has been steadily growing. In 2016, Donald Trump won 43 percent of voters in union households—a ten-point jump from the previous election. And in 2024, that figure ticked up to 45 percent, bringing support among union households nearer to political parity.
But at the same time, union leaders have remained steadfastly progressive. According to a recent study by the Coalition to Protect American Workers, the ten biggest labor unions still spend 94.8 percent of their political budgets on progressive candidates and causes. The SEIU tops the list with 98.94 percent. Just behind it were the American Federation of Teachers, AFSCME, and the United Food and Commercial Workers, all of which spent more than 98 percent of their money on the Democratic side. And even the Teamsters, whose members largely backed Donald Trump in 2024, spent almost 92 percent on Democrats.
This divide undoubtedly hurts union recruitment. Unions currently represent less than six percent of the private-sector workforce. And surveys have shown that one of the biggest barriers to expansion is political activism. By a three-to-one margin, potential members say that they would prefer a union that focused on workplace issues rather than politics. And they cite politics as the number one reason they do not join.
Given that attitude, one might expect to see an exodus of current members too. But leaving a union is hard. Once a union organizes a workplace, the law allows workers to leave only when they win a “decertification” election. To do that, they first must organize themselves against the union, gather enough signatures, and navigate the minefield of a federally supervised election. And even if they win the election, they still may not get what they want. Many workers simply want a union that focuses on workplace issues, not politics. But decertification doesn’t get them a better union: it leaves them with no union at all.
It doesn’t have to be this way. The law could give union members an easier way out. It could do away with tactics like “blocking charges,” which allow unions to slow down decertification elections by claiming that an employer committed some minor misconduct. It could outlaw “no raid” agreements, which prevent unions from competing with one another for new members. Or it could discard exclusive representation and allow employees to bargain for themselves.
Labor unions should welcome these changes. They should recognize that things are not going back to the way they were. Unions are not returning to voluntarism; Sean O’Brien and his peers are not backing away from politics. Nor are union members likely to accept their leaders’ reflexive support for policies they dislike. So what unions need is more internal flexibility. Workers need a way to voice their displeasure with union politics. And if voice isn’t enough, they need a realistic way out. If they don’t get it, the system will eventually face a constitutional reckoning.






