Note from the editors of Public Discourse: This week, while our staff takes a week off between Christmas and the new year, we are showcasing past conversations we’ve had with leading intellectuals. Oren Cass and Brad Littlejohn’s discussion was originally published on September 26, 2020. Enjoy!
Brad Littlejohn and Oren Cass recently sat down to discuss the future of labor unions in the United States. Below is a lightly edited transcript.
I was very excited to read the new statement from American Compass explaining why conservatives should ensure workers a seat at the table. Now, conservatives are not generally known for being friendly to labor unions, but a lot of this is due to the rise of public sector unions, whereas you’re looking to reinvigorate private sector unions, right?
Start your day with Public DiscourseSign up and get our daily essays sent straight to your inbox.
Yes, that’s exactly right. And I think it’s an incredibly important distinction that historically everyone understood. I mean, Franklin Roosevelt famously said that public sector unions don’t make any sense.
Over the last few decades, the US has seen public sector union membership growing and private sector union membership plummeting.
Why have private sector unions done so poorly in recent decades, and what do you think are some of the harmful consequences that have come from this weakening of organized labor?
Private sector unions have been in such terminal decline because the legal framework we have set up for them is a very bad one. Michael Lind makes this point: that we now have a lower share of the private sector workforce organized than before the National Labor Relations Act was passed during the Great Depression. So, in a sense, you could say labor law is now worse than useless. You might very well have a higher level of organizing just with no labor law at all.
There are a number of serious problems in the way that we’ve set up the system in America that I think have led to its decline, most of which come down to the way that we structure the relationship between the sides at the enterprise or firm level. We have what’s called enterprise-level bargaining, and Americans just take for granted, “Oh this is what labor means.” You have an organizing campaign in the workplace, and the union campaigns to get a union, and management campaigns for no union, and the workers vote. And either a majority say, “Yes,” in which case now you’re unionized and everyone’s in the union and you bargain, or the majority say, “No,” and then there’s nothing.
First of all, that system obviously creates an incredible amount of adversarialism right off the bat. And then, where the union does succeed, it can be a pyrrhic victory, because the union’s success almost by definition is going to weaken the employer in the marketplace. So you’re typically going to see a decline in the health of those firms, and you’re going to see capital and jobs moving away from those firms and those regions where unionization is high to other places where it’s lower.
We’ll come back to this point about enterprise-level vs. sectoral bargaining at the end. But what I hear you saying is that our system of organized labor is crumbling more from built-in design flaws than because of some coordinated opposition. Is that fair?
I think that’s generally right. There’s a fair critique from the Left that the Right has conveyed a very aggressive, anti-union message. And certainly, corporations work incredibly hard to make sure that unionization does not happen. That being said, at the end of the day, there is a vote.
If the arguments against unionization were obviously unpersuasive, and the benefits of having a union were obviously huge, you would expect to see organizing be more successful. It’s not. There was a fantastic survey done by Richard Freeman called “What Workers Want.” He asks workers whether they would prefer a situation where they have an adversarial relationship with management and some power, or a cooperative relationship with management but no power. And they choose “cooperative with no power” by two or three to one, including workers in unions. In my mind that underscores the degree to which this model that we’ve carried forth is not a substantively wise system and is not the system that workers want either.
Let’s talk about why conservatives have historically been so hostile to the idea of organized labor. It seems to me that part of it is just partisanship: because we associate it as, “This is something the other side is for, and so we don’t know entirely why we’re against it, but we are.” That becomes self-reinforcing, because as the Right defines itself as anti-union, unions tend to support liberal candidates. Then the Right sees the unions as a force for the Democrats, so we oppose them all the more.
I think part of what you’re trying to do is to say, “Let’s de-partisan-ize this issue. If we don’t want unions to be predictably supporting Democratic candidates, then let’s offer a model of unions that Republicans can enthusiastically support.”
That’s very well put. If you go back to the origins of organizing in this country, though, Samuel Gompers famously insisted on the idea of unions as politically non-partisan, and some of the first labor laws actually started out as Republican initiatives. What has happened more recently, and has been compounded by the shift towards public sector unions whose priority truly is bigger government, is that labor has become a partisan, political issue. As a result, it makes sense for Republicans to say, “Well, they’re against us, so we’re against them.”
What we have to do is separate the conceptual opportunity from the political conflict. Because there is at least as much if not more for conservatives to like in what a well-functioning system could deliver.
I think it is fair to say that there are some substantive conservative critiques that have been leveled against labor unions. We could probably put those into two categories, even if people often tend to blur together both sets of arguments.
One would be practical arguments: “Here’s the way in which labor unions fail to deliver economically, and here’s how they actually hurt businesses and workers.” Those flaws might be addressed with better labor law, as we’ll get to in a minute.
The other set of arguments is more philosophical: that there’s something intrinsically anti-free market about the idea of labor unions, that they necessarily interfere with the freedom of contract, and that they distort supply and demand and artificially raise wages.
What would you say to the conservative who opposes organized labor on the basis of ideas about freedom of contract and a view that imposing a labor union on the employment relationship tends to inherently distort the proper workings of the free market?
I think the principled objection comes in a couple of flavors. One is from the worker who says, “Well, I don’t want to be a member of the union.” It’s important to recognize that even in the US, we don’t force workers to be members of unions. The question is whether we’re going to force workers to pay dues if they are in a unionized workplace.
The reason behind that, of course, is that otherwise you have a free-rider problem. The union is negotiating benefits on behalf of workers, some of whom aren’t paying, so the only fair way to do it is to require everyone to contribute.
Yes, precisely. The argument is that if you want to have a system of labor along the lines of what we have—and you don’t have to—then you’re going to have to say that people who work in unionized places have to pay dues to the union.
Working in a unionized workplace can be viewed as part of the terms and conditions of employment. No one is forcing you to take that job or work for that company, but the contract being offered to you is one that has been established in negotiation with the union, and one that requires you to pay dues.
A separate objection that gets raised is that the actual infringement here is on the employer: that I as an employer want to go out and offer a certain set of terms and conditions or contract, and you are telling me that I may not do that because the workers voted to unionize, and so I must negotiate with them.
Someone like Richard Epstein at the Hoover Institution would say that workers should be free to unionize but employers should be equally free to fire anybody who tries to join a union. Whereas in our labor law, one of the core premises for making the whole thing work is to say, “No. Workers have a right to join a union, and you may not discriminate against or fire someone because he is a union member or trying to organize a union.” Asking the question, “Well, why should employers have to operate within this kind of framework?” is the core libertarian objection, and one that on its own terms I think is quite fair.
But the reason I think it fails is that we’re not setting this up in a vacuum. We’re setting this up in the context of the massive range of rules and regulations that both attempt to manage the employment relationship and attempt to create opportunities for employers in the first place—everything from the corporate form to intellectual property protection to the hundreds of billions of dollars we spend every year on educating the soon-to-be workers before they reach the workplace to all of the infrastructure that we build.
In other words, it’s naïve to say that we can take one piece of the market and treat it as an idealized free market that just spontaneously takes shape, when in reality the market that we have is a market that has been shaped in a way that fosters the interests of workers in certain respects and fosters the interest of employers and corporations in other respects. If you want to be a principled libertarian and deprive workers of certain protections, then you probably ought to also deprive employers of the protections of a limited liability corporation.
Exactly. If you want to go and try to start your company in the middle of a jungle or a desert somewhere and fire anyone who tries to join a union, have at it. But the social contract under which you are operating your enterprise in our country is going to be the one that comes with a set of rules that sets up the parties within the market to reach good agreements and deliver good outcomes for the nation.
Now, one of the ways in which labor law is trying to set up those rules of the market is by trying to recognize that the kind of idealized supply and demand curve of the employer’s need for labor and the worker’s need for a wage isn’t usually accurate outside of textbooks, because there is an asymmetry of power between the employer and the employee. The whole premise of organized labor is to say, “We do want a free market in which the worker is compensated at the fair market wage, but that fair market wage is only arrived at if the two sides have equal bargaining power. And the only way they have equal bargaining power is if workers can organize.”
Right, and it’s important to emphasize that this is a decidedly non-partisan view. Go all the way back to Adam Smith, who wrote about the idea of labor’s disadvantage, including the fact that the worker typically needs to work and earn the next day in a way that the employer has sort of greater reserves.
John Stuart Mill, who’s a favorite of libertarians, came to the exact same conclusion. In fact, he questioned the morals of anyone who didn’t want to see workers attaining the power necessary to get the best wage they can.
There are obviously economic constraints here; you can’t say, “Well I want everyone to earn $80 an hour.” But those economic constraints set an envelope within which there is significant wiggle room and a question of, “To whom are the gains going to be allocated?” between the owners of a firm and workers, and beyond workers of different types. Ensuring that workers have power and can bargain on equal footing should be a core principle for setting up a well-functioning market economy.
So, in principle, a well-designed system of organized labor simply equalizes the playing field so that employers and workers are negotiating on fair terms. But in practice, many people would say that the way it’s been done here in the US does tend to result in labor unions that are able to make extortionate demands, so that their companies become uncompetitive and inefficient. Is that objection valid? If so, how did we screw this up?
There are certainly cases in which it’s valid. I think the way that we screwed it up is two-fold. One is, as we talked about earlier, we set up this bargaining at the company level, where we essentially ensure that whatever firms make bargains most advantageous to labor are going to lose. Yet from labor’s perspective, the push is always to get the best bargain you can.
Labor also developed some bad habits in an era when the US didn’t really have any competition. If you look at how the UAW and the automakers were acting in the 50s and 60s, if you were the UAW, you might think, “We don’t have anything to worry about. There’s no alternative to getting cars from these companies.” As that reality changed, the UAW obviously did not change with it.
The second problem is that we’ve established a system where any sort of government regulation of employment sets the baseline, and anything you negotiate for then has to go above it. Over time, we’ve taken more and more of the things that a union might have negotiated about in the past—minimum wages, overtime, safety, leave—and we put that all into employment law. So now if you’re unionized, you have to think of a new set of things above that to ask for.
I think that gets at one of the most interesting parts of the American Compass statement on why conservatives should support unions: that it’s actually a way of achieving a conservative goal of limited government. It seems like somewhere between the 1930s and the 1970s, we got confused about what the goal of the labor movement was. It might make sense to say, “Workers are exploited. Therefore, the government ought to pass laws to protect them against exploitation,” or to say, “Workers are exploited. Therefore, the government should pass laws that enable them to organize to achieve better outcomes.” It seems redundant to pursue both goals at the same time. But, in fact, that’s what we did. So now, labor unions are negotiating over a terrain that has already been extensively regulated by the federal government, which generates lots of inefficiencies.
So, what you’re saying is effectively that if we think there’s too much government regulation, then the authentically conservative solution is not to say, “Well, let’s just try to operate a landscape of isolated individuals jostling in a competitively economic marketplace.” Instead, we should say, “Let’s create institutions of countervailing power so that where exploitation is happening, the people themselves are equipped to resist it, and the government doesn’t need to intervene to fix it.”
That’s what really excited me about your proposal.
The word “institutions” is exactly the right one there. This is a way in which genuine conservatism differs so much from what we tend to call conservative economics, which is really just economic libertarianism. A genuine conservative understanding of markets says that, yes, markets are the mechanism that we want to use to mediate private sector activity, but for markets to work well, they need to be working within the context of institutions. They need to be governed by relevant rules. Most importantly, alongside the corporate form and the marketplace, you need to have the family, and you need to have the right education system, and you need to have the right system of labor.
It is the interaction among those institutions, particularly as they are able to evolve and strengthen over time, that is going to yield good outcomes.
This whole point about reinvigorating other institutions is another important conservative value. You make this point in your statement: that reinvigorating organized labor has knock-on effects that go beyond its mere benefits within the marketplace. One of these is, of course, the fact that if workers are able to earn better wages, they’re more likely to be able to support families on a single income, which used to be a big priority for conservatives.
But you also make the point that various forms of organized labor can provide context for community identity, which is one of the things that is so missing in our society today.
You enumerated the two elements of this nicely: that one way labor can really support family and community is by delivering better economic outcomes. But economic outcomes aside, labor itself is an incredibly powerful force of community. Not by coincidence are labor unions often called Solidarity. Like, that’s the name of the union!
Unions provide a lot of these other features that are important for a healthy civil society in terms of establishing communities of mutual interest and facilitating mutual aid, giving a voice in the public square and in the workplace.
One thing that well-functioning unions do very well is take an interest in prospective members. They help to build the pathways out of high school, for instance, into training, and then into jobs. When you step back and ask, “What are communities missing or what has disappeared from our communities that policymakers have any possible influence on?” Worker-based organizations that promote solidarity and support are a no-brainer and a critical place to start.
Ok, so you’ve made the case for why conservatives should like unions in principle, and we’ve talked about why, in practice, unions have often failed to deliver a good deal for American workers, employers, or consumers. In the American Compass statement, as well as your book, you point toward some solutions that could give us the benefits of unions without the problem—“more perfect unions,” as you nicely put it. Let’s start with the idea of sectoral bargaining. You’ve made the case that workplace-level bargaining tends to put unionized workplaces at a competitive disadvantage, so this can be solved by structuring unions that can bargain with entire industries or sectors. Can you give me some examples of where this has worked well? How easy would it be to make a change like this?
Sectoral bargaining is actually the norm in most western democracies, and northern Europe in particular. It’s our American system, with workplace-level bargaining, that is an outlier. Just to elaborate briefly on what sectoral bargaining is—the idea, as you said, is that the adversarial negotiations over economic terms and conditions in the labor market occur at the sectoral level—for entire industries and geographies, or in some cases for the entire national economy. This happens between representatives for employers, like trade associations, and unions that represent workers in an industry. The agreements they reach then govern the sector as a whole, regardless of whether individual firms are unionized or not. Effectively, they are playing the role that firm-level collective bargaining agreements play in the America, but then also can play the role that a lot of federal employment regulation has to play here.
Establishing a system like this certainly isn’t easy. But there are plenty of places to start—for instance, by choosing particular geographies and industries within this country that already have elements of these processes, and building from there; or choosing particular issues, like the minimum wage, that would be much better set on a sectoral and regional basis than with a national “Fight for $15” campaign.
“Works councils” are another model for which you’ve looked to the successful example of other countries, with organizations that can work alongside a cooperative employer to set mutually beneficial policies. This sounds great in theory. Does it work in practice? Do companies actually listen to workers speaking through these kinds of organizations?
Works councils are prevalent especially in Germany, though other countries have analogs. As I mentioned earlier, it’s not just management but workers as well who say they would prefer a cooperative relationship to an adversarial one. There’s unquestionably value for all sides in worker representation and open communication, even in the absence of economic power or formal negotiations. So, as we talk about a process of transition and evolution, I think one move we should absolutely make now is to allow new kinds of cooperative relationships beyond the traditional union, whereas our laws today say you cannot set up a works council unless you already have a union.
But certainly, a cooperative employer-employee council in isolation is only going to accomplish so much. They make particular sense in conjunction with sectoral bargaining, where the big economic issues get addressed at the broader level, and then a works council can focus on workplace governance and conflict resolution.
One of the issues we talked about earlier is that with a traditional labor union, the only way to avoid the free-rider problem is for all workers at a unionized workplace to pay dues, whether or not they want to. Now, it seems like the ideal solution to this would be to construct labor organizations that workers want to be members of, because they offer other benefits for members. The key here would be loosening up regulations so that labor “co-ops” could take lots of different forms, and begin providing some services that currently the government tries to provide, like unemployment benefits. Tell me a bit more about what these might look like.
The idea of labor unions as social service providers is often referred to as the “Ghent System,” because it initially had significant success in Belgium, but this is also a structure that is present in various forms in many other countries, and even in a few industries in America. In this system, unemployment insurance is something that unions—typically subsidized by employers and the government—can provide to members. More broadly, unions can be a great way to provide so-called “portable benefits” that are attached to the worker rather than to the employer. This is something the Freelancers Union in America has worked to establish. Unions are also an ideal leader for training programs, and they typically produce better outcomes than government training programs undertaken in isolation.
Once again, though, this makes most sense in conjunction with structures like sectoral bargaining that move away from the idea of a union as something you’re either part of or not based on whether your co-workers voted yes or no in a contentious organizing campaign. The point I keep returning to is that it’s really important to think about labor as an institution and the legal infrastructure that supports it as a complex system that has become deeply dysfunctional in this country and requires quite broad reform.
That doesn’t mean a single 2,000-page bill in Congress is going to solve this. Rather, with an eye toward how a healthier system might work, we should start making progress in that direction by opening up new institutions that can serve workers, new roles they can play, and new relationships they can establish with employers.
Thanks, Oren, this sounds like quite the challenge you’ve taken on. You have convinced me that the project is an important one, and I wish you the best of success!