Rebuilding the Working-Class Base One Door at a Time
Reinvigorating class-based politics in the US depends on more than inspiring candidates like Bernie Sanders: it requires durable working-class political organization. Here’s what one group learned about organizing working people around bread-and-butter issues.
- Interview by
- Jared Abbott
Working America is among the largest member-based working-class organizations in the United States, with over 3.5 million members and thousands of local community activists across the country. The organization employs sustained contact with voters through in-person canvassing and digital engagement over the course of years — rather than just before Election Day. To date, Working America has mobilized millions of working-class Americans to take action in defense of their economic interests and to vote for candidates who will advance a working-class agenda.
We spoke with Working America’s executive director, Matt Morrison, to better understand the keys to his group’s success in building working-class consciousness and organizational capacity on a scale most on the Left can only dream of. Morrison highlights many important lessons Working America’s experience can impart to the broader left, as we struggle to build up the foundation of historic yet halting advances from the presidential campaigns of Bernie Sanders to the recent triumph of Amazon warehouse workers in New York and beyond.
Building organizing capacity and class consciousness among working Americans is a critical task if progressives and socialists hope to build majoritarian coalitions to support their candidates as well as cultivate a new generation of skilled organizers who can help deliver gains for working people on the shop floor and electorally.
While many on the Left talk about the importance of this kind of working-class “base building,” Working America is one of the only organizations in the United States focused on mobilizing working-class Americans on a mass scale around class-based issues outside of the workplace. Why do you think that is?
First, organizing takes money and organizing at scale takes sustained investment. We work at the scale we can resource, and we focus that scale on evidence-based strategies. Second, the evidence says that organizing first around class — our mission — both impacts the immediate economic goal and makes people more receptive to subsequent asks, like voting for a pro-worker candidate.
To the question how we fund the work: largely unions, rich people, some foundations, and member contributions. But each has its limits. Unions have resources, but they also have to fund organizing, member services, politics, and a range of other things. Investments from unions fund worker organizing, but those resources are finite. Rich people motivated by elections fund work that advances those interests. Foundations support education.
We need to meet the primary interest of each type of funding source, which can be a bit of an organizing salad. But we have learned that we are most effective advancing those interests among our members and people whose personal economic interests have been advanced by Working America. So what makes Working America distinct is that we are able to organize around the direct interests of working people and really focus on that. We are not a campaign for a cycle; we don’t have to bargain and service contracts like a union. We are able to engage a broad base consistently over time and do so with a real evidence-based approach.
For example, in our black worker unemployment insurance project,we increased the number of workers who successfully applied for benefits by 32,000 people, at a point in the pandemic where state unemployment agencies were not getting people through and against a backdrop of the last recession when just one in four eligible black workers received benefits. Taking an action in one’s economic self-interest, whether it is helping someone fill out unemployment insurance applications or seeking rental assistance, changes the action taker’s perspective. To do that at scale is powerful.
So for instance, we’re super effective at convincing people who we just helped pay their rent, navigate a byzantine marketplace in the health care arena, understand their options for vindicating wage theft, abuse, and so on. Those people listen to us when we say, “Oh, by the way, this candidate is on your side, and this candidate is not.” So we win votes, but we win votes by serving the economic interest of workers or helping them build agency.
But that approach also has its limitations. I mentioned our 3.5 million working-class members. Their dues are an important but small part of the funding mix, about $1 million a year. What we don’t see is the type of broad working-class funding base that we see in high-profile elections that supports this type of on-the-ground power building of the working class. There are millions of us willing to give $25 a month for candidates, but there is a disconnect when it comes to institutionalizing class power.
The only way to truly scale this type of work where the singular focus is building working-class power is for working-class people to pay for it.
Could you talk a bit about the scale of your work, in terms of where you operate, in how many states, the kind of staff you’re working with, the volunteer base you can count on, and your organizational budget?
Yeah, absolutely. At the beginning of the year, Working America had about 3.5 million members. By Labor Day, we will be at 4.2 to 4.4 million. So we’ve got a lot of growth in front of us, and we’re staffing like crazy to try to meet that goal: by the end of the year we hope to have 1,500 skilled organizers on the ground, not to mention nearly ten thousand volunteer activists distributed across the country. That is one of the advantages to our organizing model. When we talk to working people and ask them to join and take an action, they are making a values decision at their front door. Dues are voluntary — $5 per year. So a lot of the traditional constraints to growth, like getting someone’s attention in a crowded digital environment or overcoming the risk to your job when the action comes in the work site, are not factors for our work.
Once we hit our membership target, we’re going to turn back to those people we just organized as well as a portion of the existing membership base. The goal is to do around 1.2 million face-to-face conversations just for this election period. That’s a two-and-a-half-month period from Labor Day to Election Day.
In terms of where, we have full boots-on-the-ground canvass operations in Arizona, Georgia, North Carolina, Pennsylvania, Michigan, and Minnesota. In addition to that, Nevada, Ohio, and Wisconsin have our full digital organizing suite.
Our intention in terms of voters, individual human beings who are making different choices on the ballot, is going to be in the ballpark of 600-700,000 voters who have changed who they’re voting for.
To give an example of what our work looks like on a given day, tonight a couple of hundred professional canvassers will go out around the country. They’re going to knock on about 20,000 doors in places where we have offices. They’re going to get somewhere in the ballpark of around 6,500 people to open the door. They’re going to from that get around 4,000 or so of those people to sign up.
It’s a really light-touch mechanism. It’s a five-minute conversation. People appreciate the convenience of it. We’re not collectively bargaining for them in the workplace, but we are collectively bargaining in the political economy for them. That’s the value that they get.
What sorts of people do you target in your organizing work? Are you primarily focused on engaging low-propensity Democratic voters, swing voters, or some combination?
It sounds a bit counterintuitive, but our organizing targets tend to be higher-propensity voters — people with 70-80 percent turnout probability in a midterm election. Despite being regular voters, however, they are part of that vast 65 percent, 70 percent of the electorate who don’t pay attention to politics. They’re not into the sport of politics, and since they’re not committed to whatever their politicized information source is or dialogue that they’re in, they’re very receptive to getting new information. And a little bit of new information changes their mind. In other words, what we’re doing is we’re finding a voter population that is likely to vote already and maximizing that.
Do you tend to reach people who self-identify as Democratic or Democratic leaning, are they primarily independents or Republicans, or is it a mix?
It’s all over the place, but the partisan breakdown for folks tracks largely to the demography of where they’re coming from. In Southeastern Pennsylvania, they’re certainly people of color in the urbanized core, and it gets pretty diverse in the inner-rim suburbs, and then it gets pretty white the farther out you go. Partisan identity kind of tracks that, but that partisan identity is not a strong predictor of how receptive people are to change. We’ve all talked about the number of Pennsylvania Democrats who love Donald Trump, for instance.
Got it, but isn’t it still safe to say that you engage a huge number of voters from outside the Democratic base? For instance, Working America has reported that its members increased in their support for Democrats by 12.2 percentage points from 2016 to 2020. And since you’ve said that it wasn’t primarily people that were going from nonvoting to voting, you must have moved a lot of people who might have voted for Trump in the past into the Democratic column.
Yeah. I mean there’s people who voted for Trump but also people who voted for Jill Stein or Gary Johnson. To give an example, it’s pretty likely that what’s happening is we’re taking a white guy who didn’t go to college who earns $35,000 a year and lives in Ohio, who by all accounts would be voting Republican, and giving him a little bit of economic context. Therefore, he’s shifting along with his economic interests in a way that his neighbor is not. That’s probably what’s happening.
Can you say a bit more about how this persuasion works in practice?
One factor is the importance of the member concept, going and asking someone to join, getting them to internalize the decision. That’s critical. Two, you have to get them to take an action on something that’s relevant to them; that’s the iteration of the relationship. Three, you have to be an agile organizer. You can’t make it about your agenda. It’s always got to be about them. Those are the keys to reaching nonpoliticized voters.
Testing is also essential. We’re not scared to fail, and so we’ve done a lot of testing. When you do so much testing, you get better at it. So we have stronger hypotheses. We have predictive modeling for this validating out over lots of different cycles and contexts that says, “Okay, down to the door of the actual individual, this person is persuadable, not persuadable, or neutral.” We can sort the entire country in that way and prioritize our energies toward people who are going to be responsive. That same evidence-based process iterates across all of the outreach objectives.
Finally, it’s important to note that the goal is not to have long conversations on the doors. We should be done having an initial conversation with someone in five to six minutes. What’s important is not the length of our conversations but sustained contact over time. It’s key that we develop some kind of relationship with you in January before we show up at your front door in October during the election campaign. So it’s the distance in time and the consistency of engagement over time that matters, not having a long conversation.
So when you ask about persuasion, yeah, we want to win the election, but to really be effective we need to think in much broader terms — we need to change the wealth distribution in this country so that working-class people share in the prosperity. What that means in practice is that typically the ratio of income held by the top 20 percent versus the bottom 20 percent is around 4.8. What if we could narrow that ratio and get it below four in the Philly metro, the Atlanta metro, the Raleigh-Durham metro, the Maricopa County metro, and so on within five years by helping working people better use the available tools, inadequate as they are, to create economic security, and establish new tools?
If you think about what that means for wealth redistribution down to the bottom of the income distribution, this is huge. It’s going to take a concerted effort, including to continue to push the leverage that lower-wage workers have right now in the job market. That’s the trot we’re moving at, and people can tell, so they’re responsive to that.
You’ve not only made impressive strides in persuading nonpoliticized working-class Americans, but elsewhere you’ve stressed that Working America’s outreach efforts are also highly cost-effective. I read that in 2020, campaigns and super PACs spent an average of $750 to $1,000 per vote whereas you guys only spent about $50. Can you explain how this can be cost-effective given what seems to be a pretty significant staff investment in maintaining a connection with people over such a long period of time? And if your approach not only works, but is also cheap, why aren’t more people doing it?
First, those cost-per-vote outcomes in 2020 did not reflect the typical large-scale organizing staff investment because we did not run door-to-door canvass programs. But even when we account for the traditional organizing staff investment, we are still generating votes at a more cost-effective rate, usually by a more sizable margin than one would expect.
What I really think is driving this is that most people who invest in progressive political organizing don’t do persuasion and instead focus on voter registration and turnout. That is important work, but it is also comparatively saturated in terms of organizing investments and can take on an importance that is more philosophical than tactical.
The best “worst” example that comes to mind is a funder who was so ideologically committed to the philosophy of turnout — to the exclusion of all other tactics — that he asked for a strategy to go win a West Virginia congressional seat by turning out black voters. That’s kind of a fantasy, right? The black population of the state is in the low single digits.
But there are also tactical considerations. Measuring whether someone completes a voter registration application, requests an absentee ballot, or turns out to vote is just an easier task than measuring individual vote choice because we respect the secret ballot in this country.
More broadly, there’s a bit of an allergy to class-based organizing, especially among the white working class, because of the skepticism that anyone who ever voted for Trump, or other Republicans before him, could be swayed, or that to sway them one must betray the interests of black people, people of color more broadly, and the rest of the progressive coalition. Nothing could be further from the truth. The reality is that if you want to win elections in difficult places, you have to be willing to kiss a few frogs, so to speak.
The point is that incentive structures that drive organizing choices across the community have been built up around that. They’re built up around turnout. The incentive structures are built up around identity and not around persuasion based on shared class interests.
In turn, even when groups are interested in persuasion, they often lack experience. There is some sense that persuasion is really the province of TV ads, but the persuasion organizing space is not very crowded. What we have learned is that you can’t become good at persuasion in high-salience contests without a lot of trial and error. We started this strategy at scale back in 2011 in the SB 5 campaign in Ohio to save public sector collective bargaining. We tested our approach using randomized control trials — similar to the way a new vaccine would go through a medical trial — hundreds of times before we got into the 2020 cycle. When you test a lot, you fail a lot. When you fail a lot, you learn a lot, and so we’ve been “in school” for a decade. Most folks who want to do persuasion are new to the game, or they’ve relied on pollsters to tell them what persuasion is and aren’t grounded in causal metrics because polling is easier to understand. Polling is easier to implement than randomized control trials but not as useful in assessing the cause of the failure or success.
So that’s really what I think has distinguished us: we have a mechanism, the canvass, that taught us that persuasion is possible, so we leaned into persuasion and built up good practices around it. We took those practices and applied them to other communications channels that are more competitive in terms of getting voters’ attention, like digital and mail outreach. If you don’t have that mechanism and all your incentive structures are pointed in a different direction and you don’t think it’s possible, you don’t know how to do it regardless of medium.
If persuasion organizing efforts received the same level of attention and investment, or even a third of the level, that GOTV-centric organizing strategies get, we would see that the relative cost advantage diminishes as others begin fishing in the same pool of voters. What’s happening is we are first movers. That’s all. If others start to come in and replicate some of the stuff that we’re doing, our cost per vote is going to go up to $200, but that starts to give you a sense of how much runway there is because people just aren’t doing it. That’s the first thing. And then the second thing, again, is the importance of establishing a member relationship.
Circling back to the idea you were mentioning about the GOTV-focused folks or the people that maybe have somewhat optimistic views of how to win elections without kissing frogs, do you worry at all that the broad, class-based approach that you guys take may be harmful to voter turnout among the base? Maybe progressive outreach efforts should be focused more on issues that might not appeal broadly to, say, white working-class voters, but are key issues for many Democratic base voters, and may help to drive turnout among many working-class voters of color, in particular.
Actually, I haven’t heard that flavor of critique. The flavor of critique I’ve heard is about investments, right? That we shouldn’t spend all this money trying to please someone who doesn’t want you instead of speaking to someone who’s on your team. It’s kind of that. It’s more personalized in that way.
To then reframe the question, what I’ve heard isn’t concerns about our approach being counterproductive. In fact, I don’t think the evidence supports an argument that issues mobilize voters. What causes someone to increase their voting turnout more often than not isn’t the words campaigns use. The message isn’t the thing. It’s the process and understanding. Homeowners vote on local school funding levies that renters often skip because they understand that their property tax bill is on the ballot.
I also think that there is a growing awareness that voters of color are not a given for Democrats. So focusing on persuasion in these communities is arguably as important as turnout in terms of netting votes.
We as a community rally around compelling causes that translate into political causes. For example, Black Lives Matter, so do their jobs, and so does black children being able to afford to go to the doctor.
The things that are economic and social well-being challenges for the working class broadly are especially huge for people of color. Interestingly, Asian American and Pacific Islander (AAPI) voters are the most responsive to our appeals, followed by Hispanic and black voters, then white voters, who are the least responsive. So turning out the vote is important but leaves a lot on the table when we also need to persuade people to join the base.
There’s a subtle realization I’ve only recently come to appreciate. When we look at the Upper Midwest, for example, our persuasion effects are quite large among noncollege white voters. Try that same thing in North Carolina, and we get backlash after backlash after backlash. The racially polarized Southern white noncollege voter is so much harder to reach.
Likewise, the Southern nonwhite voter is surrounded by an air of religiosity, not everyone, but there’s a lot of that. And so the issues for those voters are. . . . You have abortion, women’s rights, and other kinds of conservative value issues. So in the South, your best persuasion targets tend to be people of color by a long shot.
It’s kind of counterintuitive, but if you want to win a net vote for a Democrat, go persuade a black voter in Georgia. That’s going to get you a lot of votes because blacks are a third of the electorate. If you’re at an 85-15 split, then what you’re talking about is what? The four and a half points [among the total statewide vote] that are going Republican. If you can cleave a point and a half of that away, then you’re plus-five points within that community, but statewide, you’ve essentially shifted an outcome by three percentage points.
Huge, right? That’s the ballgame right there: find the most efficient targets for winning elections. Look at their economic contexts. Go and organize them on account of their economic context, and ask them to go vote with you in the election.
Finally, can you talk a bit about where you see the approach of Working America fitting into a broader medium-term vision of progressive change in the United States? What are the limits of the approach? What are the ways in which Working America interfaces or doesn’t interface, say, with the rebirth or regrowth of the labor movement?
In terms of Working America’s role, our kind of vision, what we’re doing, and how that fits in the broader ecosystem, what we know is that there is a deep selection bias when you size up Working America. When you look at all these polls and see that 60 percent, 70 percent of Americans approve of unions, that shows up when you walk through a neighborhood that has a hundred doors.
A third of people are going to say, “Get off my lawn, kid.” Those are the folks who don’t want unions. But everyone else is like, “Oh, they mean accountability for corporations investing in jobs and democracy. Yeah, I’m for that. What do I have to do?”
So what’s really driving the composition of our membership is who can accept the labor movement in a way that they want to literally sign down and say, “Okay, I’m a part of it.” That’s a huge amount of people, and most of them would join a union today if they could. They just don’t have that opportunity.
If we do this right, the only thing limiting our growth becomes just how much money we raise, because there are, just based on the math, 75 million or so people in this country who would join a union today who don’t have one and are part of the working class. We can go and organize all of those folks, and we’d do it if we had the money. We don’t have that type of money yet, but if we keep doing things that are a value-add to different constituencies, we will find a way to fund that.
Ultimately, we’re not going to be the big, independent side of Bernie Sanders. That’s not real, but if you think about that volume of investment and that volume of energy [in Sanders’s 2016 and 2020 Democratic presidential primary campaigns], we also have a big volunteer base that does a lot of our organizing. If we could scale up that investment and scale up that level of energy and direct it, then we could change our politics.
What’s the difference between New York and Florida or between Texas and California? It’s the same democracy, but different people, different attitudes. So we can actually change the attitudes in this country to scale, and that’s really thrilling to me.
I’m an eternal optimist, because if you talk to a lot of people, there are so many reasons for optimism. So I think there’s an inevitability to this; we cannot lose in the long run. We are a sleeping giant, and the problem is we just need to get at the bulk of people waiting to connect.