What Strategy for Labor?

My debate with Slaughter, Olney, and Fong on leverage, unionization targets, and how to scale up worker power

What can we do to turn around decades of union decline? This is the key challenge of our era, because the power of a revitalized labor movement is needed to pull America off its descent into oligarchy and authoritarianism. There are no easy answers, which is why I welcome the responses of union organizer Peter Olney, Labor Notes co-founder Jane Slaughter, and labor scholar Ben Fong to my new book We Are the Union: How Worker-to-Worker Organizing is Revitalizing Labor and Winning Big.

Whereas Slaughter and Olney’s critical-but-comradely reviews accurately address my book’s major arguments, Fong’s dismissive polemic is marred by strawman claims and tendentious logic. Nevertheless, all of their pieces raise important questions about just how new worker-to-worker unionism is; the role and industrial scope of strategic targeting in our contemporary political economy; and the generalizability of the campaigns profiled in the book. I’ll do my best here to address their major criticisms.

Is the Model New?

The starting point of Fong’s critique is his case that worker-to-worker unionism “is not a fundamentally ‘new model’ of organizing.” There’s a lot of truth to this. Indeed, I explicitly highlight the roots and commonalities of recent worker-led unionization campaigns with their historic predecessors in the 1930s and earlier. As I write in the book, “I’m definitely not the first person to make the case that putting workers into the driver’s seat is the key to building a powerful mass labor movement. … We Are the Union makes a new case for an old strategy.”

But contrary to Fong’s claims that I am self-servingly “overhyping a branded ‘model,’” the actual reason I conceptualize worker-to-worker unionism as “new” is that a) the specific organizing practices and structures of today’s worker-led efforts are significantly different than they were a century ago, and b) if labor is going to scale up, it’s necessary to be very specific about what worker-to-worker unionism refers to, since critics wrongly assume it requires romanticizing “spontaneity” and since so many unions continue to frame their unscalable, staff-intensive efforts as “grassroots” or “bottom-up.”

Here’s how I put it: “we need to replace a blurry image of worker-to-worker organizing with a high-definition picture. Without such clarity, it’ll be hard to diffuse a new grassroots model or to counter the claims of its skeptics.”

I define the worker-to-worker model as one in which organizing is relatively lightly staffed, and therefore scalable, because “1) Workers have a decisive say on strategy, and 2) Workers begin organizing before receiving guidance from a parent union, and/or 3) Workers train and guide other workers in organizing methods.” Unfortunately, this cluster of attributes has been exceedingly rare from the 1950s onwards, contrary to Fong’s suggestion that a staff-intensive approach was limited to the 1990s.

Even after the post-pandemic uptick, exceptionally few unions have anything resembling the rigorous worker-to-worker training and structures of the NewsGuild’s national Member Organizer Program, which trains up worker leaders in all the responsibilities that in other unions are typically carried out by staff. It would be a game-changer for the labor movement if most unions began to organize like the Guild.

In order to help make that happen, my book seeks to clarify the urgency of adopting a new model and to specify the practices unions could adopt, including big online organizing training programs; worker coaching of new drives; nationwide “pods” of a dozen worker leaders to share encouragement and expertise; elaborate online systems for workers to track their organizing skills and campaign benchmarks; and the mass seeding of new drives through worker outreach and digital tools.

As I explain in We Are the Union, and as the list above suggests, one of the crucial differences between today’s worker-to-worker model and its historical ancients is that it’s now possible through digital technologies for workers to train, co-strategize, and initiate organizing campaigns with other workers anywhere in the country — tasks that could only be done on a local level until a few years ago. Fong downplays these novelties by quipping that “had communists had smartphone technology in the 1930s, they would have availed themselves of it.” Sure. But it’s hardly a secondary matter to specify how and why unions should transform their organizing approaches in light of major changes in terrain, such as the rise of digital tools or the decentralization of industry.

Practitioners seem to agree with my suggestion that there are new things we can learn from the recent uptick. Over a dozen unions and rank-and-file organizations have already reached out to our new Worker-to-Worker Collaborative at Rutgers University, which was recently founded to help unions and formations like the Federal Unionists Network scale up ambitious, powerful organizing by relying more on worker leadership and new tools.

Finally, even though the idea of putting workers in the drivers’ seat is not new, I do think my case for doing so is unique. Rather than fall into a common labor-leftist trap of treating militancy as a silver bullet, or downplaying the importance of resources, experienced staff, and systematic organizing training, I argue that these need to be deployed in a way that’s scalable — hardly a common theme in the labor-left literature.

Winning Battles vs Winning the War?

Although my book focuses on the question of scaling up workers’ power, Olney’s response has little to say about this issue of scale. His review focuses on the tactics necessary to win specific campaigns, while my argument focuses on the model (division of labor) and methods (mass seeding, trainings, etc.) necessary to win widely. Olney thus confusingly references “Blanc’s point that staffing ratios, while not unimportant, are not determinative.” But my book’s point is that staffing ratios are determinative — if the question is how we can potentially unionize tens of millions.

In that sense, I disagree with Olney’s downplaying of worker-to-worker unionism, which he says is just “one piece of a much larger analytical framework for success in organizing.” To bolster this claim he points to other tactics needed to win union campaigns, such as having strong organizing committees, talented staff, alliances with unionized workers, and roots in communities. I agree with all these tactics, and I explicitly advocate for each over the course of the book. (That said, Olney ignores the huge amount of evidence that community ties have declined across the board in the US since the 1960s, including among people of color. Community support is more important than ever, but it’ll take a lot of bottom-up organizing and persuasion to leverage it.)

But Olney’s listed tactics are for winning battles; they don’t directly address my question of how we inspire and lead enough battles to win the unionization war. To transform America, that’s the key question we need to find answers to. As I write in the book:

While improving the tactical quality of union campaigns is crucial, it isn’t enough on its own to organize a sufficiently large number workers for systemic change. To win the unionization war, the harder question is how to exponentially increase the quantity of unionization battles.

Olney seems to expect that scale will come relatively automatically from improved tactics for winning specific campaigns. But I show in the book why — due to exorbitant time and money costs — if good tactics continue to be paired with a staff-intensive model, labor can’t scale up. In that sense, the staffing-resource dilemma is primary.

It’s also the case that the mechanisms for scaling up via worker-to-worker organizing are the least understood and implemented within organized labor, which is why my book focuses on addressing this missing link. For example, when the Amazon Labor Union (ALU) won its union election in April 2022 at JFK8, almost a thousand Amazon workers over the coming days and weeks reached out to ask for organizing help. Unfortunately, the ALU had neither the capacity nor organizing vision to onboard these workers; failing to initiate any mass training to absorb them, ALU missed an unprecedented opportunity for scaling up.

But it’s not just independent efforts that have failed to seize moments for rapid growth. Despite their deep pockets, established unions have done little to scale up recent union election victories at major chains like Peet’s, Apple, Chipotle, and Whole Foods. Most consequential of all, organized labor has continued to fail to leverage its billions in financial assets to spark countless new drives in strategic industries via wide-scale salting and mass worker-to-worker seeding. As I write in the book, “Why not use labor’s access to a huge amount of voter data to call or knock the doors of every voter who works at Amazon, Walmart, and FedEx to generate leads for ambitious national unionization campaigns at these companies?” Without these types of initiatives to dramatically grow grassroots unionism, good campaign tactics and good strategic targets won’t be enough to defeat Trump, Musk, and Bezos.

Which Targets to Focus on?

Olney, Fong, and Slaughter agree on at least one point: We Are the Union is not sufficiently oriented to the most strategically central workers. Thus while Slaughter notes that the unionization upsurge at Starbucks is “exciting,” she disputes my contention that large service sector corporations like Starbucks are “at the heart of America’s political economy.”

I remain unconvinced of her skepticism. Services now make up over 75 percent of America’s GDP, while manufacturing contributes about 10 percent. And the overwhelming majority of our workforce is also now employed in services. Don’t these economic shifts from a century ago oblige us to adjust our unionization (and electoral) strategies accordingly? If so, how?

In my view, these changes require us now to combine strategic targeting with widespread seeding (more on this below) and to expand our organizing targets to include not just manufacturing and logistics, but also strategic public sector workers such as teachers and service sector workers in big corporations like Starbucks or Whole Foods.

In our decentralized political economy, when thinking about where to concentrate our energies we can’t only ask which workers are most economically powerful. We also need to ask: Which workers are most socially powerful? Social power refers to the ability to galvanize the public and to shape politics. For example, federal workers and teachers have a particularly high degree of social power because of the services they provide. By leaning on such community leverage, teachers have been key to successful labor militancy for the past decade, despite their lack of economic might. And federal workers, at the heart of today’s fight against Trumpist authoritarianism and austerity, will have to do the same.

Even economically powerful workers can lose without sufficient social power. From the 1870s through the early 1930s, for example, manufacturing workers in the US kept on losing most of their biggest strikes and unionization drives, in large part because state repression was strong enough to overcome their economic leverage. It was only after factory workers began receiving unprecedented public and governmental support under the New Deal that they unionized America’s mass production industries.

Finally, there’s a third strategic criteria worth considering: What sectors can grow the labor movement today? This relates both to direct union recruitment and, no less importantly, to inspiring others to self-organize.

Keeping all three of these criteria in mind is key for making smart decisions about targets. It’s not an easy task today, since not all three factors line up in the same industrial sector like they did in the 1930s. But only through such an approach can we avoid, on the one hand, a hyper-concentrated strategy that fails to spark or inspire other key layers of working people, and, on the other, the tendency of some unions to only chase after the quickest way to add dues payers to their rolls.

In contrast, Slaughter counsels a focus on “the larger workplaces” like auto plants “that collectively have the power to take on capital — and thereby inspire other workers.” I agree with her that big manufacturing plants remain a crucial target. Throughout We Are the Union, I highlight the work of the reformed United Auto Workers and I drafted a whole chapter on the Mercedes workers’ union drive in Alabama, scrapping it only after they lost in May 2024 (since the book only included case studies of campaigns that overcame strong employer opposition).

But it’s not the case anymore that powerful workers always work in large workplaces. Some employees with the most potential disruptive power these days normally work in tiny offices or remotely: the tech workers who code and direct the digital back-end upon which all corporations and the government now depend — as Elon Musk’s DOGE has recently demonstrated to the country. You don’t need to be a blue-collar worker to have real power.

Unlike in the 1930s, the largest corporations in the US are now generally dispersed into relatively small workplaces scattered across the country. The obstacle here is not that these workers fundamentally lack disruptive power, but that it’s so hard to scale up to super-majority union density in such large and dispersed companies.

As such, the key organizing challenge at today’s mega corporations is growing as widely as possible — and in leveraging consumer support in the meantime through weapons like brand damage and economic boycotts to help force management to recognize the union and grant first contracts. And this same scalability challenge is also sharply posed at Amazon, the most economically central target of our era. Olney’s references to as-yet-unspecified chokepoints don’t resolve the scale dilemma at Amazon, since, as Nantina Vgontzas’s research has shown, management can relatively easily reroute around specific warehouses or other nodes paralyzed by economic militancy.

Strategic Targeting vs Widespread Seeding?

Given the pervasive challenge of scaling up, I disagree with Slaughter and especially Fong’s framing of strategic targeting as a separate and counterposed task from seeding, i.e. tactics to spread widespread self-organization. If my book tends to emphasize the latter, it’s because seeding tactics like the following tend to be less implemented or understood within organized labor:

From Starbucks to Southern auto, some of the most productive seeding techniques include using high-publicity moments like big union elections and strikes to call on (and provide tools to) other workers to start organizing; holding big online trainings; producing viral social media content to generate new leads; posting digital ads or distributing fliers encouraging people to sign up for organizing support; and developing in-depth, easily accessible training materials for workers to start self-organizing.

UAW strategist Chris Brooks explains in the book why their new leadership sought to utilize such a seeding approach: “We didn’t know — and didn’t want to try to predetermine — where the most heat would be, so we’ve tried our best to fan the flames everywhere.”

In contrast, Slaughter portrays the spread of auto-worker self-organization after the Fall 2023 Big 3 strike as a direct grassroots response to big economic wins. These contract gains certainly were galvanizing. But so too was the masterclass in seeding undertaken by the reformed UAW leadership, which during and following the strike called on (and gave organizing tools and digital content to support) all non-union auto workers to begin self-organizing.

For his part, Fong claims that “the vision conjured [in We Are the Union] is of worker self-activity bubbling all over the country and rising to a boil thanks to throwing off the shackles of outmoded union ideas about strategic targeting.” In reality, the book argues that to organize today’s biggest companies and industries, targeting and seeding have to go hand in hand.

Here’s how I summarized my view of the relationship between these two tasks: “There’s no need to counterpose targeting to seeding, even within the parameters of a single campaign. [Starbucks workers in] Buffalo showed that winning one election by salting a small workplace can set the stage for a broader grassroots organizing spurt—provided that the union is momentum-savvy enough to seize an opening via national social media agitation, mass onboarding, and worker-to-worker training.”

Far from “punting on the question of targeting” as Fong suggests, We Are the Union repeatedly insists that salting is a pivotal tactic to scale up strategic targeting. Though I discuss salting twenty-one times in the book, the term is not mentioned once in Fong’s lengthy review. For instance, I explained that,

I’m not arguing for a return to hot-shopping, in which unions passively wait for workers to reach out and do nothing to transform initial wins into company-wide or industry-wide campaigns. Nor am I suggesting that unions stop targeting strategic sites. This is still a crucial tactic. Unions should put serious funds into efforts .... to widely train a new generation of salts capable of initiating campaigns at pivotal workplaces and companies.

Nor was this rhetorical window dressing, as Fong ungenerously insists. I’m proud to have put a lot of effort into launching the Democratic Socialists of America’s ambitious new salting project Workers Organizing Workers, which anybody looking to salt should reach out to.

We should target Amazon and manufacturing. When we recruit salts, we don’t ask them to go unionize their local bakery. But the harder question, the one I focus on in the book, is how to generate the organizing mechanisms and momentum necessary to spread unionization widely enough to beat the world’s biggest corporations on today’s decentralized economic terrain.

Self-Organization and Targeting

Which workers have the power to inspire others to fight back? As noted above, Slaughter suggests we focus on “the larger workplaces [like auto plants] that collectively have the power to take on capital — and thereby inspire other workers.” But consider the experience of Starbucks Workers United, which has done more than any other recent campaign to inspire large numbers of other workers to unionize. While Starbucks workers have won significant economic concessions, it’s primarily been the example of their militancy and grassroots organizing that has been so inspirational. In times of uptick or upsurge, material gains are not always the primary catalyst for contagious struggles.

A similar dynamic of militancy begetting militancy drove forward the red state teachers’ strike wave, which began when West Virginia’s educators walked out on February 22, 2018. What’s often forgotten today is that West Virginia’s strike lost its major demand — a funding fix to the state’s public health care system — and the demands it did win through striking (a 5 percent pay raise) barely caught teachers and service personnel up with inflation. Significantly, pro-strike Facebook groups in Oklahoma and Arizona went viral in the days before West Virginia’s strike was concluded on March 6. Both groups were launched immediately after West Virginia’s educators caught the news cycle and went viral online by going wildcat on February 28.

Obviously, it’s always better to win than to lose. But it wasn’t the size of West Virginia’s material wins that caught the imaginations of so many educators nationwide in similar states nationwide. Since big fightbacks tend to inspire other big fightbacks, you don’t necessarily need the concentrated economic might of a factory worker to inspire other workers to organize.

Nor does the sequence of mass organizing contagion necessarily pass from the most economically powerful workers to the least. Workers tend to get most inspired by workers in similar jobs to themselves. And because the vast majority of Americans work in the service sector, and because there still exists a pervasive myth that unions are only for blue-collar workers in hardhats, recent unionization efforts at Starbucks have been particularly impactful.

Service-sector effervescence, moreover, can feed into encouraging workers to unionize across the whole economy—including in manufacturing, as I show in my book’s discussion of the Yanfeng auto parts factory in Missouri. As one Starbucks barista texted me earlier last week, “winning a contract is so crucial to keep up the excitement our campaign has brought to the labor movement, and that makes us pretty damn strategic!!”

Even in relatively marginal workplaces like a neighborhood coffee shop or a small non-profit, self-initiated workplace struggles can (in addition to winning significant material gains and a voice at work) get the idea of unionization into the air, boost labor’s momentum, help recreate a class-conscious common sense among working people and, just as importantly, pressure politicians to pass much-needed labor law reform. No matter where you work, building a union with your co-workers is your most accessible onramp into the labor movement, your best way to get a crash course in serious organizing, and your best way to pressure the government to support labor.

As such, even fights in relatively non-strategic jobs can help forge the massive army of worker leaders needed to unionize corporate America’s strongest fortresses. Undergrad student worker unionization drives have relatively little economic weight, but they could be crucial training grounds for future salts — and full-time labor organizers — willing to dedicate their lives over the coming years to unionizing Amazon, auto, and our country’s largest corporations. A similar process took place in the 1960s, when large numbers of young people, including the founders of Labor Notes, got radicalized on universities and proceeded to take jobs in strategic industries.

Fights in relatively small workplaces can also play a crucial role in boosting Labor Notes’ decades-long efforts to reform calcified unions into militant, democratic engines of class struggle. As I show in the book, some of the most successful recent union reform processes, such as in the NewsGuild, emerged when groups of workers first self-organized their shops and then proceeded to push their unions in a new direction. A cannabis worker who organizes her neighborhood marijuana dispensary may not have earth-shaking economic power, but they can play a critical role in democratizing the stagnant United Food and Commercial Workers union.

It’s true that individual unions always have to make hard choices about where to immediately allocate scarce resources. Prioritization is certainly one of the thorniest sides of labor strategy. But lowering the cost of organizing via the types of worker-to-worker mechanisms implemented by projects like the Emergency Workplace Organizing Committee can make it easier to say yes to a much larger number of workers looking to organize. And there’s no reason the labor movement as a whole, with its dozens of national unions traditionally rooted in different industries, can’t try to launch ambitious campaigns in all corners of the economy.

Moreover, we shouldn’t assume there’s always a zero-sum relationship in the medium or long term between organizing in strategic and less strategic workplaces. Slaughter notes that “the new UAW turned to unionizing auto factories, not to the less strategic target of more graduate students, as their lazy predecessors had prioritized.” I agree with her characterization of the old UAW leadership, but even if we leave aside the fact that recently unionized graduate workers were one of the constituencies that helped elect Shawn Fain, the experience of United Electrical Workers (UE) suggests that the relationship between these sectors can be more reciprocal — and that decades of difficulties in organizing manufacturing are more deeply rooted than union “laziness.”

Despite a tireless commitment to militant grassroots unionism in the manufacturing sector, UE’s ranks were decimated from the 1980s onwards, as factories moved abroad and as workers faced with capital flight lost their confidence to fight back. I touch on one such emblematic campaign in We Are the Union: “UE’s Plastic Worker Organizing Committee (1989–93) was a bold worker-driven organizing project with the slogan ‘It’s Time.’ But after years of making little progress, the joke among worker leaders and staff became, ‘Well, maybe it wasn’t quite time.’” Deep structural obstacles like globalization surely have had more of an overall inhibiting impact in manufacturing than “laziness” or the one factor Olney points to: leftists not thinking much about factories once they moved south.

Fortunately, UE’s first major growth spurt in decades came in 2022-23, as the union doubled its national membership through graduate student worker organizing. Had UE focused exclusively on organizing manufacturing workers over the past five years, it’s likely that it would currently be in a worse position to unionize more economically powerful workers. Sometimes armies need to accumulate forces before they’re strong enough to take down their strongest enemies.

Downplaying successes

Olney suggests that recent worker-to-worker campaigns have not demonstrated a high degree of power. Aiming to encourage labor to focus on more economically mighty workers, he fails to acknowledge that each of the campaign case studies I profile did demonstrate real power by overcoming scorched-earth union busting and winning first contracts (Burgerville, Colectivo, NewsGuild), or forcing a recalcitrant management to the bargaining table (Starbucks). Yet Olney presents a strawman comparison of employer resistance against graduate students compared to Amazon workers: “Certainly the organizing of grad students is impressive and important but these campaigns most often don’t confront the scorched-earth resistance of Amazon and its notorious union-busting law firm Littler Mendelsohn [sic] in Garner, North Carolina.”

It’s a particularly puzzling framing since Starbucks Workers United has been by far the main target of Littler Mendelson over the past four years. It’s true that Starbucks workers have not yet won a first contract, but is it really accurate to suggest that they lack economic power? If every single Starbucks store in the US went out on an indefinite strike tomorrow, that would cut off hundreds of millions in profits and create huge pressure on corporate to immediately settle, lest it stop raking in the $4 billion in profits it receives yearly.

Again, the obstacle here is primarily one of scale: with only about five percent of Starbucks stores organized, a strike of currently unionized stores on its own is insufficient to cut off huge profits — which is why the union has also simultaneously leveraged as much brand damage as possible, and why it remains intensely and successfully focused on continuing to increase the number of unionized stores.

Is Worker-to-Worker Organizing Generalizable?

Can worker-to-worker organizing spread beyond its current occupational and demographic strongholds? Fong doesn’t think so. By insisting that worker-to-worker organizing is constituted only by the specific case studies I examine in the book — thereby rejecting my actual definition of this model as a lightly staffed division of labor in which workers set strategy, and in which the training or initial organizing is done by workers — Fong proceeds to note some of the ungeneralizable specificities of these particular campaigns. But through this reasoning, you can disprove the generalizability of any organizing model, since no two union campaigns are ever identical, even within a single company.

Thus at one point Fong suggests that worker-to-worker organizing is only relevant for “low-hanging fruit” like academia and health care where workers are already “in motion.” Yet he then goes on to suggest that a worker-to-worker approach is only valid at Starbucks, where workplaces are tiny and where there are a lot of left-leaning workers. But was Starbucks really “low-hanging fruit”? There was exactly zero motion at the company before a group of Buffalo salts launched the campaign and it unexpectedly caught fire. Up until that moment, virtually everybody in organized labor assumed it was impossible before labor law reform to unionize such a massive company with so many small scattered workplaces, such high turnover, and such deep pockets for union-busting.

Grassroots momentum always looks inevitable in hindsight. It’s hard to know ahead of time how widely worker-to-worker organizing can expand, given that labor movement has done so little to seed drives in a wide range of economic sectors and given that most unions still remained mired in staff-intensive organizing norms. How many more opportunities for scaling up ambitious worker-to-worker campaigns were missed in the wake of the pandemic due to the labor union leadership’s suicidal unwillingness to seriously fund new organizing?

That said, We Are the Union does acknowledge and grapple with the major sociological limitation of the recent worker-led uptick: its disproportionate concentration among college-educated workers, whose progressive views and occupational leverage have tended to buoy their self-organization.

As I explain in the book, this is not actually a surprising pattern. Historically, it’s often been the relatively better-off layers of workers, such as skilled tool and die makers in the 1930s, who’ve taken the lead on grassroots unionism. And it’s reasonable to wager that the scope of this type of organizing could be expanded to different industries through well-resourced seeding and salting initiatives, the fruits of which we’ve begun to see at Amazon and in southern auto.

But the current scope limitations are real. And, as I write in the book, “the speed and extent to which unions can move towards a worker-to-worker model depends not only on leadership strategy, but also on the degree of self-activity, confidence, and politicization of workers in distinct occupations and industries at a given moment in time.” In instances where workers are unable or unwilling to take on more organizing responsibilities, good staff-intensive organizing may be the only feasible short-term path forward.

Ultimately, the generalizability of worker-to-worker unionism is an open question that can be answered only through practice. And since effective organizing is always context-specific, the particular tactics and organizing structures needed for unionizing a big factory or warehouse obviously won’t look the same as a small coffee shop. But as I try to show through the book’s extensive discussion of organizing costs and scalability, the only plausible path to unionizing large enough numbers in any industry is through a relatively lightly staffed organizing model:

The only way of testing how far it’s possible to go is by actively encouraging worker-to-worker unionism. And unions need to be crystal clear about the urgency of moving in such a direction, since prevailing staff-heavy approaches simply can’t build power widely enough.

Faced with powerful corporations and broken labor law, there are no guarantees that any unionization approach will succeed, especially when authoritarian reactionaries like Trump are in power. But if we’re going to wager on the revival of organized labor — and we should — worker-to-worker unionism is our best bet to defeat Trumpism and the billionaire class.

[This post combines my responses in Convergence and Jacobin to reviews they published of my new book We Are the Union.]