Coworkers, Not Contractors: Preserving the Social Fabric of Work in a Fractured World

The Future of Work Debate—and What It’s Missing

For more than a decade now, business leaders, policymakers, and scholars have been predicting radical changes in how Americans work. The rise of gig platforms like Uber and Lyft, advances in artificial intelligence, and the economic disruptions of the pandemic have all fueled headline-generating proclamations about the “future of work”—what it will look like, who will lose and win, and what, if anything, can or should be done about it. The ensuing conversations give voice to many Americans’ hopes and anxieties about work, technology, and even the meaning of a life well-lived: Do workers need more flexibility, more security, or both? Are platforms like Uber and Lyft prosocial innovations that should be freed from outdated regulations, or exploitative conspiracies to evade labor and employment laws? What might it look like to organize our lives around something other than work?

Debates about these questions are capacious and broad. But the debate as it currently stands is missing something essential: a clearly articulated and affirmative progressive case for preserving the institution of employment itself. Those who celebrate a brave new future of work, who see automation and the diffusion of independent contracting via platform-moderated gig work as efficient improvements on a tired and costly legal form, tell a clear and persuasive story about the good life, about how meaningful our lives could be, freed from the obligation to work on someone else’s schedule.¹ In response, progressives tend to defend the status quo by pointing to the many hard-fought statutory protections built around employment: minimum wage, overtime, antidiscrimination laws, the right to organize. These protections matter enormously, and we should fight for them. But in defending just them, progressive voices have at times implicitly disclaimed any independent stake in the organization of work. If employment is just a delivery vehicle for a social safety net, then if we could decouple work from the social safety net through universal basic income, portable benefits, or other policy innovations, perhaps it does not matter how we actually work?

What is truly distinct about employment is its relational character: that it constitutes ongoing, co-productive human relationships, promoted and protected by law. Those kinds of relationships are highly beneficial: for workers, customers, firms, and society as a whole—or at least they can be, under the right conditions.

Progressives can and should make more of this opportunity: to advance a more humanistic account of why we work and the multiplicity of reasons why employment is worth defending. Employment is a complex sociolegal institution with a great deal of baggage, much of which progressives rightly want to shed. And yet compared to its politically viable alternatives—massive job loss to automation or the reorganization of human labor into brief episodic transactions via hegemonic gig work—employment is the better option, by far.

The value of employment has often been obscured by legalistic debates over its definition, by the extent to which employment has been defined based on an employer’s “right to control” its employees. But in the real world, control follows primarily from economic power and inequality and is not a useful way of distinguishing employment from independent contracting.² Rather, within an otherwise atomistic economy, what is truly distinct about employment is its relational character: that it constitutes ongoing, co-productive human relationships, promoted and protected by law. Those kinds of relationships are highly beneficial: for workers, customers, firms, and society as a whole—or at least they can be, under the right conditions. Much is wrong with employment today, but the solution is not to give up on it. The solution is to invest further in it, through public policy that makes these relational investments worthwhile, for workers and for everyone else.

As such, in this brief, I make an affirmative, progressive case for employment, for its hard-fought legal protections and its relational benefits, all under the conditions that maximize its prosocial capacity.

How Coase Shaped the Way We Think About Employment

To build this case requires revisiting foundational principles, and challenging inherited ideas about what an economy is for and what we get from working together. To fully understand the terms of today’s debate about the future of work, we need to go back to 1937, when economist Ronald Coase published “The Nature of the Firm,” one of the most influential papers in the history of economics.³ Coase’s question was seemingly quite simple: Why do “firms” exist at all? Consistent with 1930s-era ideas about economic value, Coase was not asking about the corporate form here. Rather, he was interested in what economic discourse took for granted at the time as the most valuable attribute of the firm: its workers. More specifically, he wanted to understand why employment as a legal form had proliferated. If markets are efficient, he asked, why don’t people simply contract with one another for every task, rather than investing in ongoing work relationships?

Coase’s answer to this question, his answer to most questions, was a concept he called “transaction costs.”⁴ Finding someone to do a task, negotiating the terms of the exchange, and enforcing the agreement all take time and money. Employment—which he defined singularly based on a treatise on master and servant law that emphasized an employer’s right to direct a worker’s activities (the “right to control”)—was a way of avoiding those costs. By creating an employment relationship, bosses and workers no longer had to contract over every little thing. Under the right conditions, Coase argued, it would be cheaper to hire employees than to contract for labor.

It is difficult to articulate just how important Coase’s ideas about transaction costs have been within certain academic and policy circles, and how much they continue to shape our understanding of how to best engage in economic exchange: namely, as quickly and as cheaply as possible. Whereas organizational economics had once been seen as an alternative school of economic thought, one that radically differed from the neoclassical account in its assumptions about human motivation and the role of law, Coase’s theories collapsed the distinction. In explaining firms as an efficient alternative to the market, he paradoxically also redefined the social meaning of the firm—from a socio-legal institution bound together by people and norms to a value-maximizing efficiency function.⁵

Ironically, Coase’s theories today feature most prominently in arguments against employment. In discussing how work law should evolve, a host of policymakers, business leaders, and scholars invoke Coase to argue that it is employment that has now become too costly.⁶ The claim here is that technological developments, most especially the platform, have made it cheaper than ever to “spot contract” for labor, while the multitude of rules attached to employment have made it unjustifiably expensive. Neo-Coaseans accordingly argue that the transaction cost calculus today disfavors employment.⁷ It no longer makes sense to invest in ongoing work relationships when we can exchange labor, just in time and on demand.

Coase’s account was elegant. But it was also limited by his superficial analysis of employment, its history and its legal framework, and perhaps most of all by Coase’s deeply held priors about the stakes of “social frictions,” the human interactions and relationships attendant to economic activity.⁸ In his account, the social dimensions of exchange are flattened into costs.

In other words, what this framework leaves out is that human connections and relationships can also have real and measurable value, socially, politically, and economically. And an economy that fails to recognize that value . . . well, it looks very much like the economy we have today.

Transaction Benefits: The Missing Half of the Equation

To date, transaction cost economics has focused single-mindedly on costs, how the time and expense of economic coordination might dampen levels of exchange. The field of law and economics has, in turn, focused primarily on how to minimize these kinds of costs, to make exchanges of goods and services essentially frictionless.⁹

To be clear, the social interactions that throughout history have been the medium of all economic activity are costly in the Coasean sense, ranging from just time- and effort-consuming to stultifying and infuriating or worse. Meetings that could be emails, driving to the store, managing a clueless boss or an entitled customer—these take time, energy, and effort, and the social effects are not always positive. But to focus only on the downsides, to treat social “frictions” as always necessarily costly, is shortsighted and makes for bad policy. These interactions can also be useful, pleasant, and meaningful. Talking to a taxi driver, taking the time to get to know a patient, having a work best friend, building something with a coworker whose skills complement your own—these moments can produce joy, trust, solidarity, belonging, and social cohesion; they can facilitate the best of human innovation and creation. I use the term “relational transaction benefits” to describe this flip side of the coin. Transaction benefits are just as real, and as economically significant, as transaction costs.

The concept of relational transaction benefits draws on a long tradition of sociological and economic thought.¹⁰ Reclaiming this tradition is essential to building a framework adequate to articulate all that is at stake in debates about the future of work and the future of our economy. Throughout history and across cultures, “economic” interactions among people and groups have been just as much about community building as about value maximization. Human beings are unusual among animals in the degree to which we cooperate with non-kin, form large-scale social groups, and sustain complex institutions across generations. Evolutionary biologists and anthropologists suggest that this capacity for large-scale cooperation has emerged through a long evolutionary history in which cooperative groups consistently outcompeted less cooperative ones.¹¹ We are, in a deep sense, built for cooperation.

What Relational Transaction Benefits Look Like

These theoretical insights are supported by a growing body of empirical research documenting the concrete benefits of human interaction within “economic” contexts. These benefits accrue systematically at multiple levels:

For individuals: Research shows that social interactions attendant to economic and/or instrumental activities—activities not conventionally understood as “social”—meaningfully impact our social well-being. Starbucks customers who were instructed to try and “have a genuine interaction with the cashier” left feeling happier than those who were told to just “have [their] money ready, and avoid unnecessary conversation.”¹² Work, as it turns out, is the leading social location in which adults form friendships.¹³ People are more likely to make friends at work than to make friends at school, in their neighborhoods, through religious or purely social activities, or even through existing friends. Importantly, these benefits accrue regardless of self-reported “tastes” for interaction; introverts and extroverts alike benefit from them.

For firms: Relational transaction benefits also accrue at the group and organizational level. And here, they translate more directly into traditionally measured “economic” benefits with a monetary value. As such, it is not just that economic situations foster sociability but also that social relationships foster productivity and efficacy. Organizational scholars have shown that face-to-face conversation leads to higher productivity.¹⁴ Interpersonal relationships at work facilitate knowledge transfer and collaboration.¹⁵ Invention and innovation are social processes that rely on peer connections and learning.¹⁶ Having positive coworker interactions and feeling organizational support increases employee commitment and productivity.¹⁷

For society: These benefits transcend the individual and the firm. These are truly social benefits, positive externalities. Informal social and economic ties help produce broader social trust.¹⁸ When people’s daily experiences of interacting with others are positive, they feel better about people generally and more hopeful about the future. Meaningful interpersonal interactions in workplaces improve intergroup relations across complex racial and sociopolitical divides.¹⁹ In contrast, experiences of social exclusion at work or in broader economic life contribute to the atomization, polarization, and social mistrust that now characterize American public life.²⁰

Why Work Is a Uniquely Important Site of Relational Transaction Benefits

While all economic activity can have meaningful social benefits, work stands out for its unique capacity to produce social solidarity, because work facilitates ongoing interactions, cooperation, and collective orientation.

First, work facilitates sustained social interaction and connection. Unlike most market transactions, which tend to be fleeting or at best episodic, coproduction brings people together regularly and continually.²¹ And the research on trust formation is unambiguous: Trust takes time. It cannot be produced on demand or replicated algorithmically. Work is one of the primary sites of social integration and connection in American life, bringing together people of different backgrounds, experiences, and perspectives in sustained, purposeful interaction.

Second, the materiality of working together matters for its role in building social solidarity. Working together entails experiences of mutual interdependence, vulnerability, and shared fate. These experiences have long been shown to lead to trust, solidarity, and both individual and collective well-being.²² Working together allows people to experience in concrete, material ways our interdependence.

Work is one of the primary sites of social integration and connection in American life, bringing together people of different backgrounds, experiences, and perspectives in sustained, purposeful interaction.

Finally, working together builds a sense of group belonging. Working together on an ongoing basis facilitates collective identification and social inclusion, feelings which psychologists treat as basic human needs, comparable to food or shelter.²³ Social inclusion significantly predicts mental health, while social exclusion breeds radicalism. Group identification bonds people together across difference, undercutting preexisting biases and prejudices.²⁴ Group membership leads to civic participation, community investment, and political trust.²⁵

Bringing Transaction Benefits into the Debates About the Future of Work: The Relational Case for Employment

And so, what does all of this have to do with the future of work? Honestly, everything. In a host of ways, the transaction costs framework burdens our conversations about the future of work, constraining how we articulate, even imagine, the stakes. Automation and gig work, the argument goes, reduce costs related to employment, and so we should favor them in the name of efficiency and personal freedom. Treating reduced transaction costs as the measure of how to organize work taints the discussion from the outset.

Evaluated through the lens of relational transaction benefits, however, the comparison between employment and its alternatives looks quite different. Automation can be efficient; it can also be antisocial. Amazon factories today are eerily silent. Seemingly by design, they limit human collaboration. Workers engage with robots day in and day out, but only rarely with each other. Similarly, gig work may reduce certain coordination costs, allowing just-in-time overlap of labor demand and supply. But it is exactly this superficial ease that systematically sacrifices relational transaction benefits.

The relational loss occasioned by automation, at least automation used to displace points of human interaction, is easier to observe. Yet the distinction with independent contracting, particularly as instantiated by gig work, when it comes to relational benefits is also significant, even as it is harder to directly observe. It is accordingly worth mapping out how contract work and employment compare on each of the three dimensions that make work a primary source of relational benefits: interaction, cooperation, and belonging.

Interactions and relationships: Employment is ongoing and relational. Independent contracting is episodic and transactional. This structural reality matters enormously for the kind and quality of relationships that follow—to some extent between workers and customers, but tremendously when it comes to coworker relationships. In an increasingly networked world, we know that “weak ties,” our random acquaintances and friends of friends, matter.²⁶ Yet all the weak ties in the world do not replace strong ties—close, deep relationships—which are the bedrock of social and mental health. Employment is one of the only ways that Americans today still build strong ties outside of their immediate family and across the demographic trenches of race, religion, and politics.

In contrast, gig platforms actively prevent workers from forming ongoing relationships with clients or with one another. Gig workers tend to work in isolation, without colleagues, without a shared space, and without any of the informal social interactions that constitute workplace culture. As a legal and social matter, independent contractors have no coworkers.

Coproduction, not just coordination: Employment allows workers to amalgamate their efforts through cooperation, which extends beyond mere economic coordination. Employees do not simply work in parallel or in seriatim; they build something together that none of them could build alone.

Traditional independent contractors coordinate efforts with their customers, but they do not coproduce. Meanwhile, gig workers are legally and practically market competitors, a reality platforms exploit through algorithmic management that individualizes performance and rating systems that make workers compete for customer approval. (Of course, employers have long pitted employees against each other to incentivize production, but that is a management choice—arguably a poor one, under a transaction benefits view of the firm—not a defining legal structure.)

These distinct legal and organizational forms shape the kind of work that can be performed and, in turn, its relational potential. Employment allows workers to participate in a shared endeavor. Platforms do not.

Collective orientation: Ultimately, employees constitute part of an ongoing communal enterprise; together they form, in the Coasean view, the firm. As such, employment provides collective infrastructure for group belonging—a shared group identity, a physical or virtual space, a community of practice, and an ongoing culture.

Independent contracting, in contrast, is characterized by its independence and lack of sustained community. Public policy can attempt to change this, as New York City has done by providing shared public space (and a shared collective identity) for its gig community.²⁷ Yet when workers do not actually work together on the same projects over time, a deep sense of belonging is harder to come by.

When Gig Work Degrades Essential Services: The Case of Nursing

Nursing is an important case study of how gig work undermines the relational benefits of employment—even in the most relational professions.

Nursing is in many ways a form of care work. Relational capacity is necessary to implement the technical skills of nursing: the ability to build trust with patients, to know them over time, to notice subtle changes in condition, and to coordinate effectively with other members of a care team. These capacities are cultivated through ongoing relationships between nurses and patients, between nurses and colleagues, and between nurses and the organizations in which they work.

Platform advocates have celebrated the rise of on-demand nursing platforms—which allow hospitals to fill purported staffing gaps by contracting with independent nurse contractors available for short-term assignments—as a solution to nursing shortages and a source of flexibility for nurses who want greater control over their schedules.²⁸ But critics have documented a range of serious problems that follow directly from the erosion of the employment relationship.²⁹

On-demand nurses, by definition, lack continuity with particular patients. They arrive at new facilities without knowledge of institutional culture, patient histories, or team norms. Research on patient safety consistently identifies care fragmentation, the breakdown of continuity in the care relationship, as a leading cause of medical error.³⁰

The harms extend beyond individual patient care to the lived experience of the profession. Of course, on-demand nurses, classified as independent contractors, are ineligible for employer-sponsored benefits, have no access to collective bargaining, and bear the full risk of income variability.³¹ But less emphasized is how they are excluded from the communities of practice that are essential to professional development and peer support—communities that help nurses navigate the emotional demands of care work and develop their clinical judgment over time.

A Caveat: Relational Harms at Work

In making the argument that working together can make us happier and more unified, I do not seek to minimize that the opposite can also be true, nor the fact that these benefits vary widely across social position.

Transaction benefits are not automatic. They depend on the conditions under which work is performed. Under the wrong conditions, working can be profoundly dispiriting. It can be lonely.³² It can force us away from the people we love most, and it can put our health at risk.³³ It can make us bad citizens and community members.³⁴ It can deaden our humanity. It can coerce socially marginalized groups away from the communities in which they feel most valued and force them instead into daily experiences of isolation and subjugation.³⁵ Moreover, the work that has the greatest potential for fostering relational transaction benefits, emotional labor, tends to be both feminized and, as a result, insufficiently remunerated.³⁶

Formal employment relationships alone will not save us. Exploitation, harassment, chronic insecurity, and authoritarianism all destroy relational transaction benefits even within formal employment relationships. Employment is a legal form with potential relational value. The extent to which that value is realized depends on the specifics: the broader political, socioeconomic, and cultural conditions of employment.

The Standard Employment Relationship: Built Once, Can Be Built Again

If employment has these benefits—if it offers such potential—why do gig work and automation remain alluring, even to some of the working people whose livelihoods may soon be destroyed by them? Beyond misclassification and gig company marketing, it is clear that some segment of the working population is tempted by the symbolic freedom of employee-exit. Employment can be a powerful source of social connectedness and solidarity under the right conditions. However, the reality is that we are increasingly failing to create those conditions for many working people in the United States today.

The problem with employment today is not its costs, though, and it is not that people no longer want working relationships. The problem is that terms and conditions of employment are an ongoing site of political contestation, and workers have been losing badly for decades.

Scholars use the term “the standard employment relationship” (SER) to describe the social and cultural edifice built around employment at particular points in time. The 20th-century SER was multidimensional, shaped by law, collective bargaining agreements, personnel policies, and deeply entrenched cultural norms. It was a political compromise and a psychological contract.³⁷ That contract offered economic security, regular opportunities for promotion and advancement, social connections, and a sense of identity and belonging to working people. In exchange, employers received dedicated workers with deep firm-specific knowledge. The SER—at least for those with access to it—promoted transaction benefits for workers, employers, customers, and society writ large.

Today, too many otherwise progressive scholars buy into the notion of employment or labor as a cost, rather than an asset or source of value. And in so doing, too many of those scholars tacitly accept that work’s future must be a race to the bottom.

That SER has been in decline since the 1970s, upended as a result of political choices too often blamed on technology. The broad societal benefits of the SER were the result of political struggle and political compromise, ensuring ongoing investment by workers, employers, and the government in the employment relationship.

Today employers invest much less in their employees. In part by discursively minimizing the value of workers—by treating them as liabilities instead of assets—employers have been able to demand more productivity and offer a much smaller share of the pie in return.³⁸ Meanwhile, the weakening of the social safety net inhibits quality employment too. The lack of affordable housing near job centers, inadequate public transit, and insufficient care infrastructure all make it that much harder for workers to invest in workplace relationships.

Given employment’s potential benefits for productivity and for society, this moment should be a worrying state of affairs. It should also be a call to work on a new and better standard employment relationship. Today, too many otherwise progressive scholars buy into the notion of employment or labor as a cost, rather than an asset or source of value. And in so doing, too many of those scholars tacitly accept that work’s future must be a race to the bottom. Faced with the cost savings of automation and alternate work arrangements, the argument is that employment must find ways to make itself cheaper. A transaction benefits perspective provides an alternative argument: The costs of cheap labor, when it comes to mental health, social cohesion, political stability, and economic productivity, are simply too high.

What a 21st-Century SER Could Look Like

The 20th-century SER was built around a set of socioeconomic realities that have changed: a male breadwinner model, an economy anchored in manufacturing, and a presumption of lifelong employment. Building a 21st-century SER means adjusting those particulars, while preserving the core insight: that employment is not a transaction but a relationship, and that the law should structure and support it accordingly.

A 21st-century SER would recognize that workers today need both security and flexibility. It would invest in the conditions that generate relational transaction benefits: stable schedules, reasonable workloads, genuine voice in workplace decisions, and opportunities to form relationships with coworkers and customers, without surveillance or bureaucratic control.

A 21st-century SER would adapt to the contemporary realities of caregiving by significantly reducing the hours of work expected for “full-time” pay and benefits. Ironically, the most enduring characteristic of the 20th-century SER is the characteristic that is least well suited for the needs of American workers and their families today: the 40-hour workweek as the paradigm for full-time and fully remunerated work. The 40-hour workweek is neither written in stone nor inexorably encoded in law. And it is certainly not something that employers adopted munificently or without complaint.³⁹ The 40-hour workweek was the result of massive worker activism. Workers demanded it because it worked for them and the way they organized their families and their lives.

Today, however, we organize our lives quite differently. Gender is no longer an automatic division of labor. Yet for all the other changes to the SER, presumptions about total hours, work schedules, and flexibility have changed very little. As such, platform employers argue that they must treat their workers as independent contractors in order to give them flexibility. This, of course, is not true. Workers’ schedules are not determined by employment law; they are determined in the penumbra of power, policies, and norms around that law. The 40-hour workweek was part of the SER of the 20th century; the SER of the 21st should be significantly shorter.

Finally, a 21st-century SER must protect and prioritize the workers its predecessors excluded. Domestic workers and other care workers—disproportionately women and workers of color—perform some of the most relationally intensive work in the economy. Their exclusion from the protections of the SER was enabled by, yet also perpetuates, the systematic devaluation of relational work. Relationships underlie all work. Care work is not the exclusion; it is the model.

Conclusion: Fighting for the Future We Want

We live in difficult times. Today, Americans are unhappy and lonely. Social trust is at a low.⁴⁰ Disinformation and polarization corrode the efficacy and stability of our institutions. By some measures, one-third of all adults in this country struggle with anxiety or depression.⁴¹ In 2023, over 50,000 Americans died by suicide—more than any other year on record.⁴²

Against this backdrop, it would be a mistake to contemplate the future of any American institution without contemplating its impact upon the future of American social solidarity. And yet, the impact of the “future of work” on our social relationships has largely been treated as an afterthought. How we transact is largely left out of our increasingly pained reflections on how we interact. We blame Facebook and X but let Amazon Prime off scot-free.

Debates about the future of work are often framed as if the only choices are adaptation or resistance—accepting the market’s verdict about efficiency or fighting a rearguard action to preserve the past. This framing is false, and accepting it means ceding our most compelling arguments.

Work is more than a mechanism for producing goods and services. It is one of the primary sites through which human beings build trust, form communities, develop shared identities, and sustain democratic life. The institutional form of work—whether it takes the form of stable relationships or episodic transactions—matters.

A genuinely progressive vision for the future of work must make an affirmative case for work as a social institution: for why the ongoing, cooperative, relational character of employment generates value that cannot be reduced to transaction cost minimization and cannot be reproduced through platform contracting or automated production. In turn, we must translate that case into concrete policy—rebuilding the SER for the 21st century and investing in the conditions that allow employment to generate the social connection and solidarity that our country so desperately needs.

Just as the law has been used to minimize transaction costs, so too can it be used to promote transaction benefits.