An open access publication of the American Academy of Arts & Sciences
Winter 2023

Making Decent Jobs

John S. Ahlquist
View PDF

On both normative and pragmatic grounds, I make a case for “decent jobs” over the current discourse around “good jobs.” I define decent jobs as ones that reflect sustained worker influence over the terms and conditions of work. Making decent jobs necessarily entails groups of workers capable of engaging strategically with firms and governments. Where will these groups come from? Changes in technology, the structure of production, and boundaries of the firm all point to profound difficulties in sustaining collective action centered on workplace relationships and identities. Networks of workers organized around mutual aid show some promise, but connecting these groups to concerted action on the shop floor implies numerous organizational and governance challenges.

John S. Ahlquist is Associate Dean and Professor of Public Policy at the School of Global Policy and Strategy at the University of California, San Diego. He is the author of In the Interest of Others: Organizations and Social Activism (with Margaret Levi, 2013) and Maximum Likelihood for Social Science: Strategies for Analysis (with Michael D. Ward, 2018), as well as numerous articles across political science and economics.

In many, but not all, advanced democracies, income and wealth inequality are at levels not seen since the times of top hats, bustles, and oil lamps.1 Across the developed world, the share of production going to workers—long believed to be something of a natural constant—has declined significantly over the last forty years.2 Economic mobility is slowing and jobs are polarized.3 Employers increasingly hire workers into precarious, supplier-like relationships devoid of labor protections and access to social insurance.4 Real wages are stagnant for the bottom half of the wage distribution. Many lament the disappearance of “good jobs,” a process now linked to the rise of nativist populism and “deaths of despair.”5 And this was before the COVID-19 pandemic gave us the moniker “essential worker” for people who, more often than not, are treated as anything but indispensable.6

Neoliberalism is the (hackneyed) catch-all term covering the intellectual, political, and rhetorical devices abetting these changes. At its core is a stylized vision of untethered individuals in transitory, arms-length relationships coming and going as so many local optimizers. This vision relies on an uncomfortable dualism between “society” and “market” that banishes difficult questions of fairness and mutual obligation from economic interactions while treating government as largely pernicious, something to be minimized. It provides no coherent response to the looming questions of distributional fairness or the appropriate conditions of production, provoking calls for a “paradigm shift” and “new moral political economic framework.”7

The moral rubber hits the economic road where people buy and sell that “fictitious commodity” of human labor: that is, jobs.8 In contrast to various ideas about good jobs, I make a case for “decent jobs,” which I define as jobs that reflect sustained worker influence over the terms and conditions of employment. Individual workers are almost never able to reliably secure such influence on their own. American labor unions, operating in a wildly outmoded legal and regulatory framework, are no longer up to the task in all but a handful of situations.9

Rebuilding workers’ “strategic capacity” and influence is a political project, implicating power relations, conflicting and overlapping interests, and the practicalities of sustaining agreements through time.10 Changes to the structure of production, the nature of hiring, and the treatment of workers once hired are all undermining the social and economic basis for sustained collective action on the job.11 Worker voice is therefore unlikely to emerge from shop-floor unionization campaigns under current U.S. labor law.12 Changing the laws is unlikely without sustained political pressure.

With this impasse in mind, political entrepreneurs and labor activists are experimenting with a variety of organizational forms and funding models designed to build social networks and expand workers’ “community of fate” beyond the shop floor.13 Many are taking a mutual-aid approach, reminiscent of the early history of labor organizing. These “mutualist” groups tend to organize around a location, a particular cause, or an epistemic, professional, or cultural community.14 What they sometimes lack is the consistent presence at the point of production necessary for becoming agents of decent jobs. As these networks become more densely connected and encounter conflicting interests among workers themselves, governance issues will loom large.

What do we mean when we talk about good jobs? There are numerous characterizations, emerging from interviews and surveys of workers, examination of the historical record, and lots of introspection. The U.S. Department of Labor recently launched the “Job Quality Measurement Initiative” to figure out how to measure good jobs.15 Clearly, there is no consensus analytic definition of job quality, but there are some common themes emerging.

  • A good job is multidimensional, involving an unspecified combination of a living wage; stable/predictable scheduling; stable/predictable pay; forward/upward mobility; predictability in employment; access to benefits; freedom from discrimination, abuse, and harassment; reasonably safe work environment; autonomy; voice; a sense of mission, purpose, or belonging; and recognition/status.
  • A good job is context dependent. Different types of work arrangements can be “good” for different people in different life situations. And good jobs are embedded in the larger milieu. What counts as good depends on what was initially promised, what other jobs are like, and what other employers are doing.
  • A good job is not fixed or static. What counts as a good job must necessarily change and evolve.

As a target for policy-making, good jobs suffer from conflicting goals, competing constituencies, and the imperative of continual adaptation. In any case, good jobs are ill-defined from a moral political economy standpoint. The phrase “good jobs” itself calls to mind a technical problem of quality assurance, eliding the exercise of power, threat of conflict, and questions of fairness endemic to the labor market. It is thus unsurprising that many policy prescriptions designed to increase the supply of good jobs end up turning workers into stakeholders at best and spectators at worst, rather than the ultimate arbiters of whether a job is any good.

As one example, labor scholar Zeynep Ton locates the supply of good jobs in the strategic and tactical decisions of executives and managers.16 Managers surely help shape working conditions. There are important attempts to better orient corporate objectives and management practices toward human flourishing.17 Nevertheless, expecting enlightened managers to land on a “good jobs strategy” is untenable as a policy program.18 Echoing the old literature on efficiency wages, firms pursuing the good jobs strategy exist in an industrial ecosystem with other employers successfully pursuing a “bad jobs” approach (which can make barely adequate jobs look good in comparison). Reliably producing decent working conditions across a dynamic economy is impossible to achieve solely through one-off reforms to the business practices of individual firms.

Looking to technocratic policy-making as the source of good jobs is another wrong turn. As we see in the aftermath of the “great resignation,” setting monetary and other macroeconomic policies to keep unemployment low does not guarantee good jobs nor does it provide a mechanism for translating transitory worker leverage into durable improvements that persist beyond the next recession.19 Business-government partnerships for worker training won’t solve foundational problems of credibility and management of a “common pool” of skilled workers. If workers cannot durably and systematically affect the terms of their employment and exert concerted political pressure, calls to enact “place-based” government policies and enlightened corporate strategies to “bring back” the lost good jobs ring hollow. And even when government manages to produce policies that might make some jobs better, implementation can be uneven, unreliable, and subject to political cycles.

Enforcement of regulatory standards is far more effective when workers gather and transmit information and advocate for their own interests.20 Worker power in favorable political and economic contexts transformed industrial exploitation, drudgery, and alienation into the “good, blue-collar manufacturing jobs” that are nostalgic tropes of campaign speeches.

I want to reorient away from good jobs toward a notion of “decent jobs,” by which I mean work arrangements resulting from processes that reliably, consistently, and directly incorporate workers and respect government-set standards that themselves incorporate workers’ interests. The idea of decency presupposes standards of mutual obligation and respectability broadly shared in some political community. The focus on worker voice emphasizes the political project as well as thorny questions of governance.

The notion of decent jobs may echo old calls for “industrial democracy.” But the core rationale is pragmatic, resting on extensive findings linking the fairness and transparency of decision procedures with improved well-being and organizational performance.21 As a matter of political economy, decent jobs are necessary even if the ultimate goal is a specific notion of the good job or something more ambitious as economic fairness or justice.

Ideas about industrial democracy date back to the Fabians of the nineteenth century.22 There are a variety of consequentialist arguments for workplace democracy with mixed degrees of empirical support. Political scientist Robert A. Dahl articulated a normative element, claiming that, insofar as employees are “roughly equally well-qualified to decide which matters. . . require binding collective decision,” they have a moral right to democratic voice in the firm.23 Philosopher Elizabeth Anderson reinvigorated this line of argument in her recent attack on “private governments.”24

A common critique of industrial democracy—one laid out by political scientist Robert Mayer—rejects a moral right to worker voice because the employer recruits the worker.25 The “terms of subjection are negotiated,” which obviates the worker’s claim to a voice in the firm in the manner of citizens with respect to their government.26 Making this move requires the assumption that all negotiation takes place before signing a contract, when a worker can refuse subjection in theory, if not always in practice. However, the limits of worker subjection—like many aspects of a job contract—are difficult to articulate and credibly enforce. The conditions of the employment relationship must evolve in response to circumstance, which requires adaptation and implicit bargaining.27 Once we recognize both contractual incompleteness and the “relational contract” that characterizes virtually every job, Mayer’s critique verges on the irrelevant.

Different jobs will exhibit differing levels of contractual incompleteness. Workers will vary in their desire and capacity to exercise their voices, so creating decent jobs need not imply a worker-run firm, as some in the industrial democracy tradition have argued. Rather, the contribution of the industrial democracy approach is its demonstration that an individual worker’s exit option—the only real mechanism of redress in the neoliberal political economy—is vain and certainly insufficient to produce decent jobs.

My main justification for decent jobs is practical and derives from the “procedural justice” approach closely associated with legal scholar Tom R. Tyler and his collaborators. Across several domains—including employment—they have produced extensive evidence showing that processes viewed as transparent, fair, consistent, and accessible increase a sense of just treatment, personal agency, voluntary compliance with organizational decisions, and willingness to take actions aligned with organizational goals in ambiguous or unspecified circumstances. Conversely, processes that are opaque, arbitrary, or unilaterally imposed have the opposite effects, even if the decision outcome is good from the worker’s perspective. Procedural justice is not just about getting a better outcome. People appear to value fair processes in part because they signal social standing in a group: that is, respect.28 From a procedural justice perspective, jobs are deemed decent based on the processes by which we arrive at and sustain them, not the content of the work arrangements themselves. This approach to decent jobs is both coherent and tractable. A worker, manager, or policy-maker can evaluate whether any particular organization, reform, or law will increase the decency of certain jobs or the overall supply of decent jobs.

An important part of the definition of decency is the plural in “processes.” Workers can have influence through a variety of channels in different organizational forms. This includes—but does not require—traditional labor unions and collective bargaining under the threat of strikes. Historically, other forms of worker influence include self-managed teams and “quality circles,” works councils and other consultative bodies, minority unions, ombudspersons, job rotation in and out of management positions, and worker representation on corporate boards.29 Across all these options, workers can have more (or less) influence in ways that are more (or less) procedurally transparent and neutral. As such, job decency is a matter of degree.

As one illustrative historical example, Margaret Levi and I studied dockworkers and their unions through the twentieth century.30 In the 1930s, work on the docks was bad. Pay was low and conditions were filthy and dangerous. The job was casual; you never knew if you would be hired back the next day, but it might help if you kick back part of your wages to the “walking boss.” At the time, the union for dockworkers—the International Longshoremen’s Association (ILA)—was known for its authoritarian governance and feckless leaders. Although a union was present, these jobs were clearly not decent.

In 1934, workers at almost all ports along the West Coast went on strike, contrary to the instructions of the ineffectual ILA leadership. After enduring government violence, the workers won and broke away from the ILA to establish a new union, the International Longshore and Warehouse Union (ILWU). The ILWU became known for its intense internal democracy almost as much as for its industrial effectiveness. Union meetings could be contentious, and the members regularly questioned union leaders. Through the ILWU, dockworkers transformed their jobs, introducing a system of job rotation managed through a hiring and dispatch hall largely controlled by the workers themselves. The newfound decency of their jobs is reflected in a favorite saying from the time: the ILWU “transformed wharf rats into lords of the docks.”31 To this day the work remains difficult and, at times, dangerous. But there is substantial worker influence over work conditions and the hiring hall remains. Although not perfect, the job of an ILWU dockworker is decent.

For there to be decent jobs there must be effective workers’ organizations. There is no other way. But calls for improved worker voice are a dime a dozen. The overriding challenge is how, which requires some vision of where workers might come together in ways that make collective action more likely. At the level of policy, there is a chicken-and-egg problem: existing law and regulations governing labor unions incentivize overly narrow and parochial bargaining units, enable employer resistance, hamper organizational experimentation in unions, and preclude some organizational alternatives altogether. But changing the law requires that workers’ organizations and their allies apply sustained political pressure beyond what they appear capable of delivering. This situation for many workers is, unfortunately, not all that dissimilar from that facing the dockworkers in the early 1930s.

Historically, successful union organizing rested on one of two pillars of common interest across workers: occupational identities (in the form of shared skills or occupations) and the structure of work (in the form of shared employers or buyer-supplier relationships). But a shared interest is not enough. Successful collective action is more sustainable when relationships are ongoing, people have long time horizons, it is easy to observe and share information, group membership is clear, and there are coordinating devices that can sustain reputations and resolve disputes.32

Many of the changes that make work less decent also undermine both the foundational pillars of common interest as well as conditions conducive to collective action. Thanks to improved information technology, jobs are being decomposed into tasks, perhaps performed remotely and in parallel in widely distributed supply chains in multiple countries. Some tasks are increasingly assigned to algorithms or robots and the pace of change is rapid, threatening occupational identities. Job contracts may be project-based or contingent, rather than open-ended. Changes in shipping technology and economic policy have enabled global sourcing, extended supply chains, and led to threats to shift capital investments. This same dismantling process also extends to shifting boundaries of firms. “Fissuring” workplaces sever the links of common employers among some workers while obscuring the existence of a common employer between others.33 It is far harder for workers to see where they fit in the larger production process and who might be in a similar position. At the same time, effective action requires that workers exert pressure up and down the supply chain. In many industries, especially service work, scheduling is volatile, turnover is high, and time horizons are short, reducing the attractiveness of exercising voice. Some workers are geographically fragmented and politically isolated. All these changes work against the emergence of collective action organized around stable occupational identities or single worksites.

These are not new problems. In the early industrial period, joint bargaining and extended, industry-wide political organizations were not yet conceived. Production was fragmented and barriers of language, race, and religion were real. Organized cooperation among workers took the form of mutual aid and friendly societies aimed at assisting one another or families in times of sickness, injury, and untimely death. In late-nineteenth-century Britain, friendly and mutual aid societies became the organizations providing something like “portable benefits” to upward of 70 percent of British workers, relying on intensive social ritual and regular drinking events to sustain membership and build camaraderie and solidarity.34 In the United States, mutual aid groups formed around the nuclei of shared religion, ethnicity and language (especially for recent immigrants), gender, and, of course, race. Some mutual aid societies collapsed due to their parochial nature, agency problems, and the correlated risks among workers in the same industry or city.35 Other mutual aid groups were folded into modern trade unions and used to recruit and retain union members.36 Over time, commercial insurance sometimes proved more economical. Governments stepped in, through social insurance and welfare programs, to underwrite and sometimes replace teetering mutual aid societies and union funds.37

In the contemporary period, unions are vanishing. The American health, welfare, and social insurance systems are creaking, as unprecedented demand runs up against decades of disinvestment. The fraying safety net’s encounter with the COVID-19 pandemic has provoked new interest in decentralized mutual aid at the neighborhood level, as well as increased attempts to raise and disburse funds across distributed networks.38 Recent surveys reinforce the idea that American workers are hungry for a say on the job in ways that also address failures in our social insurance systems. Mutual aid “services” and portable benefits, alongside collective bargaining, are the key areas of worker interest.39

The data are equally clear about what workers want their organizations to avoid: partisan politics and conflictual relations with management. In their survey conjoint experiment, political scientist Alexander Hertel-Fernandez and his collaborators find that respondents were less likely to support a workers’ organization described as “campaigning for pro-worker politicians in elections” and less willing to pay dues to support political engagement.40 A survey from the conservative think tank American Compass offered respondents a stark choice between an organization that “devotes its resources only to issues and issues facing you and your coworkers at your workplace” and one that also devotes resources to “national political issues.”41 Given these options, respondents preferred an organization that eschewed national politics by a 2-to-1 margin.

It is easy to object that this aversion to conflict and partisan politics should be ignored because strike threats and political engagement are necessary for workers’ organizations to make jobs more decent over the long term. I disagree. Partisan politics are always divisive and unpleasant, all the more so in the current American political economy. In the recent period, union political activities have rarely managed to deliver concrete wins for working people, especially at the federal level, which increasingly dominates all levels of political contestation. It is hardly surprising that, when asked, workers want to avoid expanding into areas in which cooperation may be more difficult or impossible. Understanding this constraint will be important for getting new organization-building off the ground.

This is not to say that political mobilization and even partisan alliances should be avoided over the longer term—far from it. Margaret Levi and I show that workers’ organizations can and will take on broad-based political commitments that extend well beyond the immediate job concerns of the current members.42 But there is an important sequencing: these organizations must first “deliver the goods” before expanding their scope of action to include national (or international) politics. Solving these initial coordination and collective action problems is critical; workers can see that their colleagues are people worth taking a risk on. Developing this kind of social capital in one domain can then alter what workers believe to be feasible and in their interest in other areas. More important, it provides a vehicle through which workers can deliberate about the political projects they think are most important. Successes must build on each other, begging the question of how to get the ball rolling. The discussion thus far points to programs that can connect workers across worksites, collaborate with management, and deliver valued benefits.

Among explicitly labor-focused mutualist organizations, the most high-profile have emerged among workers in that regulatory liminal space of independent contractors and freelancers, especially in media and tech, but also for taxi/rideshare and delivery drivers.43 Most of these organizations emphasize information sharing, training, mutual aid arrangements, and sometimes portable benefits. Some managed to coordinate job actions among a set of workers in particular cities, often—but not exclusively—in the more labor-friendly parts of the country.

If mutual aid and self-funded portable benefits organizations do manage to spread, they will begin to confront important governance challenges well before they can grow into their potential role as vehicles for decent jobs. The biggest issue is leveraging mutual aid communities (that may rely on geographic or other social ties) into an actual and durable shop-floor presence across employers. Mutualist groups might approach this problem in several ways. The most obvious is that mutualist groups provide support or a backstop for other organizing efforts that may come from traditional unions. Such support can (and does) include managing communication forums for connecting workers with support they need in the event of a job action as well as working with union activists to identify promising organizing opportunities. Successful provision of certain portable benefits can strengthen workers’ hands when they approach their employers through other organizational vehicles. For example, childcare collectives could both provide a valued service while also giving workers more time and mental space to engage with the struggle for more decent jobs. More ambitiously, mutualist groups might be the seeds that grow into something like minority unions. For example, workers connected in a mutual aid network could develop the ability to coordinate their wage demands. Whether this is ultimately scalable will depend, in part, on legal and political concerns.

Managing conflicts between groups of workers will be the second governance challenge. Most immediately, there may be different groups trying to provide services or organize the same workers. Experimentation is critical and competition between these groups can be beneficial. They will need the space and funding to explore but also the incentives to collaborate with erstwhile competitors or shut down altogether if better options are available. Workers will have differing interests in some circumstances. More senior workers and labor market incumbents may view flexible or temporary work arrangements differently from younger workers and those on the outside. Too often, incumbents view freelancers, temporary workers, and “gig” workers as either a threat to their existing job or as workers forced into a “bad job.”

Yet there is substantial evidence that many workers in nonstandard jobs prefer those types of positions. As Sara Horowitz, founder of the Freelancers Union, notes: attempts to force gig work into the existing employment law buckets in the name of worker protection become a “wedge issue” that divides workers and their organizations.44 The governance challenge will involve building coalitions across these groups and settling on plans tolerable to both, preventing some employers from exploiting this wedge while enabling others to grow and develop new technologies and work processes.

Other potential governance challenges will involve externalities as one group of workers makes demands or provides services that impact others. Historically, this was a common problem among unions, where the wage demands of some workers could affect the employment or purchasing power of others. These conflicts spawned several institutional solutions, including sectoral and centralized bargaining in Germany and Scandinavia, labor tribunals in Australia and New Zealand, and government-brokered labor agreements in multiple countries.45 Versions of these problems may reemerge, but the fissuring of employment and the difficulty in connecting workers in different parts of supply chains will certainly raise new challenges and require different solutions. Confederated or meta-organizations—like labor federations of the past and present—will be critical for identifying and, ideally, managing externalities and conflicts as they emerge.

Making decent jobs requires independent organizations representing workers’ interests at the points of production, regulatory enforcement, and policy-making. In the United States, this process will ultimately require wholesale revision of (or even jettisoning) the existing labor law, which is based on establishment-level bargaining and a 1930s vision of both households and industry. But such changes appear politically impossible now. Any program for making jobs more decent must start with the long-term project of building social capital among working people. A renewed focus on mutual aid is one promising avenue for making some immediate progress, but it comes with risks and without any guarantee of success.

What roles can academics and policy advocates play? There is value in articulating a “new moral political economic framework,” but intellectuals are unlikely to solve inherently local and contextual problems of articulating grievances and building organizations.46 It is equally important to recognize that independent, autonomous worker organizations may pursue economic or policy goals contrary to the preferences of both the populist Right and the “Brahmin Left.”47 We see this already when looking at public opinion data describing attitudes toward certain immigration and trade policies as well as a general desire to keep unions out of (partisan) politics.

There will be extensive experimentation and, perhaps, competition between different groups and organizations to address common problems. Academics are well positioned to play an important coordinating role here, by rigorously evaluating different programs and highlighting their successes, which can reduce wasteful infighting and poorly supported advocacy. When evaluating which initiatives, programs, and organizational experiments succeed, however, the moral framework should be that of decent jobs.

author’s note

My thanks to Roy Bahat, Henry Farrell, Peter Gourevitch, Josh Graff-Zivin, Steph Haggard, Nina Hall, Tom Kochan, Margaret Levi, Liz Lyons, Matthias Matthijs, Kathy Thelen, Zachary Ugolnik, and Nathan Wilmers for helpful comments. I am grateful to Suresh Naidu and Michelle Miller for agreeing to engage in conversation around the ideas herein.