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

Eudaimonic Jobs

Suresh Naidu
View PDF

John Ahlquist’s essay proposes a notion of “decent jobs” that is multidimensional, context-dependent, and dynamic. I suggest a simple test for “decent work”: work that the worker thinks is valued by people whom the worker values (a measure of exemplary virtue, hence eudaimonic). High pay in a cash society is but one such signal of value, and much labor is provided without that signal. I advocate making relationship-building, decision-making, and sheer time and care required by democratic self-government part of what we consider “decent work,” a democratic corvée we impose on ourselves.

Suresh Naidu is Professor of International and Public Affairs and Economics at Columbia University. He has published articles in American Economic Review, The Journal of Political Economy, and The Quarterly Journal of Economics, and has a book on historical labor markets under contract with Harvard University Press.

When I think of “good jobs” or “decent jobs,” I envision the multidimensional, context-dependent, and dynamic elements John Ahlquist raises in his essay.1 I am going to offer a notion of “decent work” that is perhaps more parsimonious, but encompassing of Ahlquist’s thicker definition: the experience of decent work occurs when a person expends time and effort because they believe their various social communities think the sacrifice is valuable, even if they do not value it themselves. Work binds us to each other, making us do the things that others value even when they tax us physically, emotionally, and mentally. Making that experience “decent” involves eliciting consent, but also confirming and credibly signaling the value others put on our effort. We are doing dignified work when we think what we are doing is seen as worthy by others who are also worthy to us. This second-order belief, that other people important to us think our exertions are valuable, creates a thread running through all the dimensions of decent work Ahlquist describes. These signals of exemplary work are a version of classical “virtue,” and decent jobs as characterized by “eudaimonia”: it is not the actual content of the job that matters so much as whether we think the job is valued by the various communities we are attached to. Following this classical thread, I conclude by suggesting that we begin thinking about democratic political participation and self-government as a form of labor. It requires time, effort, and care, and we may wish to compensate it appropriately, particularly in a world where the market is no longer such a reliable source of meaningful work.

Money is one (however distorted) representation of community value, summarizing the economy-wide willingness to pay for work. Good work should pay well because it is valued by society at large, even by anonymous others many degrees of separation away. And it is difficult to proxy the relationship embodied by abstract cash with other less universal tokens, although not impossible (medals and plaques can go far in some contexts). Further, cash gives a worker a claim on the social product of market society, allowing one to secure housing and livelihood. Of course, work is not the only way to get cash, but in a cash society, wages are a credible way for society to express its gratitude for work done, and high-paying work naturally gets prestige. For many jobs, pay is valued because it is a signal of how the broader society values their work, not necessarily just the consumption facilitated by wages. And so, high pay is a cultural signal of good work, and when the work is unpleasant in some other dimension (rude customers, high training costs, sweltering outdoor heat), then the pay ought to be commensurate.

An immediate pathway to creating eudaimonic jobs, familiar to economists, is to “go global” and supercharge labor market—demand, decoupled from any particular social networks and compensated with money. But there are other paths. For example, we could “go local” and provide material supports for people to allocate their time in ways that are rewarded by their extant local communities. Implementing this path entails disconnecting material subsistence from work, via a variety of basic income and in-kind supports, and letting people use existing social networks and obligations to allocate their reciprocal labors. A third path, the most radical of the three options, is to create forms of work rewarded by a larger political community, for example, encouraging democratic self-governance as a form of public employment. In the process, we would generate new social networks, new valuations of effort, and new kinds of work that ought to have social value as inputs to self-determination, but do not have it now. In sum, the three paths leading to eudaimonic jobs are: 1) dignifying market work with stable and high wages, 2) dignifying nonwage work with material supports, and 3) expanding what counts as work to include the tasks of democratic self-government.

The first path, supercharging the labor market, immediately runs into worries about the future of labor demand. I have little long-term worries about work, in the sense of doing unpleasant things that have value primarily for others, becoming obsolete. I do wonder whether the labor market will continue to be the primary institution able to deliver this work. A precondition of good work in a capitalist economy is that privately owned firms want to hire workers on favorable terms. There are three forces pushing toward reduced market labor: lower labor demand, lower labor supply, and lower ability of private employers to deliver the sources of standing, self-respect, and well-being in a complex, intersectional society via work.

Among these forces, the first is reduced labor demand. Beyond the usual bogeyman of automation—which crops up every couple of generations to offer new post-employment dystopias and utopias—increases in consolidation, an increasing share of capital tied up in intellectual property monopolies, an inflation-averse central bank, climate (and geopolitical) disruption, and a general decrease in competition all work together to reduce the market demand for labor. Like all calamities, a drop in labor demand will first impact those from low-income backgrounds, and even if restricted initially to a few sectors, it will ripple through the labor market lowering wages throughout the distribution.

Government can do a lot here via macroeconomic management, antitrust, and directing technological change, but the proximate decision-makers are still businesses. Government might also need to do more spending and less regulating in order to stimulate private sector labor demand, a mix that has limited political support. But if we are going to keep the labor market as the primary device for allocating people to jobs, we will need to prod and activate entrepreneurs and corporations, the channels through which good jobs run. These will be the entities that will need to invest in businesses, hire staff and managers, and make work happen. Effectively, we need macroeconomic and microeconomic forces driving high labor demand for all workers. On its own, a tight labor market can partially ensure some of the freedom to quit and the freedom to disobey an employer that should characterize a good job.

But it may be that the future of good work is in the public sector. The large demand for public infrastructure, social insurance, and human capital investment provided by the state will generate the labor demand, funded out of taxpayer revenue, for construction workers, nurses, teachers, and home health care workers. Public sector and building trades unions, not exactly the most popular segments of the labor movement, have a disproportionately large role to play in ensuring workers retain an autonomous veto point in a political process governing the allocation and renumeration of their work.

The second force pushing toward reduced market labor is reduced labor supply. Considered over the life cycle, we are well past Keynes’s aspirations for his grandchildren. With life expectancies approaching eighty years, retirement ages staying constant at sixty-five, and rising educational attainment, we now have forty years of work distributed over eighty years of life, while in Keynes’s day, it was closer to fifty years of work over sixty years of life. People died of old age while working. Much of the twentieth century had a tragically large number of missing Black men retirees, which has improved since midcentury. Even hours per week have trended downwards with economic development, both within developed and developing countries. It seems difficult to believe that there is an arc toward more leisure in the aggregate, but it is there in the data.

It is possible this issue is temporary. We could be on the verge of a second industrious revolution. Just like the industrious revolution of the late eighteenth century, it would be driven by new consumer goods, immigration, and new forms of work. Powered by new tasks that leverage human minds, hands, and relationships, labor demand could expand enormously. Powered by climate change, economic inequality, and conflict, a younger, non-White, and less educated workforce in the relatively tranquil and secure markets of the industrialized countries could expand. Powered by innovations in health care, entertainment, and other commodities, people want money even more than free time or maintaining relationships, and they will be willing to work for it.

To manage flows of workers from diverse multinational childhoods, we will need a more active use of nonmarket and premarket institutions for labor market integration, including families and schools. Part of this need, I suspect, entails disrupting the segregated social networks of care that naturally form within class and categorical groups. As economist Gary S. Becker pointed out, higher educational institutions could be thought of as firms that provided so much general-purpose skill that workers paid to work there.2 Policy has a role in driving out the “low-road” of low-value-added educational institutions and expanding the “high-road” of schools and childhood interventions that boost the life chances of kids born to disadvantage at home and abroad.

But the third obstacle to a vibrant labor market is more elusive and points to the third option above. It could be that the work of the future is not able to deliver either the pecuniary, the cultural, or the social rewards as work of the past, at least for workers without a college degree. It could be that the levels and forms of compensation and job designs compatible with the profit-maximizing imperatives of firms’ own cannot ensure the social standing and validation that comes with good work, nor can they guarantee the freedom to collectively reconfigure a workplace. In a world where we interact with coworkers only on terms set by (or Zoom spaces owned by) our employers, the social networks at work that are the basis of workplace norms and shared expectations of employer fairness will erode, facilitating both wage inequality and arbitrary rule within an employer. Calls for “workplace democracy” seem like weak medicine in a labor market otherwise marked by workplace fissuring, fragmentation, and segregation.

The second path to eudaimonic jobs recognizes the signals of social value are manifold, and often local to a social subnetwork and an overlapping set of organizations. People take care of their children, their friends, and their friends’ children with little in the way of pay, because these are recognized as useful work, are intrinsically enjoyable, and are a coin of repeated exchanges within these networks. Volunteers in social and political movements and community organizations contribute their time to stuff envelopes or knock on unresponsive doors because these activities validate one’s identity in a network of peer perception. People spend countless hours playing social video games or arguing with people on the internet, and I suspect some of this comes from the networks of affirmation and recognition constituted in these worlds. Maybe giving surfers and Twitch streamers unconditional cash and having them entertain us with sick videos on the internet is an unrecognized utopia-in-the-making.

So, how about that radical third option? The third route to eudaimonic work is to transform the many communities that create and communicate the social value of unpleasant tasks. Prototype communities that engineer recruitment, retention, and social norms to reward work exist: the military does this effectively, as do some religious communities. But even Israeli kibbutzim, able to enforce redistribution with shared beliefs and social sanctions, have lost what egalitarian norms they had. Few explicitly collective organizations of human labor have remained large-scale, egalitarian, and persistent across generations. Rather than trying to make the existing workplaces the basis of democracy, one thought is that we begin emphasizing the eudaimonic jobs required by widespread and regular participation in politics: the work of democracy alongside the democracy of work.

What does this look like in practice? Instead of responding to low labor demand with a universal basic income or employer subsidies or trivial and tedious “make work projects,” why not a democratic jobs guarantee, where we pay people to engage in some forms of democratic politics, and mandate other forms of those same politics? Absence of work was the classical precondition for democratic participation, and by this light, a jobless future is possibly a huge democratic opportunity. But we need to think about new institutions for channeling all kinds of people into political participation, not just leaving it to the unrepresentative population of people who intrinsically like and have time for politics.

Going to meetings, debating collective decisions, learning about policies, talking to (and adjudicating among) experts are all enormous labor requirements. Doing them well earns one standing only in our political communities, which rarely overlap with our private ones. When work and care obligations increase suddenly (roughly around age thirty-five among my cohort) volunteer-driven political activity is among the first things to get dropped, and paying attention to politics is often next. Recognition, compensation, and norms that made the work of democracy seem valuable in many social networks would hopefully make the work of democracy yield some of the virtues of prosociality and self-discipline often imputed to market work. Maybe we wind up with a new democratic corvée, like an extended and well-compensated jury service, where citizens will surrender some of their labor to doing democratic governance, perhaps via sortition. Widespread deliberation with random collections of our co-citizens will hopefully wind up building more integrated social networks cross lines of class, race, and habitus. For all the discussion around democratic socialism, there is little thought given to the time, care, and attention—the work—required to implement a thoroughgoing democracy. A basic constraint on democracy, regardless of institutional design of elections, is that many of us are just too tired and busy to participate, and so, either professional politicians and bureaucrats or those without substantial family and work obligations are running things. We should seek an arrangement of work that gives us all the time, space, cognitive assistance, emotional support, social recognition, and, yes, money that we need to show up, understand each other, and disagree.


  • 1John S. Ahlquist, “Making Decent Jobs,” Dædalus 152 (1) (Winter 2023): 105–118.
  • 2Gary S. Becker, Human Capital: A Theoretical and Empirical Analysis (Chicago: University of Chicago Press, 1964).