How should a new political economy conceive of the role of markets in a just society? Markets clearly play an important role in efficiently allocating labor and goods, disseminating information, enabling cooperation among people who disagree with one another about how to live, and allowing individuals ‘ choices about where to direct their talents and resources. But acknowledging that markets play an important role does not mean that this role is simple or conforms to the status quo in capitalist countries like the United States. In this essay, I draw on classical and modern ideas to defend a limited role for markets that is tempered by democratic concerns.

The classical political economists - such as Adam Smith, David Ricardo, and Karl Marx - held a multifaceted view of markets. For these thinkers, markets not only efficiently distributed goods and services, but also simultaneously shaped our relationships with one another, supported or undermined valuable political institutions, and affected our capacities, including our capacities for collective self-governance. These early theorists of markets were especially attuned to the ways that markets can affect relationships of freedom and equality between members of society. Indeed, Adam Smith singled out “order and good government, and with them, the liberty and security of individuals” as by far the most important effect of the widespread use of markets.1 According to Smith, markets advance freedom by replacing workers’ abject dependence on one powerful lord with exchanges involving a thousand different customers, none of whom has substantial power over them.2

But Smith also understood that some markets posed a threat to the more egalitarian social relationships that feudal critics aspired to. For example, he observed that the labor market, unlike a market for apples or widgets, shapes the capacities and preferences of those human beings whose labor power is purchased. As he wrote in The Wealth of Nations:

The man whose whole life is spent performing a few simple operations of which the effects too are perhaps always the same … has no occasion to exert his understanding or to exercise his invention in finding out expedients for removing difficulties which never occur. He naturally loses, therefore, the habit of such exertion and generally becomes as stupid and ignorant as it is possible for a human creature to become…. [He is incapable] of forming any just judgment concerning many even of the ordinary duties of private life. Of the great and extensive interests of his country, he is altogether incapable of judging.3

From this perspective, it is critically important to evaluate labor markets through a wider lens than efficiency. Even if a market is efficient at allocating human labor power between different uses, it is problematic if it fails to develop or atrophies the psychological affective and cognitive capacities of workers that are needed to maintain democratic institutions. This evaluative lens can and should be extended: markets raise other considerations pertaining to democratic social relationships beyond their important effects on human capacities. Cases to consider include the expansion of markets into legal representation, health care, and education, each illustrating the problems democracies confront when they rely on specific markets.

To set the stage for my argument, I need to make three preliminary points. First, the problems I will point to are not primarily problems of “market failure.” Contemporary economists are well attuned to the ways that actual markets can be problematic because of their distance from “ideal” markets. Natural monopolies, asymmetric information, and incomplete contracts render many actual markets inefficient, and may justify forms of regulation. For example, if asymmetric information is a problem-think of the knowledge imbalance between borrowers and lenders with respect to subprime loans-it might justify interventions aimed at increasing the information of the more vulnerable party, or caps on the rate of interest. The existence of natural monopolies might justify forms of public ownership. The presence of negative externalities like pollution might be addressed by adding additional markets to fully capture pollution's third-party costs in prices. Innovative work in economics has sought to address the ways that non-ideal markets can be made to function better-where better generally means “with greater efficiency.”4

But even “ideal” efficient markets can raise ethical concerns. Consider a market in votes. Even if all the parties participating in a vote market are fully informed, equally positioned in terms of market power, and even if the trade makes the parties better off in terms of their individual preferences (for example, I have a vote I don't care to use, and you have a strong desire to politically influence an election outcome, so selling my unused vote to you seems to be a win-win), a market in votes would distort an important norm of democratic decision-making: that is, one person, one vote. Democratic voting is the main way in which we aggregate individual views about what policies society should enact. Selling votes involves voters exchanging an asset they do not fully own as individuals.5

Second, there are cases in which the problems associated with a market arise solely because of the background circumstances the market operates within. My earlier work called attention to two dimensions of this social background.6 The first dimension is the degree of autonomy or agency of the market's participants. In some markets, some or all participants lack relevant information about the trade they are making. Consider the market in used cars, the market in subprime derivatives, or the market in health care. Agency can also be weak when market transactions are made on behalf of or with effects on those who themselves are not participants in the market. Consider the third-party effects of markets in “blood diamonds” that fuel bloody civil wars.7

The second dimension concerns highly asymmetric market power. Consider “price gouging” in disaster areas, the international trade in toxic waste, and markets in human organs like kidneys. In typical labor markets, employers also have substantial power over workers. The reasons for this power are manifold. It is generally easier for employers to find new workers than for workers to find new employers. Leaving or losing a job is often greatly disruptive to workers and their families. New technologies fissure markets, turning workers into independent contractors whose work appears to be controlled and coordinated only through a platform, making it harder for workers to organize. Workers often feel loyalty to employers, coworkers, and even customers. While goods like widgets do not care where they are located, human beings have complex preferences over their workplaces, colleagues, and living arrangements, making them vulnerable to exploitation.

In markets with weak autonomy, or where power relations dramatically skew the outcomes that the parties will accept, background circumstances might be said to “infect” the morality of the market. I have referred to such infected markets as noxious markets.8 Of course, in many such cases, simply banning the market will not address the underlying moral problems, but merely drive them underground. (For example, even though kidney-selling is illegal in almost every country in the world, there is a thriving black market.) Nonetheless, we cannot ignore the moral problems raised by the “normal” operation of such noxious markets. Instead, we need to think through targeted ways to address those background problems. In what follows, I largely set these concerns about background agency and power aside and address another important dimension of a noxious market: its effects on democratic culture and institutions.9 These effects can hold sway even when there are neither market failures, weak agency, nor highly asymmetric power.

Third, my argument is not a lawyer's brief against markets. No large democratic society can or should entirely dispense with markets. Not only are markets among the most powerful tools we have for generating growth in living standards and incentivizing innovation, but also Smith was right to see their democratic potential as ways of enabling cooperation among independent, free, and equal individuals. As tools, however, we should think carefully about where to use them and how to design them when we do. While a neoliberal worldview sees efficient markets enhancing freedom and well-being everywhere, the reality is more complex. Some markets foreclose options that would better support democratic institutions and culture. Sometimes, closing off market options makes everyone better off. Consider that if individuals are free not to purchase health insurance on the market, the cost of publicly provided insurance will increase: healthy individuals are more likely to opt out of health insurance, leaving sicker individuals in the pool to be insured and raising the costs of their insurance, leading more people to forgo holding such insurance, driving the prices up even higher.

What are the democratic problems posed by using markets in certain domains? Consider education. Market enthusiasts have long advocated for treating the allocation of children to schools as an economic problem. Although Milton Friedman did not go so far as to deny some role in education for the state-because of the third-party effects of an uneducated citizenry-he argued that parents should be given “vouchers redeemable for a specified maximum sum per child per year if spent on ‘approved’ educational services.”10 Parents then would be free to spend this sum, in aggregation with their own private money, on the school of their choice. Advocates of voucher systems argue that it would generate school competition, as parents attempt to gather information about and place their children in the best schools available. Competition gives schools an incentive to improve to attract children.

While the data on whether school choice improves school quality remain contested, I want to point to a different set of concerns: the way private goals, empowered by the market, can diverge from, prevent, and undermine the state's interest in achieving social integration. The divergence arises because parents generally care about the best interests of their own children, and as individual decision-makers, they tend to prioritize those interests. Indeed, in one sense, it is entirely appropriate that they do so. Society relies on parents to act as trustees for their children and to do what conduces to their children's flourishing. At the same time, some of the ways parents prioritize their own children can lead to worse outcomes for other children and to the furthering of educational inequities, as well as to other social ills like instability and conflict. Evidence indicates, for example, that choice schools in the United States are more homogenous than public schools with respect to social class and race. Researchers have also shown that when public school choice is available, educated parents are especially likely to factor child demographics in their school selections.11 This may be because school quality is very hard to judge and parents default to markers such as the reading and math levels of other students. These levels, in turn, are heavily influenced by social class. It is likely that some parents take race and class directly as proxies for school quality.

Even if we assume that the outcomes that arise from a voucher system satisfy many parents’ individual preferences as trustees for their own children, does not a democratic society have an interest in overcoming racial and class divisions?12 Strong class and racial divisions give rise to social instability. And these divisions often arise from and help to maintain injustices.

With respect to racial and class division, one of the most powerful and robust pieces of social science evidence we have is psychologist Gordon W. Allport's contact hypothesis: familiarity reduces prejudice under certain conditions.13 The repeated exposure of children to those who differ from them in terms of race and social class is the most powerful antidote that we have to bigotry, groupthink, snobbery, and elitism. Common schooling is also an important tool in delivering fair equality of opportunity. Friedman's own voucher system, which allows parents to add their own wealth to the value of their voucher to purchase access to more desirable schools, would replicate or worsen the unequal funding we see in America's public schools today. But no individual parent can end racial and class divisions, or maintain social stability, or achieve fair equality of opportunity, by acting on their own. Market ordering in this case makes a more “socialist” form of institutional design impossible.14

My point here is that if we want schools to serve as a source of unity and cohesion, as well as deliver fair equality of opportunity, we will need to look elsewhere than to market-based solutions. Vouchers as imagined by Friedman would produce-and in many cases already have produced-greater economic, racial, and religious segregation. In the case of public schooling, closing off individual exit may produce a better social outcome than that achieved through disaggregated individual decisions. Empowering parents as individual decision-makers over the choice of schools leads them to act in ways that undermine equitable schooling across race and class lines.

This is not to deny that current social arrangements, especially exclusionary zoning, already throw up barriers to the “common school” ideal. Additionally, I recognize that this argument has some controversial consequences; in particular, it suggests the banning of private schools, which perpetuate and accentuate the class stratification of society over generations.

Suppose, however, that it is true that democratic social arrangements are threatened when people in a society live completely different lives and rarely interact with one another, except in circumstances where some are the subordinates of others. Shouldn't that at least be a factor in evaluating the education of citizens? We can take a page here from the British socialist R. H. Tawney, who earlier criticized the system of private (“public”) British schools:

It is at once an educational monstrosity and a grave national misfortune. It is educationally vicious, since to mix with companions from homes of different types is an important part of the education of the young. It is socially disastrous, for it does more than any other cause, except capitalism itself, to perpetuate the division of the nation into classes of which one is almost unintelligible to the other.15

There are other examples in which a democratic society has reasons to aim at the common provision of important social goods and to curtail private options for individuals. Let me highlight a different case-military service-before revisiting Adam Smith's concerns about labor markets’ effects on human character and capacity.

For much of American history, serving in the military was seen as a core obligation of citizenship (although initially only for men). That view changed in 1973 when the draft was abolished and an all-volunteer force was established. In 1960, England abolished its own system of conscription after some fits and starts. Many well-known factors led to ending the draft in the United States, including an unpopular war in Vietnam. Far less well known is the fact that Milton Friedman himself played a significant role in the ending of military conscription.16 Friedman and his followers successfully argued to President Richard Nixon that the draft was the equivalent of indentured servitude. Friedman denied-adamantly-that such service was an obligation stemming from the civic duties of national membership, and his argument carried the day. The result is that in the United States-and in most developed democracies-military service has become voluntary, yielding a smaller and more focused group of enlistees, enticed at least in part by market considerations. Rather than being seen as a collective duty that all citizens must share, military service is now seen as another private decision.

Extending the reach of markets even more, war has been further outsourced to private military contractors: in 2009, there were more private military contractors in Afghanistan than U.S. military troops.17 Hiring private mercenaries and outsourcing national security to a subsection of our population might spare our citizens, but as political philosopher Michael Sandel has noted, it changes the meaning of citizenship.18 In what sense are we “all in this together” if most citizens never need to think hard about decisions to go to war? Whatever the efficiency pros and cons of the decision to outsource fighting and allocate military service through market means, doing so changes our relationships with one another and our sense of a common life.

My argument so far suffers from treating the state and market as two stark alternatives for the allocation of goods and services in society. So I now want to consider ways in which the benefits of markets can be harnessed-through design-to better serve important democratic goals. Return to Adam Smith's surprising claim that when we treat labor markets simply as we do apple markets, we will find forms of work where workers’ cognitive and affective capacities are stunted.19 While we do not typically have to worry that the market production and distribution of apples have dramatic negative effects on the character and quality of apples, human beings are different.20 According to Smith and many other classical political economists, labor markets are constitutive: the work we do also makes us.

A large body of research has found evidence in support of that claim: workers who simply functioned as cogs in machines, with little or no discretion at work, were more likely to become passive outside of the workplace. Writing in the 1960s, sociologists Gabriel Almond and Sidney Verba found that, across occupational types in five countries, those who were consulted about their opinions on the job scored significantly higher on a measure of subjective civic competence.21 Their study looked in particular at workers’ sense of personal efficacy in dealing with government bodies. Almond and Verba suggest that surrender of authority to employers at work is not good for civic engagement, or for workers’ sense that they can contribute to the improvement of their society.22

These social effects of workplaces can be attenuated without abandoning a critical role for markets in the allocation of labor. One important mechanism is providing greater roles for worker voice. This can be done through such reforms as changing labor laws to support forms of worker association, like trade unions, allowing worker representatives on company boards, and strengthening democracy at work through diverse forms of ownership including worker-managed and -owned firms. Empowering the associational organization of labor would also help redress the background social conditions that render workers vulnerable to the oligarchic power of their employers.

There are other examples in which careful design and policy can limit the “noxiousness” of a particular market for democracy. Policies such as a negative income tax can strengthen the power of workers, and campaign finance laws can diminish the power of money in elections. Others have argued for reforms to our current system of commodified legal representation within an adversarial system, and for single-payer health care systems.23

Markets have an important-even indispensable-place within modern heterogeneous large societies. But that place is bounded: we need a larger frame than efficiency and market failure for thinking about the kinds of markets we want and where we want to use them. I have argued that capitalist market relations-emphasizing efficiency, individual decision-making, heterogeneity, and decentralization-are not appropriate for certain kinds of social decisions. I have sketched the case for this conclusion by considering the allocation of children to schools. Democracy's promise is, after all, that we are each other's social equals. Delivering on that promise requires that schooling be available to all students in a manner consistent with John Rawls's formulation of fair equality of opportunity: “equal chances of education and culture for persons similarly endowed and motivated.”24 It is inconsistent with an educational system that separates rich and poor, and Black and White into different schools with unequal resources; and thereby perpetuates racial inequality, snobbishness, and servility.

Beyond education, we need to pay special attention to particular markets that affect democratic functioning and stability. Such markets include but are not limited to markets in legal representation, media and news markets, markets relating to national defense, and markets governing political rights. Politicians and other commentators usually write unreflectively, as if all markets were the same. They are not. Markets affect not only the distribution of income and wealth, but also our capacities, and our views of each other. Their strengths but also their limits depend on the fact that they are radically individualizing. But in some contexts, that individualizing threatens the practice of democracy. Markets have moral and even “spiritual” consequences relevant to our shared public life, and our evaluations of them must also attend to those consequences. A new political economy needs to take this larger evaluative frame into account.

Thanks to Margaret Levi, Marc Fleurbaey, Chloe Thurston, and to other members of the “Moral Markets” group at the Center for Advanced Study in the Behavioral Sciences at Stanford University for their helpful comments and discussion.


Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations, Volume 1 (Indianapolis: Liberty Classics, 1976), 412, my emphasis.


Ibid., 420.


Ibid., 781-782.


Indeed, one common solution to market failure is to find ways to add new markets that will take into account the externalities. If society cares about the negative effects of pollution, regulators can design mechanisms to force polluters to take those costs into account.


In Margaret Jane Radin's words, voting relates to a democracy community's normative life. See Margaret Jane Radin, “Market-Inalienability,” Harvard Law Review 100 (8) (1987): 1854,


Debra Satz, Why Some Things Should Not Be for Sale: The Moral Limits of Markets (Oxford: Oxford University Press, 2010).


This example is from Ravi Kanbur, “On Obnoxious Markets,” in Globalization, Culture, and the Limits of the Market: Essays in Economics and Philosophy, ed. Stephen Cullenberg and Prasanta Pattanaik (New Delhi: Oxford University Press, 2004).


Satz, Why Some Things Should Not Be for Sale.


My book identifies four dimensions of a noxious market: weak agency, highly unequal power, harm to individuals, and harm to democratic social institutions. These four dimensions do not always coexist.


Milton Friedman, Capitalism and Freedom (Chicago: University of Chicago Press, 1962), 89.


Jack Dougherty, “Shopping for Schools: How Public Education and Private Housing Shaped Suburban Connecticut,” Journal of Urban History 38 (2) (2012): 205-224,


I am assuming that the state does indeed have this interest.


Gordon W. Allport, The Nature of Prejudice (Cambridge, Mass.: Perseus Books, 1954).


It is of course possible to use a weighted formula in determining the worth of vouchers. That said, economic and racial integration are properties of schools and not individuals. So, an “individualist” solution to integration that proceeds via parents’ choices seems insufficient.


R. H. Tawney, Equality (London: Unwin Books, 1952), 158.


Friedman's role in ending the draft is described in the first chapter of Binyamin Appelbaum's excellent book The Economists’ Hour: False Prophets, Free Markets, and the Fracture of Society (New York: Little, Brown and Company, 2019).


See James Glanz, “Contractors Outnumber U.S. Troops in Afghanistan,” The New York Times, September 1, 2009,


Michael J. Sandel, What Money Can't Buy: The Moral Limit of Markets (New York: Farrar, Straus and Giroux, 2012).


Smith, An Inquiry into the Nature and Causes of the Wealth of Nations.


While recognizing the effects of industrial production, such as the use of pesticides.


Gabriel A. Almond and Sidney Verba, The Civic Culture: Political Attitudes and Democracy in Five Nations (New York: SAGE Publications, 1963).


See John Stuart Mill, Principles of Political Economy with Some of Their Applications to Social Philosophy (New York: D. Appleton and Company, 1848).


Shai Agmon, “Undercutting Justice: Why Legal Representation Should Not Be Allocated by the Market,” Politics, Philosophy & Economics 20 (1) (2021): 99-123,


John Rawls, A Theory of Justice (Cambridge, Mass.: Harvard University Press, 1971), 275.

Published under a Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license, please visit