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

Is There a Proper Scope for Markets?

Marc Fleurbaey
View PDF

Debra Satz’s brilliant essay highlights that it is insufficient to study markets in terms of efficiency and potential market failures, as they have deep effects on people and societies. This line of thought could inspire the project of building a general theory of social interactions, in which the specific properties of market transactions would be identified, and their influence on society at large, depending on the surrounding institutions and social structure, could be understood. In this brief essay, personal care provides an example of the complex ramifications of different arrangements for social interactions.

Marc Fleurbaey is Professor of Economics at the Paris School of Economics, and Research Director at the French National Centre for Scientific Research. He is the author of Fairness, Responsibility, and Welfare (2008), Beyond GDP: Measuring Welfare and Assessing Sustainability (with Didier Blanchet, 2013), and A Theory of Fairness and Social Welfare (with François Maniquet, 2011).

Debra Satz shows that discussing markets in terms of efficiency and potential market failures is insufficient, because markets have deep effects on people’s character, and may be morally noxious even in the absence of identified failures.1 Markets may amplify inequalities, or even generate harm, especially when one party suffers from weak agency. In Why Some Things Should Not Be for Sale, Satz identified problematic features that make some markets suspect.2 She broadens this perspective in her essay in this issue of Dædalus, reflecting on how the quality of social interactions, the level of social cohesion, and the health of democratic institutions may be influenced by the presence of market transactions in certain domains.

This line of thought suggests that we need a general theory of social interactions, in which the specific properties of market transactions are identified, and their influence on society at large can be understood, depending on the surrounding institutions and social structure. It does not seem that such a theory exists already, and working toward its development seems a worthy project. In this essay, I will try no such thing, but share some thoughts about possible bits of such a theory.

Consider the case of personal care for dependent persons, such as young children and elderly people. Personal care can be organized in various ways.3 The “family” way relies on a more or less consenting member of the family (typically, the mother or the daughter) to provide such care, with no direct remuneration, and with some expectation of positive feelings being part of the provision (with good effects on the person cared for). The “servant” way relies on hiring a person (typically female and unmarried) who becomes a second-tier member of the family and is expected to provide positive feelings at a lower level than the family member. The “private care” way involves qualified professionals who provide care in a separate facility, or sometimes at home, in exchange for fees that are paid to their private or public employer. The “socialized care” way is similar to the private one, except that the service is free, or heavily subsidized, usually with adjustment for the family’s ability to pay. It may or may not involve the free choice of the facility or service in the area.

These options combine market and nonmarket features to various degrees. Each generates specific relations between care provider and receiver, and more or less favorable conditions for certain features in these relations. Here I will consider five features that appear particularly relevant to human flourishing: 1) the feelings involved in, or emerging from, the interaction, 2) the form of reciprocity involved in the relationship, which may foster or undermine altruistic forms of mutual help, 3) multiple externalities inducing over- or under-provision and affecting social inclusion and democratic institutions, 4) internalities shaping people’s character and their ethos in social life, and 5) the presence of social hierarchies and power relations associated with various arrangements.

Consider the provision of positive feelings, which are especially important for the development of young children but are also important for the mental health of elderly people. It is impossible to make people feel for other people through extrinsic motivations. No matter how much payment is offered, someone cannot just manufacture a feeling or believe something in order to obtain the payment. This issue is not primarily due to the fact that feelings and beliefs are easy to conceal and hard to verify, because in fact, in a long-term relation, they are hard to conceal. The main reason why feelings and beliefs are not for sale is they cannot be controlled by the subjects. No payment can make you believe that the earth is flat if you believe it is not, or love a person if you do not. This is a key barrier to commodification, and explains why the family way will always retain an important place, especially for the care of young children.

But this is not meant to exclude the occurrence of good feelings in market-like relations. Actually, long-term contact between people is generally conducive to bonding, and as in the family context, such bonding may be beneficial or toxic in a variety of ways. The servant way can make the servant almost like a member of the family or, on the contrary, submit them to constant harassment and bullying.

A second important dimension of the relationship is the type and degree of reciprocity involved. Market relations are the archetype for direct and immediate reciprocity, whereas nonmarket relations allow for more altruistic interactions, where reciprocity is seldom totally absent but may be deferred or indirect. Compared with private care, socialized care may free the participants from the aura of reciprocity and conditionality of market relations. This flexibility may have positive and negative effects. The positive effects come from the fact that since there is no direct payment, the occurrence of good-quality relations is easier to interpret as reflecting genuine positive feelings and exert a good impression on the care recipient and their family. The negative effects come from the possibility that the absence of conditionality may make the care provider feel more empowered to treat the subjects in an unkind way. The option to choose the care provider within the socialized care setting may reintroduce some conditionality in the relationship and, at the same time, protect against abusive behavior while undermining the possibility to interpret kindness as genuine.

With this second dimension (reciprocity versus spontaneity), we see that market relations reduce the temperature of relations, making it hard to believe that kindness is genuine. Reducing the temperature is good when it replaces violent conflict over resources with doux commerce, but it is less attractive when it undermines the development of positive feelings and altruistic relations.

A third dimension in the analysis of social interactions is the generation of positive and negative externalities. Any interaction between some parties may affect third parties that are not part of the interaction, and this holds for market as well as for nonmarket interactions. One can, in the abstract, determine conditions under which externalities are either avoided or balanced so that, on the whole, positive and negative externalities of a particular action or transaction cancel out. But there is probably no simple recipe to handle the externality problem in the great variety of contexts in which it may arise. In the case of care, whether in the private-care or socialized-care context, there is the risk, as Satz notes for schools, that free choice may generate segregation and reinforce social stratification. There is also the possibility that social convention and conformism may lead to excesses. Interestingly, excesses can occur in any direction. There can be too much reliance on family care, forcing women to devote their time and energy to care work instead of pursuing their personal plans. When this overreliance occurs, developing a market for private care may be liberating, but expanding socialized care is likely to be even more liberating, especially for poorer segments of the population.

The opposite excess can occur when people are so focused on their personal flourishing that they rely too much on external care for their dependents. This other extreme may lead to impoverished social interactions and especially a stunting of emotional development for young children, but also a depressed end of life for elderly people whose contacts with family members are drastically reduced when they become dependent on external care. There can also be excessive reliance on servants, with the development of a labor market for such services in which employees are at great risk of abuse. Externalities underlie such excesses, because the dominant option in one’s society is thereby made easier and cheaper, and the pressure of social conformism tends to entrench any dominant option.

One can add a fourth dimension of “internalities” to the picture: namely, that certain interactions shape or transform the character of the parties involved, with spillover consequences for all spheres in which character traits may matter. Satz highlights the impact of subjugation in labor relations on the ability or disposition of citizens to participate actively in democratic life.

A related fifth dimension consists of the presence of asymmetric social roles being associated with certain market trades. While markets for slaves and for bonded labor operate underground, many legal markets do involve the submission of one party to the power of another party or to various forms of risk. In the case of care, the fact that a servant is supposed to obey their employer generates dangerous cues for the parties involved in the relationship, and abuse of power is commonplace in such contexts, as in any workplace, but with the additional risk associated with intimacy in a private home. The private and socialized care settings provide a more neutral ground for relatively cold but also relatively safe relations, although abuse scandals are repeatedly occurring in these settings as well. The private care option generates a customer-provider relation that involves less social hierarchy than the employer-servant relation.

However, this is complicated by the fact that there may be different relations at different levels. The person receiving care might not be the direct customer, and the person delivering care might not be the paid provider but a hired employee. The association between certain markets and certain social hierarchies can be regulated in order to protect the weak party. The regulation of nonmarket relations in which similar asymmetries occur is also possible, although indirect interventions, such as the provision of alternative options to the weak parties, or reforming social conventions through educational campaigns, can also be powerful while being less invasive.

In conclusion, these five dimensions, by no means exhaustive, appear relevant in the analysis of the proper scope of market-like features (such as payment or free choice) in social interactions. First, the quality of relations depends very much on feelings involved in, or emerging from, the interaction, and these can be deeply affected by the presence of market-like features. Second, by promoting direct reciprocity, market-like features lower the temperature of relations and may protect the subjects from certain forms of abuse but also undermine the development of altruistic forms of mutual help. Third, many externalities can arise and lead to insufficient or excessive reliance on market-like features. And such market-like features can reinforce segregation and social stratification, or threaten democratic institutions. Fourth, character-shaping (internalities) may have strong consequences outside the transactions in which these internalities arise. Fifth, social hierarchies associated with certain market and nonmarket interactions can be problematic and warrant various forms of regulation and intervention. These five dimensions are interdependent: for example, internalities depend very much on feelings, and so on.

The general picture that is likely to emerge from further research on social interactions will be complex, as Satz suggests, since there is no clear-cut divide between market and nonmarket relations, and market-like features have positive and negative effects that very much depend on the social environment. In the current context of massive transformations in markets and networks, the value of pursuing this research agenda cannot be overstated, and Satz must be lauded for having led the way.


  • 1Debra Satz, “Democracy & ‘Noxious’ Markets,” Dædalus 152 (1) (Winter 2023): 179–188.
  • 2Debra Satz, Why Some Things Should Not Be for Sale: The Moral Limits of Markets (New York: Oxford University Press, 2010).
  • 3In this issue of Dædalus, Alison Gopnik makes a vivid plea for recognizing the value of personal care and for facilitating the provision of care by relatives. Alison Gopnik, “Caregiving in Philosophy, Biology & Political Economy,” Dædalus 152 (1) (Winter 2023): 58–69.