Introduction
That markets play an important role in modern democratic states is no longer seriously disputed. Their reach, however, is hotly contested. A significant debate has developed around the question: What are the moral limits of the market? It is argued that votes, kidneys, women's sexual labor, and much else, ought not to be for sale.
Consider two very different explanations for such limits. The first appeals to facts about specific goods to justify restrictions. A position of this sort is taken by Elizabeth Anderson (Reference Anderson1993) who defends an account of the market's domain based on a socially grounded pluralistic theory of value. She holds that certain goods, if their value is to be fully realized, require suitable social conditions. To secure these conditions, some goods may need to be excluded from the market. The second explanation, defended by Debra Satz (Reference Satz2010), holds that limits are appropriate when the operation of specific markets, whatever the good being exchanged, exerts undue influence on citizens’ equal status. To discern which specific markets have this unwelcome effect, Satz proposes four general parameters. Markets that are extremely harmful for individuals or for society or that involve participants with diminished agency or those who are vulnerable are deemed ‘noxious’ and become candidates for intervention.
Although the content of their views differs in important ways, Anderson and Satz approach the task of market assessment in much the same way. The normative significance of a specific market is taken to be a function of the normative significance of the transactions it comprises. This approach is a natural one. If every transaction in the kidney market inflicts intolerable harm on vendors, then that market inflicts intolerable harm. If every transaction in the market for women's sexual labor is degrading to women, then that market is degrading. The case against specific markets is thus made by appeal to the objectionable consequences so adduced.
The market also has its champions. Anderson, Satz, and other ‘anticommodification theorists’ are the target of a recent entry to the debate offered by Jason Brennan and Peter Jaworski (Reference Brennan and Jaworski2016). Their position is that ‘if you may do it for free, then you may do it for money’ (2016: 10). They deny there are inherent limits on what may be sold, and thus they deny that specific markets are morally suspect. For anything that is permissibly given, they maintain, there is some suitable market arrangement under which that thing may be permissibly sold. As with Anderson and Satz, on this approach, the status of a specific market arises from an assessment of its transactions. Because kidneys and sex may be freely given, Brennan and Jaworski contend, markets in such goods are, in principle, permissible, and if they are suitably regulated, they are acceptable in practice.
I shall argue that both sides of this debate adopt the same deficient approach to market assessment. Here is what will inevitably be a somewhat cryptic statement of the error: participants illicitly proceed from an assessment of the transactions making up a specific market to a judgment of that market's moral status. This inference is unlicensed. It mistakenly reduces the normative significance of a specific market to that of its transactions. Yet, we may know everything there is to know about the transactions in a specific market but still not know that market's moral status. To discern this—to establish that an existing market should be regulated or closed or that a currently prohibited market should be permitted—one must supply a rather different assessment. What is required is a comparison of two complete states of affairs. Intervention into the market is justified only when the result is expected to mark a moral improvement over what would transpire in the absence of said intervention. None in this debate has offered such a judgment. As a result, none occupies a defensible position.
I should stress that the challenge I present here is internal to the accounts on offer. Anderson, Satz, and Brennan and Jaworski each holds a different view of what is normatively relevant for market assessment—each adopts his or her own evaluative standard. I take no position on which, if any, of these evaluative standards is correct. Rather, I highlight an error arising from the approach to market assessment on which all proceed. Thus, I do not offer an alternative account of the market's moral limits, but instead I identify a mistake afflicting the most prominent proposals. Although my central claim is negative, it has far-reaching implications for the commodification debate.
This paper is organized into seven sections. The first elaborates the work of Anderson and Satz, and the second identifies a problem facing both accounts. Neither is action-guiding. Section 3 traces this problem to a confusion between two ways a specific market may be assessed. Section 4 presents Brennan and Jaworski's defense of markets. Their account, in section 5, is revealed to be predicated on the same mistake as that of their opponents. Section 6 illustrates the practical significance of the mistake, which is considerable. Section 7 concludes.
1. For Markets with Limits
Anderson and Satz defend markets with limits. They disagree about what is normatively relevant for market assessment and thus accept different evaluative standards. But they both adopt the same approach to the task. To arrive at a conclusion about a market's moral status, both supply an assessment of that market's transactions. The details of their accounts make this crucial commonality clear.
1.1 Anderson's Account
Anderson argues that freedom and autonomy—values often claimed to recommend an expansive free market—in fact impose significant limits on its scope. Her view is doubly pluralistic: it acknowledges that there are many legitimate conceptions of the good life, and it holds that different kinds of goods are rationally valued in different ways. This value pluralism informs her understanding of freedom and autonomy. She holds that part of what it means to be free is to be able to value many things properly and part of what it means to be autonomous is to endorse one's evaluations (1993: 141). On this proposal, the market poses a problem because many goods can be properly valued only in suitable social contexts, and these contexts may be eroded or degraded when subject to market forces.
This concern for freedom and autonomy, so understood, is expressed in Anderson's evaluative standard:
Liberal Values. Market activity must not interfere with people's ability rationally to value certain goods or undermine individual or collective autonomy.
As Anderson puts it, ‘the state has a case for prohibiting or restricting commodification of a good if doing so increases freedom—significant opportunities for people to value different kinds of goods in different ways—or if it increases autonomy; that is, the power of people to value goods in ways they reflectively endorse’ (1993: 154).
There is an element of essentialism in Anderson's account. It is by reflecting on the nature of goods that we come to see how they are properly valued. Accordingly, the norms that govern a good's appropriate use and distribution may be tightly linked to its distinctive features. But notice that the focus here is on the attitudes one should take toward the goods. And crucially, the attitudes one takes toward a good are only contingently related to whether one pays for it. As Anderson writes, ‘what confers commodity status on a good is not that people pay for it, but that exclusively market norms govern its production, exchange, and enjoyment’ (1993: 156). The point to emphasize is this: that a good is exchanged on a market does not necessitate that it is inappropriately valued; likewise, it is not necessary that a good be exchanged on a market in order to be valued inappropriately.
This feature of Anderson's view has an important implication. She offers an account of the market's limits, but she is not exclusively or primarily concerned with the exchange of money. Rather, she focuses on the undue attitudes contingently arising when certain goods are subject to inappropriate norms. Thus, when a specific market violates Liberal Values, this is explained by the attitudes engendered by its operation not by the mere fact that money was exchanged.
By way of demonstration, consider Anderson's treatment of markets in women's sexual labor. Her assessment focuses on the transactions in that market and their consequences for Liberal Values. Prostitution, she claims, is ‘the classic example of how commodification debases a gift value and its giver’ (1993: 154). Central to her assessment of this market is an account of the special value one finds in personal relations. The ideals of intimacy and commitment play a special role in the personal sphere, influencing how people value each other as well as how they value the goods exchanged there. Gifts, as shared goods, are not properly valued as mere commodities. They are degraded when subject to market norms. Thus, by attending to the consequences of permitting a market in sexual labor, Anderson builds a case against it. Because the transactions making up the market violate Liberal Values, she reasons, the state may justifiably intervene.
1.2 Satz's Account
Prominent in Satz's work is the claim that markets shape who we are and who we become, an insight she traces back to classical economists, Adam Smith and David Ricardo. This concern is reflected in her evaluative standard:
Equal Status. Market activity must not undermine the conditions necessary for citizens to interact as equals.
Satz elaborates the role of Equal Status with reference to a status-oriented conception of citizenship developed by T. H. Marshall (Reference Marshall, Manza and Sauder2009). On Marshall's view, citizens are entitled to equal political rights and freedoms, but also to the material conditions of their exercise. One of Marshall's three conditions for equal citizenship is ‘the social element’, which includes the ‘right to share to the full in the social heritage and to live the life of a civilised being according to the standards prevailing in the society’ (2009: 149). When markets, even those that are perfectly efficient, prevent citizens from sharing in society ‘to the full’, Equal Status requires intervention.
On Satz's proposal any specific market may become noxious. To discern which markets qualify, Satz offers four parameters for market assessment. Markets that score highly on even one parameter are thought to violate Equal Status and thus become candidates for state intervention. The first two, harm to individuals and harm to society, focus on the consequences of some markets. Regarding individuals, the worry is that some outcomes are so harmful that citizens’ basic needs are left unmet. The concern about harm to society takes a wider view. Child labor, for example, produces uneducated quasi citizens, incapable of fully participating in society.
The second two parameters, weak agency and vulnerability, characterize the conditions out of which some markets emerge. Markets feature weak agency when their participants lack important information or adequate agential capacity, for example, when goods are marketed to children. The next parameter, vulnerability, characterizes markets whose participants bargain from disparate positions. Some participants may not be able to exit the market because the goods sought are necessary, as food items in the case of famine. Faced with few options, vulnerable participants may agree to almost any terms. Such desperate exchanges, Satz maintains, can quickly erode citizens’ equal status.
To illustrate the application of Satz's account, consider her assessment of the market in women's sexual labor. Her approach should be familiar. She focuses on the market's transactions and their consequences for Equal Status. While many sex workers are profoundly mistreated, Satz denies that the sale of sex is intrinsically degrading (2010: 153). She invites us to contrast a young ‘streetwalker’ who is desperate, subject to violence, and under the control of a pimp, with ‘a Park Avenue call girl’ who ‘engage[s] in what seems to be a voluntary activity, chosen among a range of decent alternatives . . . without coercion or regret’ (2010: 138). The weak agency, vulnerability, and individual harm associated with the streetwalker are not concerns for the call girl. Societal harm, however, is salient in both cases. A central thesis in Satz's treatment of this topic holds that markets in women's sexual labor contribute to attitudes that undermine the status of women as a class. The problem is that the market interacts with norms, attitudes, and other features of the social environment in ways that are inconsistent with Equal Status. Here, like Anderson, Satz arrives at an assessment of the market by attending to the consequences of its operation. Because the transactions making up the market violate equal status, she reasons, the state may justifiably intervene.
Both Anderson and Satz argue that some things ought not to be for sale. Although they appeal to different evaluative standards for justification, they take the same approach to market assessment. Both draw conclusions about the moral status of specific markets on the basis of a judgment of the transactions making up those markets. While this approach may initially appear reasonable, I now suggest that it is mistaken.
2. The Problem of Action Guidance
In this section I show that the proposals offered by Anderson and Satz face what I will call the problem of action guidance. Once their accounts have been applied to a specific market and a judgment has been rendered, urgent questions about that market's status remain. In the next section I offer a diagnosis of the problem.
To see what makes the problem acute, consider first what is, presumably, an uncontroversial condition of adequacy for an account of market assessment:
Not-worse. An account must not recommend actions that by its own evaluative standard are expected to make things worse.
The appeal of Not-worse is apparent. Whatever one's evaluative standard, the verdicts one's account delivers about specific markets should not make things worse from the perspective of those same values.
Not-worse is an extremely weak condition of adequacy; it requires consistency. It does not require that an account's recommendations always produce the best outcome. However, it is not vacuous. An account can easily fail to satisfy Not-worse because specific markets can be linked in such a way that a single act of intervention—whether regulation or prohibition—could have considerable unexpected consequences. In some cases, to intervene in a market that appears corrosive to some value may in fact make things worse from the perspective of that very same value. Accounts recommending such intervention fail to satisfy Not-worse.
This complication—the worry that intervention into a specific market may be counterproductive—is a familiar one. It may be instructively compared to the seemingly intractable problem Margaret Radin (Reference Radin1987) terms ‘the double bind’. The problem arises because we may regard a specific market as objectionable, but regard its alternatives with equal horror. In the context of markets in women's sexual labor, Radin frames the dilemma as follows: ‘If we now permit commodification, we may exacerbate the oppression of women—the suppliers. If we now disallow commodification . . . we force women to remain in circumstances that they themselves believe are worse than becoming sexual commodity-suppliers’ (1987: 1916–17). The problem, in short, is this. Certain markets, such as those in women's sexual labor, have mixed consequences. One may examine every transaction and conclude correctly that each is objectionable. Yet, one may imagine conditions as they would be absent this market and conclude correctly that all affected parties would fare worse. Thus, neither permitting the market nor prohibiting it appears entirely satisfactory. As long as the comparative merits of these options remain unknown, intervention in the market may make things worse.
A similar dynamic is discernible in the work of Anderson and Satz. Return to Anderson's treatment of markets in women's sexual labor. After explaining why prostitution is objectionable, she acknowledges that it may provide for the urgent needs of some women and so concedes that her opposition to the market is not unqualified: ‘If the prohibition of prostitution is to serve women's interests in freedom and autonomy’, she explains, ‘it should not function so as to drive them to starvation. It can serve these interests only where expanded economic opportunities eliminate women's need to resort to prostitution’ (1993: 156). Thus, on Anderson's account, allowing the commodification of women's sexual labor is thought harmful but so too is prohibiting it.
Two related observations are in order. First, notice that Anderson's response is thoroughly unsatisfying. When confronted with the complex dynamic of the double bind, she suggests that we permit prostitution until expanded economic opportunities make the prohibition unnecessary. Understood in this way, the prohibition is otiose. Either prostitution is of interest to disadvantaged women in poverty, and in that case the prohibition would cause them harm; or alternative economic opportunities have rendered prostitution unattractive, and then no prohibition of prostitution is needed. A second observation follows from the first. Anderson's account is not action-guiding. After it has been applied and a judgment rendered, what should be done, if anything, remains an open and urgent question.
Satz's account similarly fails to guide action. There is no straightforward means of determining how to respond to a specific market that scores highly on one or more parameters. Recognizing this, Satz explains that even ‘if a market interfered with or failed to promote certain values, banning it might be worse overall from the point of view of those same values’ (2010: 110). Consider again her treatment of markets in women's sexual labor. Satz is sensitive to the many ways the prohibition on prostitution harms women, rendering them vulnerable to the control of pimps and subject to disproportionate punitive measures. But this does not yet, on her view, justify intervention. How we ought to respond to particular objectionable markets cannot be determined a priori but requires careful consideration of empirical evidence and attention to the specific ways that the market functions to undermine Equal Status. ‘If we are troubled by prostitution’, Satz concludes, ‘then we should direct much of our energy to putting forward alternative models of egalitarian relations between men and women’ (2010: 153).
Unfortunately, even if true, this suggestion is hardly responsive to the central issue. What matters is whether we have reason to oppose markets in women's sexual labor. An answer to this question cannot be put off until we have solved the larger social and economic problems that make that market attractive in the first place. So we are left in much the same place where we started. A market has been identified as objectionable, but there remains the task of determining the appropriate response. Like Anderson, Satz faces the problem of action guidance.
3. Constitutive and Comparative Normative Significance
The problem of action guidance is, I suggest, traceable to a confusion between two ways of assessing specific markets. To support the conclusion that a market ought to be closed, only one form of assessment will do. Neither Anderson nor Satz supply it.
The commodification debate proceeds on the assumption that specific markets—collections of transactions individuated by what is exchanged—are morally distinctive. Different markets may provoke grave moral concern or no concern at all. Markets in apples and automobiles are thought innocuous, while those trading in human body parts and services are particularly troubling. Operating on this assumption, theorists focus their attention accordingly. Concerns about the market in women's sexual labor, for example, are addressed by attending to that market. If one assesses each transaction and finds all objectionable, then so too is the market. If one assesses each transaction and finds all permissible, then so too is the market. Admittedly, if one begins with the idea that specific markets are morally distinctive, then this approach will appear eminently sensible.
Understood in this way, market assessment involves an assessment of a specific market's constitutive normative significance. Thus, for any specific market we may supply a
Constitutive assessment. A set of transactions T when assessed by an evaluative standard E has constitutive normative significance S.
A constitutive assessment reveals something about a specific market, but it does not reveal much. To arrive at such a judgment all that is required is knowledge of that market's transactions.
This approach to market assessment, I shall argue, is deficient. The moral significance of a specific market is not discernible through an examination of its transactions. We can, following Anderson and Satz, evaluate specific markets in this way. But the resulting constitutive assessments are less important than supposed and tell only part of the normative story. Moreover, as I shell explain, it is precisely because they proceed from such assessments that their accounts face the problem of action guidance.
While constitutive assessments pick out something of moral importance, for the purpose of debating commodification they are of limited use. Such assessments are neither necessary nor sufficient for justified intervention into the market. What follows from a specific market's operation is morally important, but so too is what would transpire in the market's absence. And no assessment of a market's transactions will reveal this crucial information. To see the limitations of such assessments, return to Anderson's treatment of markets in women's sexual labor. Let us stipulate that each transaction, from the perspective of Liberal Values, imposes terrific moral costs and confers no benefits. Yet, this constitutive assessment does not reveal whether the market should be prohibited. Conditions in the market's absence may be worse. Consider the following case:
Sex robots. The market in women's sexual labor is found to undermine the social conditions necessary for sex to be properly valued as a shared gift. You can close the market with the push of a button. Doing so will result in the rapid expansion of the market in sex robots, which will, to an even greater extent, undermine the social conditions necessary for sex to be properly valued as a shared gift.
We have stipulated that the market in women's sexual labor, given a constitutive assessment, violates Liberal Values. Yet, it would be a mistake for you to push the button. For the increased use of sex robots that would follow would—from the very same evaluative perspective—make things worse. If you are concerned about securing the social conditions necessary to value sex as a shared gift, you should be concerned about prostitution and sex robots. Constitutive assessments, however, do not afford this wider perspective. Of course, the point illustrated by the sex robots case could, with only cosmetic modification, be made with equal force against Satz. There are, sadly, many ways to undermine the equal status of women as a class.
The lesson is simple: One cannot justify the prohibition of a specific market on the basis of a constitutive assessment. To establish that a market should be prohibited requires a different kind of judgment. One must show that relative to one's evaluative standard things would be better in the market's absence. Instead of focusing narrowly on the transactions making up a specific market, this judgment arises from consideration of a market's overall normative contribution. What is required is a comparison of the outcomes of two possible courses of action, that in which the market is retained and the best alternative available without it. To assess a specific market in this way—to evaluate the moral difference it makes—is to assess its comparative normative significance. Thus, for any specific market we may supply a
Comparative Assessment. A set of transactions T when assessed by an evaluative standard E has comparative normative significance R as a function of the difference between the set of all market transactions including T and the best available alternative set of all market transactions without T.
A comparative assessment reveals a great deal about a specific market. It indicates whether the market's total contribution to the market system is positive. Unlike a constitutive assessment, which focuses on a specific market in isolation, a comparative assessment requires knowledge of the entire market system. This more expansive judgment is required if one is to justifiably oppose a specific market.
These two forms of assessment are related in important ways. The judgments they deliver do not express incompatible propositions. A constitutive assessment tells part of the normative story. A comparative assessment completes the story. The latter judgment subsumes the former. To see this, return to the sex robots case. Consider first performing a constitutive assessment. The set of transactions comprising the market in women's sexual labor T is judged by an evaluative standard. Perhaps this reveals that T undermines the social conditions necessary for sex to be properly valued as a shared gift—it violates Liberal Values. Or from Satz's perspective we may discover that T undermines citizens’ ability to interact as equals—it violates Equal Status. Now consider performing a comparative assessment. This will involve an evaluation of all market transactions including T. And, crucially, everything that was true of T when given a constitutive assessment will still be true of T when given a comparative assessment. The phenomenon is the same, as is the evaluative standard that is employed. Thus, I do not dispute that specific markets have whatever constitutive assessments Anderson or Satz claim for them. Rather, I dispute that such judgments can play the justificatory role in which they are cast.
By analogy, consider how weight and body mass index are indicative of health. Knowing only that one weighs a lot, we might suppose a diet is in order. But we cannot be sure. Perhaps this person is exceptionally tall. Less food might make things worse. Knowing one's body mass index, by contrast, is more informative. While it does not reveal one's actual weight, it does reveal whether the relationship between one's weight and height is healthy. In acting on the assessment supplied by one's body mass index, we do not deny the significance of one's weight as a determinant of health. Weight matters, and its importance is contained within the measure of one's body mass index. Similarly, knowing that a market has an unfavorable constitutive assessment is insufficient to justify its prohibition. Closing it may make things worse. Knowing its comparative assessment, by contrast, is more informative. While it does not reveal the market's constitutive assessment, it does reveal whether closing that market would make a positive normative contribution. In acting on this assessment we do not deny the market's constitutive normative significance. That matters, and its importance is reflected in the complete picture delivered by its comparative assessment.
With this distinction in place we have a plausible diagnosis of the problem of action guidance. The accounts offered by Anderson and Satz proceed from constitutive assessments. Such judgments tell only part of the normative story. One may, on the basis of such a judgment, conclude that some specific market is objectionable. But in comparing conditions with the market operating to those arising in the market's absence, we glimpse the market's comparative normative significance. This reveals more of the moral story. And this part of the story may be rather different from what was suggested by the constitutive assessment. And it may support a rather different conclusion. It is then perfectly unsurprising that Anderson and Satz do not insist upon their accounts’ initial pronouncements. What motivates both to hedge is the realization that acting on a constitutive assessment may be counterproductive. Rather than violate Not-worse, both refrain from making any recommendations.
If this much is right, then neither Anderson nor Satz has offered a defensible case for limits on the market. They focus unduly on the transactions making up specific markets. Both arrive at a judgment of a market's moral status through a constitutive assessment. But specific markets have normative significance that outstrips that of their transactions. Part of what matters when assessing a market is what would happen if that market were closed. It matters if prostitutes will be replaced by sex robots. To show that there are some things that ought not to be for sale, as Anderson and Satz contend, one must supply a comparative assessment.
My argument to this point has focused on the reasoning offered to support the outright prohibition of certain specific markets. This emphasis is fitting as Anderson and Satz, and anticommodification theorists generally, defend the philosophically interesting claim that some things ought not to be for sale rather than the trivial claim that some things ought not to be for sale except in regulated markets. Nonetheless, one may regard market regulation as a natural response to the concerns that animate opposition to commodification. And some anticommodification theorists may be tempted to think that constitutive assessments, even if imperfect, are instrumental for identifying problematic markets and finding suitable regulatory solutions. This temptation may be especially strong for those who are moved by the assumption on which much of this debate proceeds, namely, that specific markets are morally distinctive. The proposal is that even if we do not conclude that the market in women's sexual labor should be closed, we may by way of a constitutive assessment discover what regulations are appropriate. The aim of market assessment, on this suggestion, is to ensure that each specific market has a permissible constitutive assessment. Whatever its appeal, this suggestion is mistaken. As was the case for prohibition, regulation cannot be justified on the basis of constitutive assessments. To explain why this is so, I will turn to those on the other side of the debate who defend markets without limits.
4. For Markets Without Limits
Brennan and Jaworski defend markets without limits. Their view is captured by the slogan: ‘If you may do it for free, then you may do it for money’ (2016: 10). This proposal has some intuitive appeal. There is nothing about any specific goods that makes them inherently unfit for exchange. Philosophical questions about the market's moral limits are reduced to questions about what is permissibly possessed. Because women's sexual labor may be permissibly given—because gift transactions are permissible—it may also be sold. Brennan and Jaworski reason from the permissibility of the exchange qua gift to the permissibility of the exchange qua transaction. The market is inert. It ‘does not introduce wrongness where there was not any already’ (2016: 10).
While their position may appear distinctively libertarian, the normative assumptions accepted by Brennan and Jaworski are not. Rather than invoking contentious claims about individual choice, they aim to defend markets without limits by appeal to values all can affirm (2016: 22–23). This is reflected in their evaluative standard:
Common Sense. Market activity must not violate commonsense moral principles.
The content of these commonsense moral principles is unspecified. This is consistent with their concessive approach. Brennan and Jaworski seek to vindicate markets without limits without reliance on any controversial moral or political commitments. They intend to grant the normative assumptions accepted by anticommodification theorists while still undermining their anticommodification conclusions.
To be sure, their position is more nuanced than their slogan. Brennan and Jaworski deny that there are inherent limits on the market. This is compatible with affirming other limits. In some cases one should not buy something, not because the nature of the good is incompatible with market allocation, but because of some incidental moral constraint. For example, if you promise not to sell something, then you do something wrong if you sell it. You break a promise. We do not need to appeal to the market to explain why that is wrong. Or if someone is dangerously drunk, then you do something wrong if you buy her a drink. You put someone in danger. We do not need to appeal to the market to explain why that is wrong. In such cases the market makes no interesting contribution. What is objectionable is the violation of a prior moral commitment. These are limits on the market, but they are merely incidental.
Brennan and Jaworski also recognize a second kind of limit on the market. Some things ought not to be for sale because some things ought not to be possessed. As they put it:
Principle of Wrongful Possession: If it is inherently morally wrong for someone to possess (do, use) X, then (normally) it is morally wrong for that person to buy or sell X. (2016: 11)
Child pornography and nuclear weapons, they explain, should not be sold because such things should not be possessed. The suggestion seems to be that the possession of certain objects is inherently wrong because their production or use essentially involves a rights violation or intolerable harm. If mere possession of such things is wrong it follows as ‘a trivial consequence’ that their sale is wrong (2016: 11). As before, these are limits on the market, but uninteresting ones. No appeal to the market is needed to explain what is wrong with the sale of such things. All of the normative work here is done by the wrongness of possession.
Brennan and Jaworski's defense of commodification is in one other way crucially qualified. They defend markets without limits, but not free and unregulated markets without limits. Their position is compatible with the need for significant regulation. Indeed, they suggest that much opposition to commodification stems from the deplorable conditions from which some transactions arise. Still, they caution that we must not confuse what is sold with how it is sold. We cannot conclude that a specific market is objectionable because some instantiation of it is objectionable. To show that some good ought not to be for sale—to establish an inherent limit on the market—it must be shown that its commodification would inevitably violate their evaluative standard, Common Sense. It must be shown that there are no conditions under which that good could be permissibly exchanged.
Having clarified the content of their thesis, Brennan and Jaworski announce their approach to defending it: ‘Our strategy . . . [is to] articulate, explain, and then debunk the various arguments anti-commodification theorists have produced to try to show that commodification is wrong’ (2016: 22). The strategy reflects the dialectic. Anticommodification theorists argue, by way of constitutive assessment, that some specific market is objectionable. Brennan and Jaworski respond, also by way of constitutive assessment, by showing that a market in that good—suitably regulated—may be made to operate consistent with their evaluative standard. Thus, they defend markets without limits by defending each market. Although this approach is piecemeal, they reason that ‘if we can repeatedly show that the critics’ complaints are unfounded, this builds a case for our thesis’ (2016: 22).
With these essentials in place, we should appreciate how modest Brennan and Jaworski's view actually is. First, their account states only a sufficient condition for doing something for money, namely, that one may do it for free. Second, while they defend markets without limits, they need not defend any existing markets as they operate. They maintain only this: for anything permissibly possessed, a suitably regulated market in that good may operate consistent with Common Sense.
5. The Same Mistake
Brennan and Jaworski adopt their own evaluative standard and defend commodification rather than oppose it, but they take the same mistaken approach to market assessment as Anderson and Satz. They seek to justify their support for specific markets by showing that, suitably regulated, such markets may enjoy a permissible constitutive assessment. This proposal has much in common with the suggestion proposed earlier—that constitutive assessments are instrumental in determining how best to regulate problematic markets. Here I show that neither proposal should be accepted. Both cast constitutive assessments in a central role. Such judgments, however, are neither necessary nor sufficient to justify support for or opposition to any specific market. Nor do they offer adequate guidance for market regulation. Accounts that take this approach to market assessment will either violate Not-worse or fail to guide action.
Recall that Brennan and Jaworski accept limits on the market that follow from the principle of wrongful possession. They hold that some things, such as child pornography and nuclear weapons, should not be sold because they should not be possessed. This principle is portrayed as philosophically innocuous. ‘“As far as we can tell’, they explain, ‘everyone in this debate agrees to the Principle of Wrongful Possession’ (2016: 11). And given certain assumptions, it is not hard to see why. If we assume that constitutive assessments are what matter in market assessment, then the principle would indeed be appealing. Markets in goods violating the principle, when given a constitutive assessment, will necessarily violate Common Sense. From this perspective, the principle appears irresistible.
Yet, the principle of wrongful possession would amount to an unjustifiable limit on the market. Brennan and Jaworski, in accepting the principle, would exclude as impermissible specific markets that would make a positive normative contribution. Their account, accordingly, violates Not-worse.
An example can clarify the point. Let us assume that a market in child pornography, when given a constitutive assessment, violates Common Sense. Suppose we know everything there is to know about the transactions in that market, and they are all absolutely bad. According to the principle of wrongful possession such a market is unacceptable. But this is too hasty. Consider
Hell. Sexual abuse of children is at level 10. A market in child pornography emerges and sexual abuse of children decreases to level 2. You can close the market with the push of a button. Doing so will return sexual abuse of children to level 10.
Adherence to the principle of wrongful possession requires that you push the button. But that is clearly a mistake. That specific market would make a positive normative contribution to those horrific conditions. You ought not to push the button. In arriving at this conclusion, we do not deny that a market in child pornography is truly terrible. Its constitutive assessment quite correctly reveals it to be profoundly objectionable. Rather, we recognize that some things are even worse.
The foregoing suggests that because Brennan and Jaworski accept the principle of wrongful possession they accept an unjustifiable limit on the market. But perhaps there is an easy fix. If the problem is traceable to the principle, then they should jettison the principle. They do, after all, explicitly set aside the question of whether some things that should not be possessed might still be permissibly bought and sold. They write that ‘it's an open, empirical question whether commodifying those goods and services might improve upon the status quo. . . . Examining just when this is so goes beyond the scope of our book’ (2016: 18).
There is no easy fix. The problem is not that they accept the principle of wrongful possession. The problem is that their account proceeds on constitutive assessments, of which that principle is but one instantiation. Depending on background conditions, a specific market may warrant a miserable constitutive assessment yet still make a positive normative contribution. In the Hell scenario the market was prohibited by the principle, but we can imagine similar results in other cases. Consider
Short-range weapons. Gunfights are common with harm to bystanders at level 10. A market in short-range weapons emerges and harm to bystanders decreases to level 2. You can close the market with the push of a button. Doing so will return harm to bystanders to level 10.
Let us stipulate that although short-range weapons are not wrongful to possess, the market in which they are exchanged, given a constitutive assessment, violates Common Sense. Despite this, the market's contribution is welcome. It has a favorable comparative assessment. You ought not to push the button.
Perhaps the short-range weapons scenario is not so unacceptable. Strictly speaking, Brennan and Jaworski offer only a sufficient condition for permissibly doing something for money, namely, that you can do it for free. Thus, that a market does not meet this condition, yet is judged permissible, is not immediately a challenge to the account.
No matter, even on this modest interpretation their view is problematic. The Hell scenario and, less spectacularly, the short-range weapons scenario both illustrate how a specific market may have a negative constitutive assessment but still make a positive normative contribution. Yet, a specific market's assessment may diverge in the other direction as well. A market may enjoy whatever constitutive assessment is necessary to qualify as permissible—to operate consistent with Common Sense—but still make a negative normative contribution. Consider
Bad apples. Everyone is happy, each at well-being level 10. A market in apples emerges and everyone's well-being decreases to level 2. You can close the market with the push of a button. Doing so will return everyone to well-being level 10.
Let us stipulate that the apple market, given a constitutive assessment, is consistent with Common Sense. It meets the sufficiency condition defended by Brennan and Jaworski. Were their account correct, the market in the Bad apples scenario would be permissible; it should remain open. But plainly that is not the case. That specific market's normative contribution to the market system is unwelcome. You ought to push the button.
If this much is right, then Brennan and Jaworski's account is mistaken. Like the anticommodification theorists they critique, they assume that specific markets are morally important only insofar as their transactions are. Their account proceeds exclusively on constitutive assessments. Such judgments, however, are simply not suitable for market assessment. We do not establish that a market is permissible by establishing that it has a permissible constitutive assessment. Were we to act on their account's prescriptions we may make things worse.
We can also for similar reasons reject the earlier suggestion that constitutive assessments play an instrumental role in determining how to regulate specific markets. Such assessments are inadequate. Suppose you could, with the push of a button, ensure through regulation that a specific market, given a constitutive assessment, operates in a way consistent with one's evaluative standard. Should you push? We cannot say. Even if regulation would result in a local improvement—the transactions making up the specific market, considered in isolation, would be rendered morally impeccable—it remains an open question whether or not this would make things worse overall. Perhaps the market in question features in the larger market system as the apple market does in the bad apple scenario. It may warrant a favorable constitutive assessment but still be morally objectionable.
6. The Practical Significance of the Mistake
I have argued that specific markets have both constitutive and comparative normative significance, and I have claimed that confusion between the two forms of assessment is pervasive in the commodification debate. Much of the discussion, however, including that of the improbable scenarios, has focused on abstract possibilities. Therefore, it may not be obvious that the mistake matters much in practice. To make this obvious, I turn to an instructive example.
The moral status of kidney markets is a contested matter. The variety of objections to sales is impressive, but I will focus on the perennial challenge that vendors would be harmed. This simplification delivers the following evaluative standard:
Welfare. Market activity must not be exceedingly harmful.
This evaluative standard is not intended to exhaust what is morally important about markets in general or kidney markets in particular. It is intended to capture only one consideration.
Many argue that concern for vendors’ welfare speaks against kidney sales. Consider an oft-cited study of the kidney market in Chennai, India (Goyal et al. Reference Goyal, Mehta, Schneiderman and Sehgal2002). Researchers found that most vendors regret their choice, suffer negative health and employment consequences, and do not improve their economic conditions. Having cataloged these harms, the authors conclude, ‘The sale of kidneys by poor people in India does not lead to a tangible benefit for the seller’ (Goyal et al. Reference Goyal, Mehta, Schneiderman and Sehgal2002: 1592). The reasoning on display is straightforward. If, as the evidence suggests, the harms of vending will likely exceed its benefits, then vending will likely harm vendors. Thus, concern for the welfare of those who would sell supplies reason to oppose the market. This inference is, to many, irresistible.
This reasoning is mistaken, and policy predicated on it may be counterproductive. To see this, let us stipulate, for the sake of argument, that the empirical evidence establishes a constitutive assessment of the kidney market, relative to the evaluative standard Welfare, namely,
Bad on balance. The welfare costs of vending exceed the benefits conferred by compensation.
This is the judgment on which much opposition to kidney markets is predicated. Yet, the inference— from the judgment of Bad on balance to the claim that concern for vendors’ welfare counts in favor of prohibiting kidney sales —is illicit. Such an assessment focuses narrowly on vending and thus is insensitive to the quality of options from which people choose. Thus, even if vending is bad on balance, it may remain the best option for some.
To hold that consideration of vendors’ welfare counts against permitting kidney sales requires a comparison of two possible outcomes. It must be shown that the kidney market has a certain comparative assessment, relative to the evaluative standard Welfare, namely,
Non-optimific. Vending imposes greater welfare costs than those that would arise when would-be vendors take what they judge to be their next best options.
A judgment of Non-optimific establishes that vending is bad for potential vendors given the options they face. Here, the benefit to the vendor is a function of the difference between the expected welfare after vending and the expected welfare after taking their nonvending option. With this judgment secure, removing the option to vend would not risk making things worse.
Crucially, one should not assume that the two forms of assessment will often agree. A judgment of Bad on balance focuses narrowly on the consequences of vending. A judgment of Non-optimific, by contrast, takes account of much more. Depending on the quality of options from which one chooses, the two assessments may diverge significantly. And the divergence is likely to be greatest where market participants are most desperate. The worse one's options are, the more likely it is that the best one is, in absolute terms, bad. It is thus a particularly cruel consequence of adopting this mistaken approach to market assessment that the greatest burdens are imposed on those most disadvantaged. Those who suffer the most from a misguided prohibition are the most vulnerable and desperate among us.
While no evidence reveals what prospective vendors would do were the option foreclosed, we have some notion of the desperation that makes the option appealing in the first place. Tong and colleagues, synthesizing available evidence, found that ‘Selling a kidney was perceived as the only means for survival, to repay debts they owed, or to assist a family member in financial need’ (Tong et al. Reference Tong, Chapman, Wong, Cross, Batabyal and Craig2012: 1142). Given the strength of these motives, what are the likely consequences of prohibiting sales? We can only speculate. Perhaps desperation makes prostitution more appealing. Perhaps it leads one to see one's children as economic resources or to think criminal activity a worthy choice. Perhaps one is drawn to dangerous labor. These options may be even more damaging to welfare than vending. And perhaps, as Goyal and colleagues found to be the case for vendors in Chennai, those who take these options do not escape from debt, later regret their choices and would not recommend them to others. The sadly realistic suggestion on offer is that even if the judgment of Bad on balance applies to vending, in some cases, vending may be the best option. To assume otherwise would be dangerously naïve.
7. Conclusion
I have argued that the commodification debate proceeds on a mistake. Participants have sought to justify opposition to or support for specific markets on the basis of an assessment of the transactions these markets comprise. Implicit in this approach is the assumption that specific markets are fully assessable independent of the rest of the market system. This assumption is false. What matters morally about a market is not limited to what one may discover upon inspecting its transactions. It also matters what would follow from its regulation or prohibition. This essential information is only available when one takes the more expansive view afforded by a comparative assessment. It is then unsurprising that accounts predicated on constitutive assessments either fail to guide action or violate Not-worse. Such judgments tell only a small part of the normative story.
Correcting the mistake that I have identified requires a dramatic rethinking of our approach to market assessment. In closing I will note just two of the more important changes. First, we must reconsider the significance attributed to our intuitive reactions to certain markets. It is perfectly natural that exchanges involving body parts and bodily services provoke greater moral concern than trades in apples and automobiles. But these intuitive reactions belie the complexity of these phenomena. Markets interact with other markets and other features of the environment in surprising and unpredictable ways. The moral valence of a market can flip when new markets emerge or close or when behaviors or circumstances change. As a result, our intuitions about specific markets are less instructive than is often supposed and are prone to lead us astray. Second, we should not aspire to regulate specific markets such that, assessed in isolation, they conform to an evaluative standard. We do not alleviate moral concerns about the market by ensuring that each specific market enjoys a favorable constitutive assessment. It is quite possible that the optimal market system will contain specific markets that are, in certain respects, objectionable. An exchange may, on balance, be bad—it may impose great moral costs and confer no benefits—yet blocking that exchange may be even worse.
Where does this leave the debate? We can still ask sensible questions about the moral limits of the market, but we should not expect familiar answers.
