Markets of all kinds have proven to be of tremendous benefit to society, so much so that there is a debate within business ethics about whether there are any moral limits to markets. Given the benefits, and thinking just of those markets constituted by a stream of voluntary transactions between mutually consenting parties, it can be hard to even imagine how a market would in any sense be immoral or impermissible. At the same time, though, most of us find the idea of markets in kidneys, or sex, or nuclear weapons somehow repugnant. Markets in these and certain other goods strike many as corrupting, exploitative, or to express a form of disrespect. So, the questions at the heart of the so-called ‘limits of markets’ debate in business ethics is, first, whether there are any limits on what can permissibly be bought and sold, second, precisely what those limits are, and third, what considerations lead us to draw the line where we do.

Many within business ethics (and society at large) argue that such a line can be drawn; there are ethical limits to markets. Moreover, this often involves goods or services where it is okay to have or do these things for free; there is just something bad about selling them. So, for example, while all of us agree that consenting adults can permissibly engage privately in sex acts, many of us feel that adults cannot permissibly sell sex.

In Markets without Limits, however, Jason Brennan and Peter Jaworski argue to the contrary. They argue that, for any good or service, as long as it is permissible to give it for free, then it is permissible to give it for money. Answering the apparent objections to various unsavory markets at length, they suggest that as long as a successful market is possible (as long as it can be regulated to avoid exploitation or other bad elements), then we are permitted to pursue it. Brennan & Jaworski make a compelling case, and they have continued to do so in response to a number of recent objections. Here, however, I will provide a challenge to their lucrative thesis sometimes gestured towards but never quite fully articulated.

The intuition behind the challenge is that markets can be dangerous and hard to control. Even if Brennan & Jaworski are right in arguing that if something can be permissibly sold, that there is some possible way of arranging a market for it that would be acceptable, it is not clear that we are going to get that version of the market if we try. A morally appropriate market for some good may be possible, but not feasible. And this provides a substantive constraint on our activity in pursuing markets.

To this end, in the first section I lay out more explicitly Brennan & Jaworski’s controversial view and its motivation. Although the view suggests a presumption in favor of pursuing new markets, in the second section I show how the possible danger of prospective markets together with the open question of the feasibility of moral markets undermine this presumption. I argue that feasibility is a limit on markets. Then, after carefully disambiguating the Markets without Limits thesis in the third section, I show how this limit of feasibility can be used in an argument against it. Having laid out this argument, I consider several replies. Answering them uncovers the specific danger of these markets and the relevance of feasibility, but it leaves open the question of what we should actually do when confronting the issue of pursuing an infeasible but possibly beneficial market. Thus, I conclude by considering a number of ways to proceed given infeasibility.

The paper is structured broadly as a challenge to the central claim in Brennan & Jaworski’s book. This seems appropriate insofar as that claim marks a significant view in the field, with several other full-length articles addressing it. However, the purview of this article is broader than a mere response to Brennan & Jaworski, or even as a new objection to a central view on this topic. This article will additionally serve as the beginning of a discussion of determining how the feasibility of various market conditions is relevant to when and how we should pursue market solutions. Feasibility as an object of study has only recently gained popularity within political philosophy. This article serves to introduce the work done on that topic to this debate and to business ethics generally. When thinking about the ethics of markets, we need to be thinking not only about the ethics of particular transactions or of various regulations, but about the risks involved in the systemic changes we might try to make.

Markets Without Limits

Much of the previous limits of markets debate involved arguments surrounding the permissibility of particular markets. For example, Holder (1985) argues against markets for surrogates in particular, as does Anderson (1990, 2000). Further authors such as Radin (1996), Held (2002), Satz (2010), and Sandel (2012) have argued against the commodification of sexual labor, child labor, kidneys, votes, the right to name babies, and any number of conceivably commodifiable individual goods. We are left with a broad array of concerns about how markets introduce the opportunity for certain wrongs (e.g., exploitation, corruption) or constitute various wrongs (e.g., that they are disrespectful).

Brennan & Jaworski (op. cit.) set themselves up to argue against these ‘anti-commodification theorists.’ Their book aims to address the concerns of their opponents as a part of the defense of their own thesis:

“Markets without Limits: If you may do it for free, then you may do it for money.” (10).

There are many things Brennan & Jaworski admit that it is impermissible to buy and sell. It is wrong to buy or sell child pornography or nuclear weapons or to hire assassins. But these are also things that it is wrong to have or do for free. One should not have child pornography or murder someone, and so it is not okay to pay others for these things. Brennan & Jaworski are careful to note that their opponents in this debate are not concerned about the sale of things everyone knows to be bad to have, but with the idea of markets for goods that otherwise it is not wrong to have or use.

Crucially, Brennan & Jaworski are also careful to avoid what they call ‘incidental’ limits on markets. There is nothing inherently wrong with selling baseball cards, but it would be wrong to do so if you happened to have promised your brother that you would not. Of course, what is wrong in this case is not engaging in a market for baseball cards per se, but breaking your promise. Similarly, there is nothing inherently wrong with selling your car, but it would be wrong to sell your car right now if your child needed to go to the hospital (14).

This caveat is critical for their argument. They go on to argue that when it is bad to engage in some market, what is bad is not merely having the market for that good. If engaging in the market is even bad to begin with (and its apparent badness is not simply a product of our inappropriately feeling disgust), then it is likely bad because of how we have chosen to set up the market in question, not simply that we have set up such a market.

For each objection to particular repugnant markets, Brennan & Jaworski show how the bad-making features of these markets is a matter of how opponents are imagining that the markets in question incidentally happen to be structured. As such, these features can be undercut by structuring the market in different ways meant to avoid the problems. Key to their optimism is an appreciation for the sheer volume of ways in which market interactions can occur. On this point, Brennan & Jaworski draw an interesting analogy:

Markets are a bit like guitar amplifiers. Just as guitar amps have various knobs (gain, volume, bass…etc.), so markets might have a range of variables that can be put to different settings. Just as some guitar amps sound good only on very specific settings, some markets might be good only on very specific settings. (2015a:39)

Markets can vary depending on the participants, the means of exchange, the price, the mode of exchange, and in many other ways, suggesting literal thousands of possible ways of structuring any given market. The implication is that this bounty of market structures increases the likelihood that there will be certain ways it that will be both permissible and beneficial to society. And they suggest that as long as one of those market structures avoids the problems suggested by anti-commodification theorists, then they have shown that it is permissible to make a market for this good—a market with the allowable structure. As they say,

It won’t be enough for them to show us that some markets are bad—morally impermissible—on some or even many “settings”. They need to show us that markets in certain commodifies are bad on all the possible settings. Otherwise, if we can find even one “setting” for a particular market, then our thesis stands. (40)

If they are right, the bar is very high for anti-commodification theorists. If there is just one right way of facilitating a market in kidneys, say, but our having a market in kidneys of that structure would both not wrong anybody and would save the lives of thousands of people annually, then we might well be obligated to make such a market. As Brennan & Jaworski note, that a market may only be permissible if conducted in very particular ways does not indicate that it would not be a powerfully good market to have (ibid.).

Putting all this together, it’s hard to be confident about the inherent wrongness of a market for any good. Sure, a market in kidneys might seem likely to lead to exploiting the worst-off among us, but this may just be incidental. If we are careful in thinking about who gets to participate in the market and how consent for participation will be demonstrated and so on, then our concerns may evaporate. Without considering all of the possible market structures (or at least a significant number of them), it would seem to be impossible for anti-commodification theorists to demonstrate that a market is inherently wrong. Given that Brennan & Jaworski themselves go through many of the seemingly most problematic markets to argue for ways that they could be structured to be morally acceptable, this provides solid inductive grounds for concluding that where it is permissible to have some good, there will be morally acceptable ways to sell it.

In this way, the Markets without Limits thesis appears to deliver on a central idea motivating Brennan & Jaworski, that there is nothing bad about markets as such. Selling something instead of just giving it away does not (in itself) change the moral landscape. As they say, “…the market does not transform what were permissible acts into impermissible acts. It does not introduce wrongness” (ibid., emphasis in original). This is an important point for them. At times it can feel that there is something nefarious, evil, or at least unsavory about markets such that we really should avoid them if we can. They are not things to be proud of; they are elements of inherent alienation. With that perspective in mind, Brennan & Jaworski provide a welcome and necessary point of view, reminding us about the power of our markets for tremendous good.

Two Disanalogies Towards a Limit

If the Markets without Limits thesis is vindicated, and with the benefits of markets foregrounded, we are left with what feels like a strong presumption in favor of pursuing markets. Even where they may feel repugnant, we can trust that there will be ways of structuring those markets that avoids moral hazards and secures some payoff. Further, if we grant that markets do not ‘introduce wrongness,’ this gives the impression that it is relatively costless (morally at least) to implement markets. So, why not go ahead and introduce them?

Despite this impression, I do not think such a presumption is justified. To see why, I want to recognize two disanalogies with the original analogy concerning guitar amps. Consider again that analogy: There are many settings for amps, and some amps only sound good on particular settings. And so it is with markets. There are many market structures, and certain markets may only be good with particular structures.

First, this analogy conceals just how dangerous markets can be. If you pick the wrong guitar amp setting, it’s not a big deal. You may experience some discomfort when you strum those first notes, but you can trust that your mistake will not lead to something as grave as children being exploited for sex. When we are thinking about these markets, however, this is the kind of serious issue that we are concerned about – that legalizing prostitution may lead to child sex trafficking; that legalizing kidney sales may lead to coerced organ harvesting. Perhaps the dangers associated with every repugnant market are not so grandiose, but the general idea is that pursuing a market in a good can be pretty dangerous. We are risking exploitation, unjust pricing, unsafe work environments, employees drawn into work meaningless work, customers drawn to buy worthless products. In other words, we risk serious harms or rights violations, and these are elements that must be considered. Brennan & Jaworski’s analogy may thus be fine-tuned to how amorphous and adaptable markets can be, but it is tone deaf to the dangers involved in pursuing them.

One feature to focus on here concerning the amount of danger involved is the size and speed of markets. Brennan & Jaworski may be right in the central thought that markets do not transform the nature of transactions from morally acceptable to unacceptable, from right to wrong, from good to bad. Markets are not inherently bad things, and so moving from a free exchange to a market exchange may not be the kind of change that introduces wrongness or evil.Footnote 1 But the introduction of a market does have a particular effect. It increases the velocity of transactions, ensuring wider access to the good and more participants able to transact. This is part of why markets can be wonderful. However, it also makes markets liable to amplify whatever harms or risks come through voluntary transactions insofar as the rate of transactions increases. Moreover, it will have knock-on effects for other parts of the market, perhaps cannibalizing some of the market while facilitating others.

So far, this is all just to say that the costs of failing to implement a moral version of a market can be high. This shows the significance of ensuring that the market will have an acceptable market structure if there is to be a market in the good. Again, though, we might have thought this was easy to secure. Brennan & Jaworski’s argument provides for the conclusion that there are possible good market structures, but it would not be too much of an extension for commodification theorists to think that there are likely many such market structures that would neutralize the moral issues for a market in a certain good. Even granting that it would be morally disastrous to have a market in one of these goods not in one of the permissible structures, the natural thought might have been that which market structure we implement is up to us; it is no more challenging to implement the permissible market structures than any other.

Here, there appears to be a second critical disanalogy with guitar amps. If we are choosing between settings on a guitar amp, not only is the cost low of failing to choose the right setting, but it is equally easy to pick one setting or another, and it takes minimal effort to change settings if a mistake is made. The way that market ‘settings’ are presented by Brennan & Jaworski, it may appear that the same is true of them. The market for a good can be all in cash or done through credit, or it can be done through a set price or barter. It appears that the choice is ours. But this appearance is misleading. It may be much easier for us to make a market that is all done through cash (since we will not have to work with credit card companies). Or it may be harder to make a market that is entirely done through bartering in a western economy.

Ways of making and maintaining markets generally can be easier or harder. And making a market in some particular way can be quite difficult, as can remaking a market once constituted in a specific way. Since markets consist in many transactions rather than one, controlling a market as a collective endeavor is that much harder. For us to make a market even for guitar amps regulated in some particular way will be much more challenging than my action of changing the setting on my amp.

To avoid looming harms, markets may have to be very carefully constructed. And it’s not just that they have to be construed precisely. How they must be constructed may be fairly challenging to achieve. A market for sex, for instance, would presumably require a significant amount of regulation and enforcement. We would need just the right set of rules to avoid horrible human rights abuses. This would likely require new regulatory bodies or watchdog organizations poised to create incentive structures and enforcement mechanisms to ensure compliance. We would need to create paths for whistleblowers within the industry and to ensure that industry participants could not unduly influence their regulators. We would even need extralegal, cultural mores around virtues and vices in this market, taboos and best practices. The list goes on.

Given just how much is likely required to avoid the moral hazards of certain markets, we cannot take it for granted that we will end up with a moral version of those markets if we try to implement them. There may be clear barriers to the proper implementation of these moral market structures along many dimensions, and the looming immoral versions of the market may be catastrophic. Thus, it is a live question whether the moral versions of these markets are achievable.

This sense of the importance of achievability suggests a natural limit on the permissibility of markets. What I want to offer is that it is plausibly captured in terms of the recently much-discussed concept of feasibility. Where the question is whether it is feasible to construct a market for a good in a market structure that avoids wrongdoing, we can see the following limit: If a morally acceptable version of a market is infeasible, then we are not permitted to construct it.Footnote 2

Scholars have long been sensitive to the issue of what is practically achievable or feasible,Footnote 3 but political philosophers and theorists have recently taken more care to explicate this generally intuitive concept of feasibility (Gilabert & Lawford-Smith, 2012; Lawford-Smith, 2010, 2013; Southwood, 2018). Gilabert & Lawford-Smith (op. cit.:809) give a first-pass characterization of the notion when they say, “Roughly, a state of affairs is feasible if it is one we could actually bring about.” Of course, this rough characterization could be taken any number of ways.

One way of thinking about feasibility and a feasible policy is to contrast it with this idea from above of a policy’s being merely possible in the broadest sense. A feasible policy, then, would be one which is possible in some more restricted sense. For instance, Wiens (2015a) suggests that feasibility is a matter of a state of affair’s being possible given our current stock of resources (institutionally, technologically, motivationally, etc.). Nailing down the exact delineation of these resources is a large challenge. However, once we think about not only what we have in society concretely but in terms of political capital and political will, it is fairly intuitive to judge when policies appear clearly beyond our current resources.

Another way of understanding feasibility would be to tie it closely with our understanding of ‘ability’ and to think of what is feasible roughly in terms of what we are collectively able to do. Thinking of it this way, we can recognize that there is a long literature within the philosophy of action that attempts to analyze having an ability conditionally (see Maier, 2018). As the simple conditional analysis goes: S has the ability to A iff S would A if S tried. So, as an approximation: a policy is feasible iff we would successfully implement it if we tried. Understanding feasibility in this way also has challenges.Footnote 4 However, it also promises a way of using information about our abilities to judge the feasibility of various policies.

It is not my aim here to settle on the right account of feasibility. Indeed, we may agree with Hamlin (2017) or Chahboun (2017) that there are distinct notions of feasibility worth distinguishing. And there are more questions about feasibility to answer besides. For instance, there is an ongoing discussion of whether feasibility as a concept is binary or scalar – whether something is feasible/infeasible full-stop or whether feasibility comes in degrees, or whether this captures two senses of feasibility (as Gilabert & Lawford-Smith [2012] suggest).Footnote 5

Regardless of the exact account we take on, it seems apparent that feasibility will be in some way normatively significant.Footnote 6 In particular, feasibility is often understood to be a constraint on justice, the idea being that justice cannot demand what we cannot deliver. Scholars have debated the merits of this kind of idea, but we can see how it naturally extends to provide the constraint on markets suggested above. Unless a morally acceptable version of a market is feasible, we are not permitted to construct it.Footnote 7

Interpreting the Thesis Towards an Objection

What exactly does the foregoing mean for the Markets without Limits thesis? Insofar as feasibility is a limit on markets, it clearly indicates that Markets without Limits is false. But how exactly? The thesis may be called ‘Markets without Limits,’ but it does not actually state that there are no limits on markets. Thus, our question is how the considerations raised provide an objection to the thesis, or whether the arguments given by Brennan & Jaworski secure the Markets without Limits thesis after all. How we answer this depends on how we interpret the thesis itself, and the right way to do so requires engaging with a bit more work in linguistics and philosophy of language than one might expect.

Take another look at the thesis. It seems to assert something straightforward: It’s permissible to sell goods that it’s permissible to have. But, what, like right now? Am I permitted to sell my kidney today, for instance, since it is permissible for me to donate my kidney? That does not seem right; it is illegal for me to sell my kidney. Assuming an obligation to obey the law, I am not permitted to sell my kidney.Footnote 8

This confusion is answered by recognizing that within the thesis lies a generic. Linguistically, generics “…express generalizations, but unlike quantified statements, generics do not carry information about how many members of the kind of category have the property” (Leslie & Lerner, 2016). Crucially, propositions expressing generics may tell us something that is only sometimes and not always the case. For example, it’s true “Birds fly” and “Chickens lay eggs,” even though not all birds fly (e.g., penguins), and only female chickens lay eggs.Footnote 9

When we say of some activity that it is permissible, what is often left sotto voce is an acknowledgement that the same behavior could technically be done in ways that are impermissible. Given that I have a license, it is true to say that I am permitted to drive. But, of course, I am not permitted to drive illegally. The permission to drive admits of exceptions. Applying this to the thesis, we can see that although the thesis tells us that it is permissible to sell the goods that it is permissible to have, the thesis does not commit us to the claim that, for some good that it is permissible to have, that it is always permissible to sell it.

So far, this is all to the good for Brennan & Jaworski. Their discussion of the incidental features that lead markets to go wrong shows that they do not want to be committed to the claim that if you may do it for free, then you always may do it money. They can accept something weaker, that these markets are only permissible a certain proportion of the time. The problem comes from determining what proportion of the time these markets would need to be permissibly pursued to vindicate the thesis as they really intend it.

As we saw, Brennan & Jaworski suggest that the thesis is vindicated if there is a single permissible version of the market. On one reading, that’s right. If you say, “Birds lay eggs,” and I respond, “Mammals lay eggs too,” many of us will accept my response as true, though this is the exception to the norm for mammalian reproduction. This is the so-called ‘existential generic,’ as discussed in Cohen (2004). Similarly, if someone says, “You may have this good,” we may evaluate the response, “You may sell the good too,” as true given what Brennan & Jaworski show, that there is a version of that market that would be permissible. Insofar as part of their aim is to show that markets are not evil, the existential reading of the generic may be sufficient. (Look, there is a way to have a market in kidneys that does not infringe rights! So, just having of a market is not problematic.) Nevertheless, I suspect this is not quite what Brennan & Jaworski (or at least commodification theorists broadly) are afterFootnote 10.

For one thing, anti-commodification theorists may not deny the thesis so understood.Footnote 11 According to Koplin (2018), such theorists never meant to argue for the strong view that certain goods should not be sold no matter the market conditions. Koplin shows how a number of anti-commodification theorists (including Michael Sandel, Deborah Satz, Elizabeth Anderson, and several others) have endorsed (or not condemned) the buying/selling of things like sex and surrogacy in idealized circumstances (437).

More importantly, this reading is insufficient to secure how Brennan & Jaworski seem to want to use the thesis. One gets the sense in reading that they want to apply the thesis to purportedly repugnant markets, galvanizing us to pursue and implement these markets in manners that are morally acceptable. That is, they want the presumption discussed in the last section—a presumption in favor of setting up markets. And if this is not precisely what they want, then it is at least something a reasonable commodification theorist would want, as it can be convincing.Footnote 12

For this reaction to be appropriate, however, it needs to be more than merely possible that markets in these goods can be structured in moral ways. These possibilities need to be salient possibilities, sufficiently relevant to our situation. As a reinterpretation of the thesis, then, I suggest reading it as asserting that, ‘for every good that it is permissible to have, it is permissible to engage in the market as it would typically be structured were we to construct the market while intentionally trying to avoid moral hazards.’

We need not argue over the details of this reinterpretation, as doing so involves engaging even more with the literature on the semantics of generics. In several ways, this interpretation is favorable for the commodification theorist.Footnote 13 Yet even the most charitable rendering of the thesis requires that there is at least one feasible moral version of the market for each good that it is permissible to have. And this just is the feasibility constraint from the last section. If a market is to be permissible on a charitable reading, then there must be a feasible way of constructing it morally. Implied is that if there is no feasible way to construct the market morally, then pursuing it is not permissible.

A commodification theorist needs it to be that there are feasible moral market structures, as it would otherwise be inappropriate to gesture towards these possibilities to suggest pursuing them. On this point, several recent authors have suggested that Brennan & Jaworski have not shown this by arguing for the mere possibility of morally acceptable market structures. For example, Koplin (op. cit.) ultimately maintains the Markets without Limits thesis is true but practically irrelevant.Footnote 14 Furthering this kind of skepticism, David Dick writes,

The message the market sends might cause a grave harm, the violation of an important right, or another serious moral wrong. Contingent or not, it might be beyond our power to successfully revise both the message this market sends and the connection it has with a serious nonsemiotic moral wrong. In a case like this, it hardly helps to notice that these connections are merely contingent and could conceivably be different in another possible world. Bullet wounds are only contingently lethal, but pointing this out does little to help those about to be shot. (2018: 235)

I think that this is exactly right. Dick is concerned that achieving the moral versions of these markets may be ‘beyond our power.’ Where feasibility is interpreted in terms of what we are collectively able to do, another way of expressing this concern is to say that these markets may be infeasible.

This alone suggests that the Markets without Limits thesis has yet to be sufficiently argued for. The commodification theorist who wants to use the thesis to get us to go out and make markets has a high argumentative burden. They need to show generally that if you may have some good, then there is some morally acceptable version of a market for that good that is feasible. Contrarily, opponents only need good reason to think that there is one instance in which a good is had permissibly but where a morally acceptable version of a market for it is infeasible. Taken together, this gives us the means of the following argument against the Markets without Limits thesis:

  1. 1.

    If a morally acceptable version of a market is infeasible, then we are not permitted to construct it.

  2. 2.

    There is at least one good that it is permissible for you to have and for which it is infeasible to implement a morally acceptable version of a market for that good.

  3. 3.

    Therefore, there is at least one good that it is permissible for you to have but that you are not permitted to buy or sell. (The falsity of Markets without Limits.)

In the next subsections, I want to consider two natural replies to the argument. The first reply challenges the first premise. That premise offers feasibility as a limit on markets, and it is in some way connected to the idea of feasibility as a constraint on justice, but we may deny that it is such a constraint or that this actually does motivate the premise. These concerns provide an opportunity to examine the reasoning behind the feasibility objection.

The second reply denies the second premise; it maintains that moral markets in typical cases are feasible after all. I flesh out a few ways of giving this reply, raising challenges for each. Ultimately, my aim is not quite to defend the second premise and so the argument. Instead, what I will show is that the burden is on commodification theorists to demonstrate that it does not go through, and discharging this burden is challengingFootnote 15. Given the stakes of implementing immoral markets, I think this provides for a presumption against introducing markets until moral versions of them have been shown to be feasible.

First Reply: Challenging Feasibility as a Constraint

I have suggested that that the concept of feasibility can be helpful in providing a constraint on moral markets. Though some market structure for an otherwise repugnant good might be extremely beneficial to society, we should not pursue it if it is infeasible. And we saw above how this was in some way related to the idea that feasibility is a constraint on justice. If a policy is infeasible, the thought goes, then justice cannot demand that we adopt it. This is the sense in which feasibility is taken to offer a constraint on justice: it is a necessary condition on justice that just policies be ones that it is feasible for us to adopt.

However, as a part of a much larger conversation over the role of ideal or non-ideal theory within political philosophy,Footnote 16 some scholars deny that feasibility can play this role of offering a constraint on justice. So-called ‘ideal theorists’ maintain that the demands of justice transcend questions of feasibility (Estlund, 2011).Footnote 17 While feasibility may be something to consider, the infeasibility of a market need not be an obstacle to its permissibility. On the contrary, we might be obligated to pursue such a market insofar as it is morally superior and possible. We can trust that there is a system of regulations that can be discerned, legislated, and enforced that would constitute this moral market. There may be challenges to appropriately motivating us to enforce the right regulations, say, but these are challenges to be overcome. They do not justify abandoning the possibility of life-saving markets in kidneys, for instance.Footnote 18

This objection may be appropriate, but in the end it is not very helpful or illuminating. Many among us do think that feasibility places a constraint on justice, it would be a pity if one had to be an ideal theorist in order to accept Markets without Limits. (Exegetically, I do not think it is fair to assume that Jason Brennan or Peter Jaworski themselves would advocate for ideal theory as a means of supporting their thesis.) Besides, even if one were a proponent of ideal theory, and infeasibility could not be used to argue against the permissibility of certain markets, it may still be normatively significant if a market is infeasible. We may be permitted to pursue them, but still gain a significant reason not to.

Another interesting, though less direct, reply is to put pressure on the exact connection between the impermissibility of infeasible markets and the claim that feasibility is a constraint on justice. One natural thought is that our first premise is an instance of or derivable from the claim that feasibility is a constraint on justice; however, the connection between them cannot be entailment. The strongest version of the feasibility constraint has it that ‘ought’ implies ‘feasible;’ feasibility is a necessary condition for it being such that you ought to do something (Chahboun op. cit.) on at least some understanding of ‘ought’ (Southwood, 2016). But this is not to say that if something is infeasible, then you ought not do it. Learning that something is infeasible only tells us that there is no obligation to do that thing.Footnote 19 It entails the absence of an obligation, not the presence of an obligation against performing an action. But if infeasibility at best implies the absence of an obligation, then it doesn’t on its own entail that you are not permitted from engaging in certain conduct, as the first premise claims.

It is easy to see why this would be. It may be infeasible that the middle school dance committee will arrive at a single theme for this year’s Homecoming Dance, but the committee is surely permitted to try to come to a consensus on the matter, and not obligated to avoid pursuing a consensus just because it’s infeasible. That may make it inadvisable to pursue, but not unethical. Thus, we may worry that the infeasibility of our achieving the moral versions of repugnant markets does not make it impermissible to pursue these markets.

In response, I can agree that infeasibility alone does not entail impermissibility. Critical to seeing how it does in this context is to remember the relevance of the clear danger posed by many markets. If immoral versions of these markets are introduced (and there are many such versions), then the harm will be great. Even if we have a near miss and mis-structure the market in question just a little bit, this may result in great harm.Footnote 20

Pursuing moral versions of these markets thus involves the risk of achieving one of the immoral versions of them.Footnote 21 And this risk will not typically be borne by policy-makers and advocates. If we attempt to make a moral prostitution market, we risk making a market that instead severely harms vulnerable populations. So, it seems that we are not permitted to pursue these markets if they are infeasible, as pursuing the infeasible in this context brings with it these risks, and acting with these risks is morally unacceptable. This is just as premise one claims. Still, let’s use this opportunity to think a little bit more about why it would be true.

In thinking about how to factor in the risk involved here, we might be tempted to suggest that the value of pursuing these markets should simply be risk-adjusted. A successful version of the market may involve minimal harm and a massive benefit to society, but it is not guaranteed that we will successfully implement this version. The suggestion would thus be to evaluate pursuing the market not only on the basis of the benefits of successful implementation, but also with the costs of failed implementation, with the values of these costs and benefits adjusted on the basis of the likelihood of the various outcomes.Footnote 22

If this were the right way to arrive at a decision about whether we should pursue a particular market, this alone is enough to tell against the Markets without Limits thesis. Granting that there is a morally acceptable market structure for a good is still well off from an argument that we are justified in pursuing it. Once we appropriately factor in the possible harms of pursuing infeasible policies, I doubt the presumption will be that markets for all goods (that it’s permissible to have for free) will be judged permissible. Pursuing many markets remains risky, and these risks can be significant.

However, I am skeptical that this is even how we should be thinking about the decision at hand. For one thing, as is well-known in decision theory, we tend to be risk averse, and our risk aversion complicates if not invalidates this standard decision-theoretic model for how to make decisions under risk (e.g., Allais, 1953; Buchak, 2013). More to the point, though, it does a disservice to treat the kinds of harm at issue as something to be simply tallied up and weighed against the benefits.Footnote 23

It is not just that we run a risk of harming individuals by pursuing infeasible markets. We may well think that imposing that risk of harm itself constitutes a kind of wronging. It seems wrong for me to do something that risks your life, and there has been a recent flourishing of work on how to think about the ethics of imposing risk on others (Athanassoulis & Ross, 2010; Bovens, 2019; Hayenhjelm & Wolff, 2012). For example, imposing risks has been construed in terms of infringing rights, rights not to have certain risks imposed on us (Hansson, 2003; Oberdiek, 2017; cf. Holm, 2016).Footnote 24

Agreeing that for some good there is a market structure that would not violate any rights if it were to be executed, the point is that pursuing that market in the current economic and political environment may well violate certain rights. We have rights not only against being harmed but being imperiled in various ways. Pursuing a market that could easily end in mass exploitation would constitute such an imperiling. Thus, we should not do it; we are not permitted to pursue a market when it is infeasible that we will be able to implement the moral version of it.Footnote 25

Second Reply: Challenging the Infeasibility of Moral but Repugnant Markets

If it really would be infeasible to implement a version of a market in kidneys, say, that avoids the massive harms likely to accompany any such market, then perhaps we are not permitted to pursue it. But is it infeasible? I think Brennan & Jaworski themselves are likely to push back on this very point. Generally, a commodification theorist may maintain that there is nothing so infeasible about moral market structures.

Taking kidneys as an example, though there are possible problems with a market in kidneys, Brennan & Jaworski think that there are concrete and plausible steps we can take in the design of such a market to avoid them (op. cit.:206–207). As they suggest: To avoid exploitation, we could prohibit selling your kidney if you make less than a certain amount of income. To avoid coercion, we could institute a Universal Basic Income (UBI). To maintain the sanctity of the body, we could mandate various practices around the sale of kidneys. To ensure fair access, we could institute a lottery system for who would be eligible to buy the kidneys. There may be genuine moral concerns raised by the prospect of a market in kidneys, but none of them, as far as Brennan & Jaworski can tell, cannot be met with clear and even uncomplicated regulation.Footnote 26

Are they right? Well, let’s stick with the example of kidneys and consider it. First, I am skeptical that the needed regulations would be this simple. More regulations will of course be required, as well as regulatory bodies, enforcement agencies, audit mechanisms, legal infrastructure to provide compensation for abuses, cultural practices to disincentivize deviance, and so on. Perhaps much of this could be extended from systems already in place, and perhaps not. Even if we had sufficient and fairly simple policies in hand, though, their simplicity only marks surpassing one hurdle to their proper implementation.

The real challenge is not to the simplicity of the necessary regulations. After all, a ninety percent tax on inherited wealth or a law that bans swearing are simple enough, but how feasible are they? A policy may be perfectly simple, communicable, understandable, yet absurd. From the perspective of the current global economy, political environment, regulatory regime, and so on, it is my sense that many of the repugnant markets at issue are infeasible in this deeper sense.

For any market, we should recognize that the feasibility of acceptable versions of it depends on the feasibility of the various policies that we would need to implement to facilitate it. But then we have to ask: How feasible are these individual policies? Have we ever advocated for, implemented, and successfully enforced policies like these in the past? Do we even agree with all of the policies on their own, and would we be happy to see them applied in other apparently similar contexts? Moreover, since morally acceptable versions of some of these markets will require a whole suite of policies meant to address different possible concerns, we should further wonder: If these individual policies might be infeasible, is it not even more infeasible to attempt to implement all of the policies necessary for facilitating a moral market? And to what extent does the safety of these policies depend on their being collectively implemented?

I will not blanketly claim that these questions cannot be answered in favor of pursuing a market in kidneys or any other repugnant market. It is hard to imagine, though, that the discussion is likely to come out this way. Let’s go back to the example of kidneys. Brennan & Jaworski offer several possible regulatory solutions to apparent moral problems with these markets. How feasible are these proposals?

Although advocacy for a Universal Basic Income has grown recently, many hurdles to its successful implementation remain: It has not been shown to be economically practical. There are still unsettled debates about whether it would have the desired effects, whether it is an inefficient allocation of government funds, or whether the government is even justified in using its revenue this way. So, supposing it is true that UBI is an appropriate allocation of government funds and would have the desired results, we would still need to become confident ourselves that this is the case, convince the public, and bring it into law. As a policy, then, UBI does not seem terribly feasible.Footnote 27

We could say the same thing about the policy of disallowing economically disadvantaged citizens from selling their kidneys. This policy is simple, but what precedence is there for a policy like this? Apart from needing a certain amount of money to get a mortgage, or laws requiring you to buy lottery tickets with cash, I cannot think of one. And how feasible is it to make a restriction like this but to do it just for kidneys? Wouldn’t the introduction of such a regulation be much more sweeping, causing us to rethink other markets? If we say that you have to have a certain amount of money to sell a kidney, why not to sell blood plasma, or why not even to buy alcohol? These cases do seem unduly paternalistic, but we would need a way of distinguishing the unduly from the acceptably paternalistic policies, then we would again need to convince the public and properly implement it. Successfully engaging in this process sounds infeasible, and this time it also sounds dangerous insofar as it threatens paternalism.

Finally, we can quickly turn to the policy suggestion of a lottery for buyers of kidneys. There is precedence for such policies, so it does seem feasible. But now we have to worry about the dangers that arise purely from this policy, such as the creation of less well-regulated secondary markets for kidneys. Without the right further policies in place, this policy will only forestall high prices for kidneys, and the necessary further policies may themselves be infeasible. For example, we could disallow the resale of a kidney gotten through a lottery. However, we generally do not stop the resale of goods owned. So, we would again be thrust onto a challenging policy decision about which scholars disagree.

None of this is to suggest that the right policies aren’t out there. Even stipulating that they are, it is far from known is that such policies are feasible.Footnote 28 Brennan & Jaworski quickly make the point that regulations for addressing this or that moral concern seem ready to hand (ibid.). And so, the presumption should be that there will always be some regulatory fix that in principle could correct for our moral concerns with a market. I agree. My point is that even granting this, we have not earned the presumption that these markets are feasible. The scope of the feasible is much narrower than the scope of the metaphysically possible. As such, the burden is actually on the proponents of a market for a good that seems repugnant to show us how a market in it is feasible to achieve.

For a final pushback on this point, a commodification theorist may resist giving the details and instead appeal to induction in the abstract: We have successful markets in certain goods right now that used to be thought of as repugnant. And so, we should infer that the markets that seem to be repugnant today should be viewed optimistically. We should expect that we can have successful markets in them, eventually ameliorating our sense of disgust. Though we cannot point to exact policies, the process of finding sufficiently good policies for repugnant markets is feasible. If this is right, and repugnant markets are the hard case, then we can infer that generally moral versions of markets are feasible.

Something similar to this approach is taken up by Brennan & Jaworski (ibid.:210–214). They run through a surprising list of paid professions (including laborers, teachers, lawyers) as well as other commodities (interest on loans, children’s insurance). These are things we today think should be paid for, but that scholars used to debate and find repugnant. Brennan & Jaworski are trying to undermine our expectation of continued repugnance once we set up a market, but we could imagine this to also be a way of arguing that acceptable versions of currently repugnant markets are feasible.

However, I am not hopeful about this approach. Southwood and Wiens (2016) have argued that the successful implementation of a policy is not a good guide to how feasible it was before its implementation. This does not show that feasibility does not matter; instead, it suggests that, for these supposedly successful markets, we could have just gotten extremely lucky in how we implemented them, blithely risking moral catastrophe along the way. Perhaps we could disabuse ourselves of the intuition of repugnance if we were to successfully implement moral versions of repugnant markets. But this is not to show that achieving these versions of these markets is feasible from our perspective today.

Separately, it is also not clear that the markets being pointed to were all that successful. Don’t we still find systemic abuses sanctioned if not institutionalized by the law for these commodities? Our system of pro bono defense law, for instance, is a patchwork fix as far as most of us are concerned given the unfair difference in legal representation. These commodities may be valuable to us, and we may not be able to do without them at this point, but we cannot pretend that we got the morally permissible versions of these markets. If it is true that we no longer find these markets repugnantFootnote 29, what this shows is just that repugnance is not a good guide to morality, a point with which Brennan & Jaworski are likely to agree.

If anything, then, the past battles over commodification should teach us that it is feasible to commodify things in ways that make those markets entrenched. A battle to commodify sex may be a battle that we can win. But this is all the more reason to proceed with caution, as it is highly infeasible that the market we will be stuck with will be a market that we can accept.Footnote 30

Facing Infeasibility

Consider that there may be some sense in which a novice can hit a bullseye; they might get lucky. Still, they are not permitted to throw if someone happens to be standing next to the dartboard. It is in this spirit that I have argued that Markets without Limits is mistaken. Perhaps we really can construct a market for sex or kidneys that would avoid infringing anyone’s rights or facilitating massive harms. But we have not yet seen how pursuing these markets avoids unjustly imposing risks, and so we are not permitted to pursue them. Though an important result, what we are left with is quite negative. It would be fair to wonder what we should be doing or how we could use this information about infeasibility. To conclude, then, I will consider several ideas about how to proceed in the face of infeasibility.

First, we can note a way around the apparent impermissibility of infeasible markets. The markets are impermissible because they risk harms to citizens, and these individuals have rights against having these risks imposed on them. However, we would be permitted to pursue these markets if individuals waived these rights.

If everyone agreed that a market in kidneys would be so beneficial that it was worth the risk, then it would seem permissible to pursue this market despite the risk. And the agreement may not need to be explicit. Driving is risky to both drivers and pedestrians, but we accept this risk as a necessary part of living in our society. This seems so even if we did not ever consent to such a system explicitly. Similarly, if we found ourselves with a market in kidneys because a government representing us pursued it, then this may count as our collectively consenting to the risks involved. There are big assumptions here: This involves tying state legitimacy to consent of the governed, and assuming that the state is in fact legitimate. Still, this could be a way to permissibly pursue infeasible markets.

As a second point to make, we can see how the possibility of conflicting obligations is relevant here. Assume that a market for kidneys is impermissible because we have an obligation not to risk certain harms. Still, if a market for kidneys would save many lives, then we may think that we actually have an obligation to set up the market for it (perhaps following from a duty to rescue). Thus, then we would have conflicting obligations.

As is sometimes noted, ‘ought’ does not entail ‘must’ (e.g., Vranas, 2018). So, our being obligated not to pursue the market in question may not mean that we must not pursue it; that obligation can be overruled by stronger countervailing considerations. In the case of kidneys, if many lives will be saved by a market in kidneys, then perhaps we all-things-considered ought to implement a market for kidneys. But this is no vindication of Markets without Limits. This is just to say that rather than being stalemated about how to proceed given this conflict, it may be that the proper course is to implement the imperfect market for kidneys.

Since we will have failed to meet an obligation, we may incur additional duties: We may have a preemptive duty to take precaution as we set-up the market to attempt to mitigate the possible damage. (See Sandin [2004] and a vast literature on how to think about a precautionary principle such as this.) We may also incur obligations of moral repair or corrective justice to compensate victims (Hayenhjelm, 2019; Lazar, 2008; Walker, 2006). Thus, when learning that a market is infeasible, this may not recommend against trying to implement it; instead, it could warn us about what further steps we would need to be prepared to take to mitigate the damage and correct for damage caused.

Even if in the end we decide against implementing these markets because of their infeasibility, a final point is that this does not mean that the information about their infeasibility is useless. What counts as feasible/infeasible can change over time as our abilities expand or contract, as Gilabert (2017) recognizes. There, he suggests that the notion of feasibility can be used to inspire our political imagination, and that we can even have ‘dynamic duties’ to work to change what is feasible for us. I think this is right, and it follows from a natural but walked-back version of Markets without Limits.

When we hear the Markets without Limits thesis, there is something deeply appealing about it. Engaging in market exchanges can at times lead to tremendous societal gains, and there may well be versions of repugnant markets that are acceptable and only leave us with these gains. The problem is that it is not clear that these versions of the market are accessible to us; they aren’t feasible. But, well, they should be.

Given this, another version of the thesis that I would endorse would be the following: ‘If you may do it for free, and it would add tremendous value to be able to do it for money, then you ought to be permitted to do it for money.’ I take this to provide a standing obligation. We ought to explore policy options towards the end of making feasible those markets that it would provide substantial benefit to allow, even if at present they may seem repugnant.