3.1. The Generality Principle of Exchange-Desert

Having clarified the metaphysics of wages, I want to now raise the first puzzle for the idea that wages are deserved. To be employed is to be a party to a certain kind of agreement to exchange labour for income. In the ideal case, both sides benefit, and the exchange is fair. Determining whether the exchange is fair is, of course, a difficult matter, and depends on which principles of distributive justice one accepts with respect to wages. On a desert-based approach, the fairness of a wage will depend on, at least in large part, whether or not it is deserved. Whether or not the wage is deserved, in turn, will depend on one’s theory of the desert of wages. It might be, for instance, that the wage is deserved on the basis of the employee’s productive contribution, or perhaps their efforts, or perhaps as compensation for the sacrifice they make in performing their work (rather than doing something else that they would rather do.)Footnote ⁵

I do not intend to wade into the details of these competing views about what makes one deserving of a wage. All such views presuppose that wages are deserved, and then try to give an explanation of why that is the case. But the plausibility of those explanations depends on the possibility of deserving wages in the first place, and that is what I intend to investigate.

Starting with the idea that an employee can deserve a wage, and that a wage is just one half of what is exchanged in an employment relation, a simple ‘generality principle’ suggests itself:

Generality Principle of Exchange-Desert: If an employee can deserve a wage, then parties to other kinds of economic exchanges can deserve what they get in those exchanges.Footnote ⁶

This principle already raises some interesting questions. In particular, is there anything special about jobs qua economic exchanges, or do parties in other kinds of economic exchanges very often deserve what they get as well? If I purchase seeds from the garden shop, do I deserve those seeds once I pay for them? Does the garden shop owner deserve my money once they give me the seeds? Despite the extremely varied usage of ‘deserves’ in ordinary discourse, I do not think it is common to use the language of desert when one talks about their ordinary purchases. Thus, if we suppose that wages can be deserved, and that the generality principle is true, then we get some counterintuitive consequences with respect to what can be deserved.

And these counterintuitive consequences should not be dismissed so lightly. For most desert theorists, one of the goals of a theory of desert – as is typical in philosophy, generally – is to justify common claims of ordinary discourse. However, because in ordinary discourse ‘deserves’ is often used where the language of rights or obligation is most appropriate, theorists also very often have to discount such claims as what Miller (Reference Miller1999: 133) calls ‘sham desert’ – i.e. uses of ‘deserves’ which are probably best interpreted as appealing to some notion other than desert. Because desert claims are so ubiquitous and wide-ranging, it is often taken to be necessary for a theory of desert to greatly circumscribe the justifiable desert claims, so that the notion of desert can be made orderly enough to be of use in normative theory. Thus, that the generality principle proliferates the justifiable desert claims in ways that are not typically recognized in ordinary discourse ought to be cause for concern, especially if, as I argue, the consequences are problematic, at least for those who think of justice in wages in terms of desert.

Of course, the generality principle might be false. In that case, there would need to be some non-arbitrary reason for thinking that jobs are special with respect to the economic exchanges they involve, such that wages can be deserved, but what is received in other kinds of exchanges is not. I will take up an objection to that effect in ${\rm{\S}}$ 3.4. For now, though, I will simply assume that the generality principle is true so that we can further explore its consequences.

3.2. Double-Counting Injustice in Exchange

If one thinks that justice in wages is getting the wage one deserves (the meritocratic or ‘desertist’ view of wages) – or at least partly consists in getting what one deserves – then the generality principle is particularly problematic for the idea that wages are deserved. Consider a case where I pay for the seeds at the garden shop, but then after handing over the cash, the owner refuses to give me the seeds I have paid for. In this case, an injustice has occurred. But it would seem that the injustice can be fully explained in terms of the ethics of agreement, entitlement or promise. The shop owner has committed fraud, has broken the terms of our agreement in the exchange, and perhaps has failed to give me the seeds that I am now entitled to.

While the fact that one deserves something neither entails, nor is entailed by their being entitled to it, this doesn’t rule out the possibility that I also have failed to get what I deserve in this case.Footnote ⁷ But, at least for the meritocrat, if I have also failed to get what I deserve, then it would seem that this simple case of fraud is a greater injustice than a simple fraud case. After all, for those who understand distributive justice in terms of desert, it is a considerable injustice if an employee gets less than they deserve. But again, it would seem that appealing to the ethics of agreement and/or entitlement would suffice to explain the injustice that occurs in the garden shop case, and similar cases in the context of simple purchases. Thus, either I deserve the seeds, and desert does not play the same explanatory role in the context of simple purchases as it does in the case of employment, or else parties to simple purchases do not deserve what they get in the exchange. The challenge, once again, is to explain what makes employment a special kind of exchange with respect to desert – either because desert only plays its role in explaining injustice in the context of employment in particular, or because employment is unique in that parties to the exchange can deserve what they get in the exchange.

3.3. The Contingency of Injustice in Wages

Now, there is an interesting disanalogy between a case where an employee gets a wage that is less than they deserve, and a case where one pays for an item in a simple purchase, but does not receive the item. Pointing to this disanalogy highlights an independent problem with the idea that employees deserve wages, and so puts additional pressure on the idea that desert plays a special role in explaining injustice in wages.

In the case where the garden shop owner commits fraud, the agreement to exchange seeds for cash is not honoured by the garden shop owner. But in the case of employment, an employer can give an employee less than they deserve while still honouring their agreement. All that is required is that the employee agree to work for less than what they will deserve, whatever that wage may be. This, presumably, is a common occurrence. But desert, generally, is taken to require some sort of proportionality.Footnote ⁸ And so in the context of employment, the wage that is deserved will be proportional to whatever is one’s preferred basis for deserving wages – the value of the employee’s contribution, or the magnitude of their sacrifice, for example.

It might be thought that the disanalogy shows how we could preserve some special role for desert in explanations of distributive injustice in wages, since appealing to entitlement or agreement would not account for the injustice in these kinds of cases. But considering the more apt analogy to simple purchases raises more problems. The more apt analogy, here, would be a case where one agrees to buy an overpriced item. The very idea of an ‘overpriced’ item suggests that the exchange is not proportional – the seller is getting more and the buyer is giving up more in the exchange than they ought to. Likewise for the employee who gets less than they deserve (regardless of one’s preferred account of the basis of deserving wages). Once again, however, if a failure to get what is deserved explains injustice in the case of wages, we might think it should do so in the case of simple purchases as well.

But suppose I buy grossly overpriced seeds from a garden shop which is located in a shopping plaza full of more reasonably priced garden shops. I buy the overpriced seeds, let’s say, because I am wealthy and it makes no difference to me how much I spend on the seeds. We can also suppose the seeds are overpriced not because of any greedy intentions, but just because of poor business sense. The exchange is imbalanced in the relevant sense, but it doesn’t seem as though there’s any injustice here.Footnote ⁹

And in fact, it seems that the same kinds of intuitions arise in the relevantly similar employment contexts. Suppose I am independently wealthy, and my skills are in high demand. I have several job offers from various employers, some of whom offer to pay me a wage which is what I would deserve in doing the job, and some less. After considering the options, I choose (for whatever reason one likes) the one that would pay me less than I deserve. Once again, it does not seem, intuitively, like there is injustice here. But given a desert-based approach to justice in wages, it’s hard to see why not. Such an approach would seem to be indifferent to facts about available alternatives – unlike an account of exploitation, for example, which would presumably be sensitive to both the terms of the agreement and the availability of reasonable alternatives. The desert theorist can say that it is an extra source of badness when people have few choices, because they independently value autonomy, but it’s not clear why the injustice would be less in the kind of case where there are lots of good options available.

Now, it might be objected that these examples fail to show that the desert-based conception of justice gives us the wrong results, because while someone does not get what they deserve they, in effect, waive their claim to what they deserve, and so injustice is avoided.Footnote ¹⁰ Their waiving their deserts would be reflected by the fact that they had the option to be paid what they would deserve by taking a different job, but chose a job which would pay them less. But I do not think this is a viable response for defenders of desert-based views of the desert of income. In general, facts about desert do not entail or ground facts about rights and vice versa. The fortunate child might have a right to the inheritance, but does not deserve it, and the hardworking employee (it is supposed) might deserve a raise, but has no right to it until it is granted. Thus, desert-facts are not thought to give rise to claims the way that rights do, and so the employee who deserves higher pay has no claim to waive. Just as the retributivist requires that the criminal get the punishment that they deserve – and neither the criminal nor the state may waive any claim to the criminal’s being punished – the desertist thinks that justice is a matter of getting what one deserves.Footnote ¹¹ At best, the desertist can maintain that the employee who freely chooses to get paid less than they deserve simply chooses to bear an injustice. But it seems to me that in the case I describe, there is no injustice to explain in the first place.

3.4. Is Employment Special Qua Exchange?

I’ve argued that the generality principle gives rise to unusual and potentially problematic consequences for the desertist. But thus far, I have simply assumed the principle in order to explore its consequences. Let me now defend the generality principle against an important objection.

The desertist might argue that we have good reason to reject the generality principle because, in fact, employment is an importantly distinct kind of economic relation, and labour is an importantly distinct kind of economic good, such that income from employment can be deserved, but goods in ordinary purchases cannot.Footnote ¹² Indeed, they might argue, the uniqueness of employment and labour can be traced to the very features that they posit as desert bases for deserving employment income: effort, productive contribution, sacrifice, or subordination of the employee’s will to that of the employer. Of course, it’s not a settled matter just which of these features of labour and employment ground desert of labour income, but on the assumption that one, or some combination of these features, grounds desert of employment income, this would suffice to distinguish employment as an importantly distinct kind of exchange in which desert can operate. Thus, since employment has desert-grounding features which other kinds of exchanges lack, employment income can be deserved while the goods in other kinds of exchanges typically cannot, and so the generality principle is false.

Responding to this objection requires some care, as I think it can proceed in two different ways. I will describe both ways and then argue that neither provides a satisfying reason to reject the generality principle.

In considering the first way that the objection might go, it is worth reflecting on the standard methodology that the desert theorist employs in arguing for a particular account of the desert of income. When desert theorists attempt to specify a desert basis for income in employment, they typically do not begin by identifying some desert base which, generally (i.e. outside of the context of employment) makes one deserving of income or some comparable benefit. The idea that the virtuous deserve to be happy, for example, is perfectly general in this way: virtue makes one, pro tanto, deserving of happiness, wherever and however that virtue is exemplified. Rather, they begin with the assumption that employment income is the sort of thing that can be deserved, and then consider what would provide the best grounding explanation for the desert of income – perhaps it is the value of the productive contribution, or the sacrifice made by the employee, and so on. These explanations are checked against intuitions about who deserves what and how much, and against other principles of desert, such as the proportionality constraint.

To be clear, this is not an unprincipled or unusual philosophical methodology. After all, it is very common in ordinary discourse to speak of someone deserving a certain wage, or not getting the wage they deserve, and so on. One of the tasks desert theorists (and many philosophers, generally) set for themselves, very often, is to vindicate these kinds of ordinary claims.Footnote ¹³ A benefit of this methodology, too, is that the theorist does not need to generalize the desert-grounding feature of employment to non-employment contexts, which, depending on the theory in question, will very often lead to counterintuitive consequences. For instance, making efforts or sacrifices, doing unpleasant tasks, and so on, are not thought to ground the desert of anything comparable to (or proportional in value to) the relevant sort of income outside of the employment context. Likewise for making productive contributions to others: we contribute to the well-being of others in all manner of ways – as friends, as parents, as consumers, as workers, as entrepreneurs, as citizens, and so on.

But if the desert base for desert of employment is not to be generalized beyond the employment context, then the objection is question-begging. If the desert base is not generalizable, it’s not the properties of having made sacrifices, or having contributed to the well-being of others, and so on, which ground desert of employment income, but rather the more complex properties of having-made-sacrifices-in-the-course-of-employment, or having-contributed-to-the-well-being-of-others-in-the-course-of-employment, which ground the desert of income. But this approach begs the question against the generality principle. It is simply built into the theory of what grounds desert of income that employment is special in this regard. Thus, merely providing a grounding explanation for the desert of employment income, and then noting that other kinds of exchanges lack the relevant desert-grounding feature, fails to offer a non-arbitrary reason to reject the generality principle. To reiterate, it would be tempting to say, for example, ‘employment typically involves making efforts and applying one’s skills, and so income is deserved, while simple purchases lack this feature, and so the generality principle is false’. But on the approach I’ve just sketched, this is misleading, because it is not merely making an effort and applying one’s skills which makes one deserving of income, but rather making those efforts in the course of employment. Thus, it is trivial that other kinds of exchanges lack this feature, and the resulting objection is question-begging.

If the objection is to work, then, the relevant desert-grounding feature of the desert of employment income must, in principle, be generalizable beyond the employment context. This kind of argumentative strategy will be effective against the generality principle if either (a) the desert-grounding feature of employment income rarely, in practice, extends beyond the employment context or (b) the desert-grounding feature of employment income extends beyond the employment context without too many counterintuitive (or otherwise problematic) consequences, but does not typically extend to other kinds of exchanges like simple purchases.

Consider, for example, the view that it is one’s productive contributions that make one deserving of employment income. This view would seem to require approach (b): as I noted above, we contribute to the well-being of others in all manner of ways beyond our roles as employees, if we happen to occupy such a role. One concern, of course, is that the consequences of generalizing the productive contribution view are too counterintuitive, as it might entail that we deserve benefits outside of the employment context that are proportional to the pay we might receive for making the contributions in the course of employment. Good parents and great friends might, perhaps, deserve to be quite wealthy. If one thinks that these kinds of consequences are objectionable, then they might not want to generalize their theory of desert of wages beyond the employment context. In that case, again, the objection to the generality principle is question-begging.

On the other hand, a desertist might simply accept these consequences. Mulligan (Reference Mulligan2018a: 214), for example, suggests that he is at least willing to accept that parenting and acts of altruism might ground desert of income. If this move is made, however, it may limit the range of what counts as a counterexample, but it would also embrace the generality principle, as there is no reason to think that by engaging in other kinds of economic exchanges such as simple purchases, one is not making a contribution to the well-being of others. Indeed, if we assume that exchanges are typically mutually advantageous, then this will almost always be the case. Thus, appealing to the contribution view of desert of income fails to provide a successful challenge to the generality principle.

Sher’s (1987: Ch. 6) proposal for grounding the desert of wages is, I think, the most promising way to give a response of type (a). He argues that it is the fact that employment typically involves the subordination of the employee’s will to that of their employer, and thus that the employee is treated as a mere means, which grounds the fact that an employee deserves income for their work. When we do something for a friend, we do so because we value furthering their ends, and so ‘make [their] purposes our own’ (102). According to Sher, in a typical case of employment, ‘the worker functions merely as a means’, and thus, compensation is deserved (102). The subordination of the will that can happen in the context of employment certainly is not typically present in the context of simple purchases, and so this might point to a reason for thinking that employment is unique qua economic exchange.

Without going into too much depth here, I will just note that there are reasons that Sher’s view is not particularly popular among desertists. First, it seems implausible that an employee’s deserved income is just a function of the degree to which their will is subordinated to the employer’s. Facts about effort and contribution, for example, which often drive our intuitions about how much is deserved, will be mostly irrelevant. Second, it’s not clear that all employment involves functioning merely as a means. In general, an employer may treat employees in such a way that respects them as ends, or employees might value the ends of an employer and so make the employer’s ends their own. Indeed, a firm, via effective propagandizing of their workers, might get the wills of their employees to align with the wills of the shareholders, and so the employees would deserve no income.

While I do not have the space to consider every view on the desert of employment income, I think similar kinds of responses can be made for them. And it also should not be lost in these details that, presumably, when one performs their labour and comes to deserve income, they deserve income from the employer, and justice requires that the employer pay the employee. In other words, if an employee works their job, and for whatever reason, a family member gifts them something proportional in value to their deserved income, justice is not yet served and the employee still deserves their income from the employer. But in that case, the fact that the productive contribution (for example) is performed in the context of employment is critical to explaining what the employee deserves, and so it is, once again, simply built into the explanation that employment is special with respect to desert.

At minimum, I hope to have shown that rejecting the generality principle is not so simple as pointing to differences between employment and other kinds of exchanges. Indeed, if I am right that the productive contribution view does not have a good, non-arbitrary reason for rejecting the generality principle, then this would be a significant result, as that view is probably the dominant one among contemporary desertists.Footnote ¹⁴ If rejecting the generality principle required rejecting this view about the desert of income, this might be a reason to simply accept the generality principle, and to try to deal with the consequences it brings in other ways.

3.5. Summarizing the Dialectic

Let me sum up so far. If the arguments of this section are good, then there are, in effect, three main options that we have with respect to the generality principle. The first option is to simply reject the generality principle. This option requires, however, some non-arbitrary, non-question begging account of why wages, in particular, can be deserved, but parties to other kinds of exchanges such as simple purchases, cannot deserve what they get in those exchanges. If the meritocrat (who understands justice in wages at least partly in terms of an employee getting what they deserve) takes this option, then they also need some answer for the double-counting of injustice problem, and also an explanation of why the availability of good alternatives in accepting a certain wage seems to sometimes override considerations of desert when considering the justice of a wage (the contingency of injustice problem). The second option is to accept the generality principle, accept the counterintuitive consequence that buyers can deserve their simple purchases, but reject the idea that desert has any connection to justice in wages. This option, however, is not open to the meritocrat, who sees desert as having an essential connection to justice in wages. The third option, which I support in this paper, is to accept the generality principle as vacuously true. In that case, employees do not deserve wages, and neither do parties to simple purchases. In this case, all of the counterintuitive consequences I raised are avoided, but some alternative account of injustice in wages must be given, which I offer in section 5. This option, also, is clearly not available to the meritocrat. Thus, the meritocrat must either find some non-arbitrary way to reject the generality principle, or else argue that what follows from the generality principle is not, in fact, problematic for their view.

4.1. The Symmetry Principle of Exchange-Desert

In this section, I want to consider another principle of desert which focuses on the symmetry of the employment relation qua exchange. I will argue that it, too, leads to conceptual difficulties for the idea that employees deserve wages, and extra difficulties for the idea that justice consists in getting the wage one deserves.

From the fact that a job involves an economic exchange between employer and employee, it follows that the employment relation has a certain symmetry. Of course, there are plenty of senses in which employment is typically asymmetrical. As Sher (Reference Sher1987: 101) notes, for instance, being employed means (at least much of the time) subverting one’s will to that of the employer. There are also, of course, very often asymmetries in status and power involved as well.Footnote ¹⁵ Nevertheless, there is always a simple metaphysical symmetry that obtains in any employment relation, in that both employer and employee are parties to an economic exchange.

Consider the following ‘symmetry principle’:

Symmetry Principle of Exchange-Desert: If one party to an economic exchange can deserve what they receive in the exchange, then the other party can deserve what they receive in that exchange as well.Footnote ¹⁶

The idea here is simply that if having a job involves being party to a particular kind of exchange, then it is unclear – at least without further elaboration – why only the employee could deserve what they get in the exchange. In an ideal case, we hope that the exchange is equal (in whatever the relevant respects are), and in such cases, it’s hard to see why if one party can deserve what they get in the exchange, the other couldn’t as well.

But then, just as with the generality principle discussed in the last section, we immediately run into some counterintuitive consequences and some difficulties. The most immediate consequence is that it would seem that just as an employee can deserve a wage, the employer can deserve the employee’s labour. This, already, is at least unusual. As natural as it is to think that an employee deserves a wage, it is very unusual to hear the idea that an employer deserves an employee’s labour. If we combine the symmetry principle with the generality principle, then the observation extends to other kinds of exchanges, such as simple purchases, as well. Not only do I deserve the seeds from the garden shop when I purchase them, but the garden shop owner deserves my money as well. These consequences, once again, not only proliferate justifiable desert claims well beyond the bounds of ordinary discourse, but will also generate problematic consequences about desert and injustice for desert-based theories of justice in wages.

As with the generality principle, I will begin by supposing the symmetry principle is true so that its consequences can be explored. I take up the objection that the employee occupies a special role in the employment relation, and thus that the symmetry principle is false, in section 4.4.

4.2. The Asymmetry of Injustice in Wages

Now, again, if some seemingly ‘unnatural’ consequences follow from these principles, we might be inclined to simply accept those consequences. Perhaps employers do deserve the labour of their employees, and it is simply unusual to hear such claims because, given the power asymmetries in typical employment, we are usually concerned with justice in wages on the employee’s side. Maybe, for instance, if we lived in a world where employers were plentiful and labour was scarce, so that employees regularly secured exploitative arrangements for themselves, it would be much more typical to hear that an employer deserved more labour from their employees.

But just as we saw with the generality principle, the idea that employers deserve the employee’s labour conflicts with our intuitions about injustice. In our world, where employers very often have plentiful options in hiring, there would be no failure of distributive justice if, say, a firm decided, for whatever reason, to overpay all of its employees, relative to what they deserve.Footnote ¹⁷ From the standpoint of desert, however, this is as bad as the employees being similarly underpaid, as the employer is getting proportionally less labour than they would apparently deserve. But even if this is non-ideal from the standpoint of justice, it certainly does not seem as bad a case where an employee is underpaid.

This particular consequence will be most problematic for those who think of distributive justice solely in terms of desert (e.g. Mulligan Reference Mulligan2018a). It will not be a problem for pluralists, who can simply note that employees are much more often in positions of vulnerability, for instance, and so additional problems of justice typically arise in cases where employees are underpaid. Nevertheless, the pluralist must also accept – perhaps counterintuitively – that there is injustice when the employees are overpaid, as the fact remains that they get more than they deserve.

4.3. The Temporality Problem

Additional problems arise as soon as we ask, ‘In virtue of what does each party deserve their share of the exchange?’. Typically, it is thought that an employee cannot deserve their wage until they actually perform the labour that they have agreed to perform. Sher (Reference Sher1987: 99), for example, writes, ‘it is uncontroversial that labour is generally considered a burden, and that wages are never deserved until work is actually done.’ The standard theories of deserving wages will then say that it is something about their labour – either the productive contribution that results, or the sacrifice that they make in performing it, for example – that makes them deserving of the wage. The natural thought, then, would be that the employer deserves the employee’s labour in virtue of paying the employee’s wage. And just as an employee might deserve a higher or lower wage depending on facts about their labour and wage, perhaps an employer might deserve more or less labour if they are paying their employee too much or too little.

But if an employee deserves their wage in virtue of performing the agreed upon labour, and the employer deserves the labour in virtue of paying the agreed upon wage, then it would seem that either the labour or the wage could be deserved, but not both. If the wage is paid after the labour is performed, then the labour could not be deserved, since it is presumably in virtue of paying a wage that the labour would be deserved. If the wage is paid first, then it could not be deserved, since it is only deserved on the basis of performing the agreed upon labour. Similar considerations apply to simple purchases as well – if an item is paid for before entitlement to it is transferred, then the payment cannot be deserved, and mutatis mutandis for a case where entitlement is transferred before it is paid for. It would seem, then, that it cannot be the case that both parties to an exchange can deserve what they get in the exchange. Note too that this does not constitute an argument that the symmetry principle is false – but rather either that wages are deserved and the principle is false, or that wages are not deserved, and so the principle is vacuously true.

In a case where wages are distributed simultaneously with, and in proportion to the employee’s labour, so that neither side accumulates a ‘debt’ first, then it would seem that neither side deserves anything. After all, once the employee is paid for their labour, they don’t deserve anything more. A wage is only deserved, on the standard view, when they have performed labour that has not yet been compensated.

Now, one could object that this line of argument assumes that a wage could only be deserved on the basis of past actions (either performing the labour or paying the wage), and that this assumes the more general idea that one can only deserve something on the basis of past actions (or present qualities). While most desert theorists accept this general principle, some do not, so perhaps one could reject the principle and avoid this particular problem that the symmetry principle raises (Feldman Reference Feldman1995; Schmidtz Reference Schmidtz2002; Kinghorn Reference Kinghorn2021).

As far as I know, such a view has not been defended for the case of wages, however, and so even if desert is sometimes forwards-looking, I see no reason to think that it might be in the case of wages.Footnote ¹⁸

A better option might be to appeal to Schmidtz’ (Reference Schmidtz2002) view on the desert of wages.Footnote ¹⁹ Schmidtz argues that while a wage is not deserved on the basis of what one will do, if a wage is paid before the work is done, then it will become deserved upon completion of the work (785). In that case, the temporality problem is avoided. If an employee performs the labour before getting the paycheque, they may fail to get what they deserve initially, but once they get the paycheque, they get what they deserve. If we apply the symmetry principle, then the employer, on the other hand, does not deserve the labour when it is first performed, but once the employee is paid, they come to deserve that the labour was performed. Mutatis mutandis for the case where the wage is paid before the labour is performed.

The fundamental question at play here is whether the ‘compensatory notion’ of desert (in Schmidtz’s sense) is correct or not in the case of wages. Schmidtz’s view rejects it, while my presentation of the temporality problem assumes it. On the compensatory model, one deserves something when the ‘moral scales are put out of balance’ (779). There are two important consequences of this view. First, objects of desert (i.e. the goods that are deserved) are always what are needed to restore the moral balance. Second, desert is strictly present or forward looking when it comes to the objects of desert. To deserve something is to be such that we have some reason to ensure that the deserving gets it.

If desert works like this, then when the employee performs the labour before getting paid, the employer can never come to deserve that the labour was performed. To be deserving of that, the ‘moral balance’ would have needed to be tipped against the employer at some point, but it never was. The employer only incurs a debt to the employee, and never the other way around. If, on the other hand, the employee is paid first, then they do not come to deserve the wage, as the moral balance is never tipped against them.

If Schmidtz’s non-compensatory view is right, however, then objects of desert can work quite differently than they do on the compensatory model. First, desert objects need not be what are needed to restore the moral balance. In fact, provided moral balance is restored, desert objects can be what put things out of balance in the first place (e.g. labour that is not yet paid for). Second, desert objects can be backwards looking. Once the employer pays the wage, they now deserve the historical fact that the labour was performed in the past.

These are very fundamental questions about desert at stake here, but let me give some reasons for preserving the compensatory model in the case of wages.

At minimum, there is a concern that Schmidtz’s proposal, applied to the case of wages, is ad hoc. Suppose, for example, that A is injured due to a malicious or negligent act by B. In that case, it would be typical to think that A deserves compensation (either from B or the state, let’s say). But then let’s suppose B provides compensation, so that A is ‘made whole’ and the moral balance is restored. We would not then say that A came to deserve to be harmed by B. Even if they now no longer harbour any hard feelings and they are better off, all things considered, it seems completely wrong to say they now deserve that they were harmed by B. The compensatory model seems clearly correct in cases where someone deserves compensation in virtue of their being wrongfully harmed. Thus, if Schmidtz’s conception is correct in the case of wages, then it would have to be that there is a different notion of desert at play, or that desert, and desert objects, work in fundamentally different ways in the case of wages and economic exchanges.

But it’s also not clear to me that Schmidtz’s view works in the case of wages either.Footnote ²⁰ Suppose an employee goes well above and beyond what is typically expected of them, and comes to deserve much more than their agreed upon wage. Suppose then that the employer, recognizing this some months later, compensates the employee and gives them what they deserve. If Schmidtz’s view is correct, then the employer will come to have deserved the supererogatory work from the employee. But this just seems to get the moral facts wrong. While the employer might ‘do justice’ to the employee and their supererogatory work, it seems wrong to say that the employer now deserves that supererogatory work.Footnote ²¹

The reason that this seems wrong, I think, is that when someone deserves something, we typically think that this gives us reason to think that they ought to get it, and perhaps we ought to help ensure that they get it. But even once the employer compensates the employee for their supererogatory work, it doesn’t make it the case that the employee ought to have done the supererogatory work – indeed, then it would not be supererogatory after all.

And this points to a final problem with Schmidtz’s view, at least from the standpoint of desert-based distributive justice – namely, it complicates the connection between desert and justice. Consider that, on the meritocratic view of distributive justice, when someone deserves something, we typically have good reasons of justice to ensure that they get it. But suppose Schmidtz’s view is correct. In that case, when labour is performed, but the wage has not yet been paid, justice demands that the wage is paid. We have reasons of justice, then, to ensure that this happens. But the paying of the wage also makes the employer deserving of the labour, so we can also say that before the wage is paid, we have reasons of justice to ensure that the employer becomes deserving of the labour that was performed. But then when the employer becomes deserving of the employee’s labour, that desert fact lacks the normative push to do anything. What is deserved is a certain historical fact – that the labour was performed. Thus, there is nothing to do, and ‘deserves’, here, is just a way of saying that justice has already been served, and that nothing needs to be done. Thus, I think Schmidtz’s notion of desert, even if it can be defended, is a poor fit with desert-based views of distributive justice.

Thus, I don’t think that appealing to Schmidtz’s non-compensatory conception of desert will help the meritocrat dodge the temporality problem. In that case, the temporality problem shows that the idea that wages are deserved requires rejecting the symmetry principle.

4.4. Is Labour Special?

Assuming that the symmetry principle leads to problematic consequences for the idea that wages are deserved, the question becomes whether there are good grounds for rejecting it. Rejecting the symmetry principle would require defending the idea that labour is special with respect to desert. And once again, the desert theorist might point to the familiar features of labour that purportedly ground the desert of wages – effort, sacrifice, contribution, etc. They then point out that what the employer does – simply paying the employee – lacks these features.

Just as with the similar objection to the generality principle, the objection must be careful not to simply assume that labour is special. The concern can be brought out, in this case, by pointing to the fact that proportionality would require that, for a wage to be deserved, the moral benefit of the wage should be proportional to the moral value of the labour. Thus, it cannot be argued that giving the wage is of less moral value than performing the labour, otherwise it just follows that a higher wage was deserved in the first place. And indeed, while giving someone money is a qualitatively different activity than performing most kinds of labour, in other contexts, such as when one donates to charity, giving money does very plausibly make one deserving of some benefit – perhaps some praise or recognition at minimum. We would need some non-question begging explanation, then, of why it is that giving money to someone does not ground desert of something of proportional moral value – i.e. the employee’s labour – in the employment context.

Furthermore, the concerns from section 3.4 still apply here. It cannot be argued that labour makes one deserving while paying a wage does not merely by pointing to the fact that labour has desert-grounding features that paying a wage lacks. If this difference cannot generalize beyond the employment context, then the argument begs the question, as it simply assumes that labour in the context of employment is special. Similar remarks apply here as in section 3.4, and so I will not go through them in detail again. I will note, however, that I think it will be particularly difficult for the contribution-based theorist to avoid the symmetry principle. After all, on that view, it is the employee’s contribution to the social good, through their labour, that makes them deserving of a wage. And while it’s true that the labour is most directly causally responsible for the contribution, the employer can rightly object that the employee never would have performed the labour if not for the wage that they would get as a result. Thus, the contribution is also clearly causally dependent on the payment of the wage, and so it is the agreement between employer and employee (and their following through with it) that leads to the social contribution.

4.5. Summarizing the Dialectic

Just as with the generality principle in the last section, then, the symmetry principle presents us with some options. One option is to reject the symmetry principle: perhaps only one party to an exchange can deserve what they get. In that case, perhaps an employee does deserve their wage, but the employer does not deserve the employee’s labour. This would avoid the unnatural consequences of the symmetry principle, and also the problems of the asymmetry of injustice and temporality problems that the symmetry principle generates when conjoined with the idea that wages can be deserved. The other possibility, if my arguments are good, is that the symmetry principle is vacuously true, because neither party to an exchange deserves what they get in the exchange. The primary pull in favour of the first option is just the common practice of thinking that employees can deserve a wage. In favour of the second option, however, it seems at least prima facie arbitrary to think that only one party to an exchange (which, in many cases, might be an equal exchange) deserves what they get.

5.1. Killmister’s Three Strands of Dignity

Of course, one wants to know, in virtue of what do these apparent injustices in wages constitute indignities? Here, I appeal to Killmister’s (2020) account of dignity in Contours of Dignity, on which dignity has ‘three strands’ – personal dignity, social dignity and status dignity – which correspond to three kinds of dignitarian norms. Our focus will be, primarily, on social dignity and status dignity. These (non-personal) dignitarian norms, generally, are sui generis norms which concern our standing and status in a community. Transgressing these norms typically involves shame or shaming, humiliation or a loss of respect (as examples will illustrate).

On Killmister’s account, social dignity is defined in reference to the dignitarian norms that apply to everyone in a given community.Footnote ²² There typically are social dignitarian norms, for instance, concerning one’s appearance in public and, for example, against public nudity. Thus, if one is stripped of their clothing in public, their social dignity is violated, regardless of what their personal dignitarian norms are (e.g. even if they personally embrace public nudity) (Killmister Reference Killmister2020: 31). Similarly, if someone is seen urinating in public, they are thereby degraded, and ‘invite the disdain’ of the community (29).

Social dignity norms are ubiquitous in social life. They might concern how and what we eat, how we speak and what we say, what kind of employment we have and whether we have employment, how we appear and behave in public, what our dwellings look like, how we are treated in a hospital, and so on. The severity of social dignity transgressions also varies greatly. For example, it is very often a social indignity to be intoxicated in public, but the kinds of dignity transgressions that are involved in public shaming, or certain kinds of torture, are far worse. The torturer, very often, is not merely seeking to cause pain or suffering, but to destroy the victim’s dignity, and to elicit great shame and humiliation (20).

Status dignity works similarly to social dignity, except that it is specific to certain social categories, and concerns the ways that members of those categories must be treated. Killmister’s example, here, is the way that it would be seen as being disrespectful to a judge, to fail to address them as ‘your honour’ in a courtroom. Showing the proper recognition respect to a judge qua judge requires that we speak to them with a certain formality that would be odd or even inappropriate in other contexts (33–34). Likewise, a female doctor might be assumed to be, and treated as, a nurse, when in fact she is a doctor (100). In such a case, her status dignity as a doctor is threatened.

Dignity has important conceptual ties to respect, shame and humiliation, and these ties help to explain the distinctive social harm that comes with dignity violations and frustrations. These ties also help to make clearer what makes a particular norm a dignitarian norm, rather than some other, non-dignitarian moral norm. Status dignity violations, for instance, reflect a lack of recognition respect for someone qua their membership in some category to which status dignity norms apply. When one’s status dignity is violated, they are likely to feel humiliation. If they fail to uphold the relevant status dignity norms on their own (i.e. without others violating one’s status dignity, or without them frustrating one’s ability to uphold those norms), feelings of shame will typically arise.

Social dignity, however, has important ties to appraisal respect:Footnote ²³

Just as with status dignity, we see that violations of social dignity will naturally elicit feelings of humiliation, while shame might be warranted if one fails to uphold social dignitarian norms without anyone’s interference.

Violations of dignity, then, constitute social harms, the severity of which will vary according to how significant the violation or frustration is. Such harms also, plausibly, very often constitute injustices, even if there is no additional physical or psychological harm.Footnote ²⁴ Someone who is publicly humiliated, for instance, and has their social dignity violated, is harmed and treated unjustly, even if (for whatever reason) they lack the capacity to recognize the humiliation and the harm (Killmister Reference Killmister2020: 31).

Now, as I noted above, Killmister’s account of dignity is a social account, in that dignity is understood in terms of the existence of contingent social norms that obtain in a given community. Having dignity, on this view, is not grounded in having some natural property or capacity (rationality or agency, for example). For our purposes, there are two important consequences of the social nature of dignity.

First, as social norms, dignitarian norms are contingent. Thus, we can easily imagine acts that would be violations of social dignity in our communities, but not in other possible (or actual) communities – e.g. public nudity in a nudist community. This feature of Killmister’s account of dignity, I will argue, makes the dignity-based account better positioned to explain the apparent contingency and asymmetry of injustice in wages than the desert-based account.

Second, dignitarian norms are, in some sense, socially basic, in that their obtaining does not depend on their being justified by more basic principles. Thus, many dignitarian norms may very well be morally unjustified – we might be better off if, for instance, it wasn’t an indignity to need help to use a toilet, so that those who need such assistance would not experience this unnecessary social harm (and its harmful psychological consequences). Thus, when one’s dignity is violated, and they are victim to this kind of social harm, the full explanation of the harm is simply that a relevant dignitarian norm obtains, and that norm is violated. So to ask, ‘but why should this constitute a social harm?’ is not to challenge the fact that it is such a harm, but to challenge whether the norm that is implicated in the harm should obtain.

6.1. Illustrating the Dignity-based Account: Pay Gaps and Glass Ceilings

Let me now give an illustrative application of the dignity-based view to a case of injustice in wages, before arguing that the dignity-based view offers a better explanation of injustice than the desert-based view.

That one’s income would have important connections to dignity in many communities is, of course, quite obvious. In general, income and wealth tend to be sources of social esteem, and having more wealth, or being paid more than one’s peers, will tend to lift one up in the eyes of the community.Footnote ²⁵ Thus, there is a general connection between income and social dignity. But being paid a certain wage might also very often be part of what it is to be acknowledged as occupying a particular social category. In that case, we can see that one’s status dignity will often be either upheld or threatened by one’s wage.

Suppose, for instance, there is a gender pay gap among the engineers at some firm. The female engineers get paid considerably less than their male counterparts for similar work. Furthermore, they have a much more difficult time getting raises and promotions relative to their male peers.

There are at least two distinct dignity-related concerns here. First, since earning more is a way to elevate one’s social status, and to be afforded greater appraisal respect, the female engineers’ social dignity is frustrated:

An individual has her social dignity frustrated if she is prevented from upholding social dignitarian norms, thereby being blocked from doing or being something her community takes to be ennobling. To frustrate someone’s social dignity thus denies her a mechanism by which she could enhance the esteem in which her community holds her (Killmister Reference Killmister2020: 61–62).

If women, for example, are systematically blocked or discouraged from getting raises or promotions, or occupying high-paying positions, then the normal pathways to attaining greater social esteem are systematically blocked. Thus, certain kinds of pay gaps and ‘glass ceilings’ will constitute injustices at least in part because they threaten dignity.Footnote ²⁶

The second dignity-related concern is that the status dignity of the female engineers is being threatened. Part of occupying a certain position at a firm very often involves receiving a certain kind of salary. So the fact that the female engineers are paid less might signal that their employer doesn’t consider them ‘real engineers’, for instance, and their colleagues are invited to do the same.Footnote ²⁷ In such a case, their ability to occupy the social role of engineer at the firm is threatened. And if such a pay gap pervades a society, then it threatens the status dignity of women, generally, with respect to the relevant positions.

6.2. Advantages of the Dignity-based Approach

Of course, desert-based accounts will count these kinds of pay-gaps as injustices as well. On an account where one deserves a wage on the basis of their productive contributions, if there is a difference in pay merely on the basis of gender, the lesser pay will not be deserved, and the pay gap is unjust. But I will now argue that the dignity-based account has certain advantages over the desert-based account, and it avoids the problems that I raised for the desert-based account in earlier sections. In doing so, I will also elaborate on the areas where the two accounts overlap, and where they diverge.

The first advantage of the dignity-based approach is that it helps make sense of the fact that injustices in wages very often elicit feelings of humiliation and shame, and also demands for respect. If a female employee is paid less than her male colleagues for the same work, for instance, it would be entirely natural for her to feel disrespected and humiliated as a consequence. When workers go on strike for higher wages or better conditions, for instance, they often frame their demands in terms of respect: respect very often seems to require that the wages and conditions be better. Dignity violations and frustrations, paradigmatically, elicit these kinds of feelings and responses. To be clear, the idea is not that there is injustice whenever someone feels shame or is humiliated because of their pay. The point is that feelings of shame and humiliation are sources of defeasible evidence that someone’s dignity is implicated, and a dignified wage is a necessary but not sufficient condition for a just wage.

Mere failure to get what one deserves does not so clearly produce these kinds of responses. It is a commonplace in the desert literature, for instance, that the winner of a competition is not necessarily the most deserving of victory. They might have benefited from an unusual degree of luck in the competition, for example, while the best competitor lost due to very bad luck. When the most deserving does not win and so does not get the deserved prize, they may be frustrated, curse their bad luck, and so on, but shame and humiliation are not so fitting. Even in the case where their loss is due to bad refereeing or judging, they may rightly feel like they’ve been treated unfairly, but it would still seem unusual and unwarranted for them to feel shame or humiliation.

Second, the dignity-based explanation of injustice avoids the asymmetry problems that, I argued, the desert-based explanations face. On the desert-based explanations, for example, it was unclear why it is far less natural (or maybe even incorrect) to talk about the labour that an employer deserves from their employee. If one side of an exchange can deserve what they get in an exchange, so should the other, or so it would seem. Why, then, is it not an injustice when the employer overpays their employee, for example? The desert-based account would seem to require, counterintuitively, that this is an injustice.

On the dignity-based approach, the answer is simple: there is no dignitarian norm which requires that, for the sake of the dignity of the employer, an employee must not be overpaid. And this should not be terribly surprising. Again, in a society where the appraisal respect one commands is intimately tied up with how much wealth they have, an employer will typically be in a better position than their employees, and so their wealth, and thus their social dignity, are often much more secure.

As Killmister (Reference Killmister2020: 96) notes, those in ‘precarious social positions … have a much stronger interest in avoiding social dignity frustrations. What few avenues remain open to these individuals to be raised in their communities’ eyes are deeply precious, and should be guarded with care.’ Workers, of course, are very often in this kind of precarious position. For many, employment, itself, constitutes a social indignity. Not only does one often not have much of a choice when it comes to which sort of job that one occupies, but they also experience a lack of autonomy and powerlessness in the workplace.Footnote ²⁸ Then, of course, some kinds of work are more dignified than others, and bring with them more or less appraisal respect. Despite all of these ways in which workers’ social dignity can be frustrated, having such employment is more dignified than having none at all. Much more can be said about the relationship between work and dignity, but the point is just that given the precarious state of the social dignity of many employees, it should be no surprise that there are asymmetries between employers and employees when it comes to our intuitions about justice in wages. Thus, the dignity-based explanation can explain the asymmetry, while the desert-based explanation seems to predict symmetry.

Indeed, a third advantage of the dignity approach is just that in cases where a worker’s dignity is not threatened, it is no injustice when they get less than they deserve. Just as it does not seem to be unjust when one buys overpriced garden seeds when there are plenty of more reasonably priced seeds available, it does not seem to be an injustice if, say, an independently wealthy employee chooses to work for much less than they deserve. On the desert-based view, this is a clear injustice. But the dignity-based view will be sensitive to the fact that the employee already has secured considerable social esteem. Furthermore, that they don’t need the income from the job (let’s suppose) shows that, contrary to what is often the case, the employee has considerable autonomy in accepting the job, in a way that many others might not because of their financial precarity. Insofar as this kind of autonomy is illustrative of an absence of powerlessness, it is indicative of someone whose social dignity is considerably more secure than in a typical case where an employee is underpaid. Thus, their low salary is no threat to their dignity, and we do not intuitively register it as an injustice.

Appeals to dignity might also justify policies which appeals to desert, by themselves, might not. For instance, consider a case where employees are paid exactly what they deserve, in accordance with their productive contribution, say. When an employee sustains an injury, and cannot work for a month, they are paid nothing for that month. After all, this is what they deserve for that month. It would not be unreasonable for the employee, in that situation, to feel disrespected, and to feel like their status as an important part of the firm is being threatened. Recognizing this, the firm might, instead, have some sort of paid leave policy to prevent the kinds of indignities that might arise in those cases.Footnote ²⁹

The foregoing examples show that the two approaches are not extensionally equivalent with respect to injustice in wages. There is, nevertheless, a good deal of overlap. In fact, it is possible that – as a contingent fact of the matter – certain dignitarian norms might mirror the kinds of considerations that are central to desert. For instance, as I will discuss below, it could be that in some communities or firms, dignity requires that people be paid the same wage for the same work. Indeed, I think this is quite plausibly sometimes the case.

But even in these cases, it’s worth emphasizing the ways in which the explanations differ. The desert-based explanation of injustice is more metaphysically simple and rigid in comparison to the dignity-based approach. The injustice obtains simply in virtue of a mismatch between the value of an employee’s productive contribution (or sacrifice, or effort, etc.) and their wage. Since theories of deserving wages specify desert bases for deserving wages – i.e. they give grounding explanations for what it is to deserve a wage – the relation between the desert base and deserving a wage holds of metaphysical necessity.

Injustice on the dignity-based approach, by contrast, appeals to certain contingent norms, their relations to certain social categories, facts about the employee, their work and their wage. Thus, there should be no concern that the dignity-based approach is giving the same explanation as the desert-based approach, just under a different name – even in the case where dignity requires that people be paid the same for the same contribution, for example.Footnote ³⁰

The contingency of the dignity-based explanation also allows for greater flexibility than the desert-based approach, and makes it so that – even where the two views are extensionally equivalent – they have very different modal profiles. This flexibility, I argue, constitutes a fourth and final advantage of the dignity-based approach.

We could easily imagine a community, for instance, where receiving a high wage constituted an indignity. Perhaps what is most dignified in such a society is sacrificing oneself for the sake of others. It’s not clear that there’s an injustice in getting a low wage – less than one deserves – in such a community (assuming they have other viable options). Or consider a post-scarcity world where human labour is no longer needed to sustain a good standard of living for everyone. In such circumstances, one’s ability to sustain themselves and enjoy the goods of society would no longer need to be tied to employment. The connection between employment and dignity, in that case, might be severed. And if so, it would explain why, in such circumstances, it might not be any kind of injustice to hire a friend or family member over a more qualified applicant, for example. Instances of taking on employment for a wage that is less than would typically be thought is deserved, might also be common, and would not obviously constitute injustices (barring concerns about exploitation).

The desert-based explanations of injustice cannot easily accommodate these intuitions. The most-qualified applicant deserves the job, and the wage one deserves is fixed by one’s productivity (or sacrifice, or effort, etc.). None of this would be altered in a post-scarcity world, or in the society that values sacrifice more than wealth. Thus, the injustices would remain the same as in our society. But I think that this is intuitively incorrect, and fails to see that the ways in which we conceive of justice in employment are shaped by the contingent social and economic conditions that we find ourselves in, and which are subject to change.

6.3. Does Dignity Require Getting the Wage One Deserves?

Now, it is true that when people suffer injustice in wages, they very often reach for the language of desert (in addition to the language of respect). Indeed, one possibility is that sometimes – assuming there is a sense in which wages could be deserved – dignity requires that one get what they deserve. In that case, dignity-based explanations need not completely preclude the possibility of desert playing some role in the explanation as well.

However, there are features of this possibility that should be unpalatable to those who defend a desert-based view of distributive justice. First, dignity would be the more fundamental notion with a tighter connection to distributive justice – it would be an injustice that one fails to get what they deserve only because (in some cases) it is an indignity to fail to get what one deserves.

Second, and relatedly, the possibility preserves no necessary connection between getting a wage one deserves and distributive justice. Again, dignitarian norms are contingent, so even if dignity requires that one gets what they deserve in some cases, it need not require it in all cases (and indeed, in some social contexts, the connection between dignity and desert might be completely severed). Justice in wages, then, could not consist in getting the wage one deserves, since grounding relations hold as a matter of metaphysical necessity.

Third, if dignity sometimes requires that one gets the wage they deserve, then the notion of desert that is operative in the dignitarian norm must be the ordinary notion of desert, and not a theoretical notion. The reason is that the dignitarian norms emerge out of community practice, and so must be formulated in terms that community members can appreciate. This distinction is important, since, again, desert theorists typically (and rightly) reject certain portions of ordinary ‘desert’ discourse as being merely ‘sham desert’. Without making this methodological move, the notion of desert is too sprawling and unwieldy to make good use of in theories of justice, and so ordinary desert judgements can smack of arbitrariness from the theoretical perspective. Further, this approach would be necessary to dodge the problems generated by the generality and symmetry principles, as concerns of arbitrariness need not arise for ordinary notions.

Thus, while it is certainly possible – perhaps even likely – that in some cases, dignity requires getting the wage one deserves, I don’t think this possibility is particularly helpful to the meritocrat, who conceives of justice in wages as getting the wage that one deserves.

In conclusion, then, I think a dignity-based account of injustice in wages has considerable promise as a competitor to desert-based theories. Not only does it avoid the difficult conceptual puzzles that I raised for the desert-based approaches in sections 3 and 4, but it better explains our intuitions of injustice in certain cases, and correctly predicts the contingent and context-sensitive nature of those judgements. In that case, and pending answers to the puzzles generated by the generality and symmetry principles, I think there is good reason to doubt that wages are the kind of thing that can be deserved, and that injustice in wages is best understood in terms of desert.