Talents and Wages (Just Wages Series)
Andrew Lister’s post is the first in our four-part series on “Just Wages,” which will be running this month.
Gregory Mankiw argues that those who are more productive should get a higher income not only as an incentive, but because this income is rightfully theirs (295). In a competitive market, factors of production are paid the value of their marginal product (32), which is the change in output associated with adding an extra unit of that input. Firms hire workers of a given type up to point at which the revenue the firm gains from hiring an additional worker is equal to the cost of that worker. Thus, in competitive equilibrium “each person’s income reflects the value of what he contributed to society’s production of goods and services.” In this way, the theory of “just deserts” gives “a new normative interpretation” to the economic theory of competitive equilibrium (295).
Joe Heath responds that wages in a competitive market aren’t intrinsically fair. Wage-setting is not a unidirectional process from unequal skill and effort to unequal contribution to unequal wages. Where workers of a given skill-level are more abundant, they will be cheaper, and so hired for lower-value tasks, and so less productive even if equally skilled and diligent. The rationale for pricing labour according to supply and demand is that it directs people’s talents to where they can best be used, generating prosperity that can benefit everyone. As Heath says, “[t]he market has one job to do, and it does that job very well [allocating resources efficiently]. Producing ‘just’ wages, however, is not that job” (31).
This critique of the just deserts rationale for capitalism has a long history among proponents of private property and free markets. Frank Knight, Friedrich Hayek, and Milton Friedman all emphasized the importance of limited information and luck in the operation of markets, generating inequalities that don’t reflect differences of merit or desert. Heath consciously drops one of the main forms of luck they identified, however: the genetic lottery. Here’s Milton Friedman, from Capitalism and Freedom:
Most differences in status or position can be regarded as the product of chance at a far enough remove. The man who is hard working and thrifty is to be regarded as ‘deserving’; yet these qualities owe much to the genes he was fortunate (or unfortunate?) enough to inherit (165-66).
On this point, Friedman followed Hayek and Knight. The inheritance of human capital is just as arbitrary a basis for distributing the social product as is the inheritance of physical capital. Robert Solow took this line against Mankiw.
You do not “deserve” that part of your income that comes from your parents’ wealth or connections or, for that matter, their DNA. You may be born just plain gorgeous or smart or tall, and those characteristics add to the market value of your marginal product, but not to your just deserts.
This rationale for redistribution poses a threat to self-respect, however, if interpreted as compensation for inferiority. Elizabeth Anderson argues that it would be stigmatizing for people to think that lack of talent is the ground upon which citizens can claim assistance (306). A better strategy, she suggests, is to “question the very idea that inferior native endowments have much to do with observed income inequalities in capitalist economies” (325). Heath adopts the same approach. Symphony musicians are obviously talented, but they don’t earn a lot (20). Inter-industry wage differentials within the same occupational category can’t be explained by differences in innate ability (3). Superstars like Lionel Messi obviously do earn high wages, but that’s in part because they have market power (18), so the markets for their talents aren’t truly competitive. Heath concludes that the question of innate ability is largely “orthogonal” to the debate over just wages.
I have doubts about the evidence Heath offers for his orthogonality thesis. And I worry that the avoidance strategy will undermine egalitarian conceptions of social justice, such as that of John Rawls.
The example of symphony musicians shows that markets reward scarce ability to produce goods and services people with money want to purchase, not ability in general. Productive potential is not equally distributed however. Developed productive ability could be more equally distributed if conditions of development in childhood were more equal. But it would still take special talents to become a brain surgeon, raising a question about the fairness of higher rewards for such roles, beyond what would be justified by heavier burdens.
The evidence on wage-differentials across industries shows that secretaries and managers in high-wage industries aren’t systematically better than secretaries and managers in low-wage industries. The explanation for wage differences within occupational groups must lie elsewhere, e.g. in norms of fairness or reciprocity within firms (as discussed in Peter Dietsch’s contribution). Those with greater talents will tend to earn more within a given occupation, however, and will be attracted to occupations with higher ratios of rewards to burdens.
My broader worry about Heath’s avoidance strategy is that it obscures the basis of egalitarianism. Knight, Hayek and Friedman were of course not egalitarians. In fact, they used the moral arbitrariness of the distribution of natural talent to accuse egalitarians of inconsistency. If the inheritance of ordinary capital is so unfair, why don’t you object to the inheritance of human capital too? As Ben Jackson explains, however, many on the left did object, not to the inheritance of ability as such, but to the fact that in a market economy, those born with scarce talents are able to command higher wages even without greater effort or diligence. Knight, Hayek and Friedman took it for granted that allowing income to be distributed according to one’s inheritance of property or talent was justified all things considered, since it contributed to aggregate prosperity (by incentivizing saving and the development of ability). Rawls argued that wasn’t enough. Morally arbitrary inequalities in shares of the products of collective labour are justified only if they benefit the worse off.
If markets don’t reward talent, then it may seem that all justice requires is protection of basic liberties and equality of opportunity, without any need for inequalities to benefit the worse off. However, even if we had eliminated unequal conditions of development during childhood, and even if we had perfected achievement-based selection to positions, it would still be unjust for positions with equal burdens to be rewarded unequally, unless such inequalities benefit the worse off. Fair equality of opportunity is sufficient for distributive justice only where there is no significant variation in the innate bases of productive ability. Denying the existence of such differences risks camouflaging the fact that inequalities between positions need to benefit the worse off, and would still need to do so even if conditions of development in childhood and adolescence were more equal.
Andrew Lister is an Associate Professor in the Department of Political Studies at Queen’s University in Kingston, Ontario. Before coming to Queen’s he was an FQRSC postdoctoral fellow at the Centre de recherche en éthique de l’Université de Montréal. He works on topics related to democracy and distributive justice, such as public reason, and more recently reciprocity.
What about the two ways in which, according Eric Nelson, Rawlsians want to have it both ways?
1) Desert should have no role when it comes to distributive justice, but it should have a role when it comes to retributive justice.
2) We don’t deserve the fruits of our labour since, when it comes to production, *we* aren’t responsible for our choices as they’re overly dependent on things we’ve inherited (e.g. talents). But our capacity to choose responsibly is one of the main reasons why liberalism affirms the respect for the individual.
Thanks, Charles, for these questions. I’m going to respond to the first one first, and hopefully come back to the second one later tonight.
The claim that desert should have NO role when it comes to distributive justice certainly seems wrong. People deserve things based on their conduct and choices, and fair distribution can’t be insensitive to choice. If principles of justice apply primarily to the rules of institutions, however, then people *will* be entitled to things based on the choices they make within the rules. So, for example, if the rule is that people are paid according to how many hours they work, and I choose to work more than you, then I will earn more money than you, and this inequality won’t be unjust. It won’t even have to benefit you, assuming the rules in place are just.
What about the design of these rules, then; couldn’t desert be a criterion for assessing social institutions? We can’t use “reward according to overall moral desert,” because a person’s overall desert will depend on whether they responded appropriately to the rules in place, which will depend on whether the rules are just, and we’re trying to identify just rules. We might be able to identify particular elements of deservingness, however, and design institutions so that they promote or at least don’t undermine those dispositions, other things equal.
Some conceptions of deservingness might be too controversial or in appropriately perfectionist for inclusion in a conception of social justice. It would be wrong, for example, to design rules of property and contract to promote industriousness for its own sake, or to discourage materialism, e.g. by refusing to enforce agreements – a suggestion Socrates makes in the Republic, at one point (556b; people should lend money at their own risk). Nonetheless, there might be dispositions that all citizens ought to have, dispositions that we think are essential to relating to other citizens as equals, and which we think institutions ought to promote, or at least not undermine (other things equal), for example the willingness to repay favors and the disinclination to free ride.
If we set aside incentives, the plausibility of the maxim of rewarding contribution derives from the idea of reciprocity. Generally speaking, reciprocity is a non-instrumental tendency to respond in kind to the conduct of others – for example, a duty to return certain kinds of benefits. One good turn deserves another, we say. We might think it good, other things equal, if social institutions promote or at least do not undermine this tendency, especially if we think of justice *as* reciprocity, as Rawls sometimes said. So I don’t accept (and I don’t think Rawls showed) that desert is irrelevant to the design of social institutions.
I do think Rawls was right, though, to follow Knight in thinking that inheritance of scarce productive abilities is not a basis of desert (though it may benefit everyone to permit reward to track talent, as a side effect of incentivizing development and application of abilities).
Here’s my response to your second question, which I see will now cross your response. I’m too slow!
I agree that the capacity to choose responsibly (based on reasons one is prepared to defend to others) is one of the main reasons for respecting individuals. Once past a threshold of autonomy, individuals’ choices merit respect even if wrongheaded, unless those choices need to be regulated for the sake of justice. A just society would provide all individuals with a real and sufficiently equal opportunity for autonomy. In such a society, it would not be unjust to hold individuals responsible for the outcomes associated with their choices. In actual societies, I think it can be, if opportunities for autonomy are sufficiently unequal and sufficiently bad at the bottom end. The fact that character depends family and social circumstances for which one can claim no credit cannot block desert in general, but in a society with seriously unequal conditions of development in childhood and adolescence it does tend to undermine justifications of inequality based on choice.
The claim that got Rawls into trouble was that “we do not deserve our place in the distribution of native endowments, any more than we deserve our initial starting place in society. That we deserve the superior character that enables us to make the effort to cultivate our abilities is also problematic; for such character depends in good part upon fortunate family and social circumstances in early life for which we can claim no credit. The notion of desert does not apply here” (89, Rev.) Commenting on this passage, Nozick said that it was risky for Rawls to try to “block” autonomous choice from influencing just distribution by “denigrating a person’s autonomy and prime responsibility for his actions.” Nozick interpreted Rawls to be saying that the notion of desert doesn’t apply unless all of the conditions making meritorious action possible are themselves deserved. I prefer to interpret Rawls’s “doesn’t apply here” comment in light of his earlier remark that the basic structure of society creates social positions into which people are born, positions that determine very different expectations of life, and that these inequalities in initial life chances “cannot possibly be justified by an appeal to the notions of merit or desert” (7).
Joe has some interesting posts on inequality in the capacity for prudent choice, for example in relation to the debate on nudging and libertarian paternalism:
http://induecourse.ca/waldron-sunstein-and-nudge-paternalism/
But I was just taking it for granted that, when we talk of desert in the context of distributive justice, we’re talking about desert tied to productivity, not to respecting just rules whatever they happen to be. If it were no more than the latter, then one could just as easily offer as an example the rule that people should be paid according to how many hours they work and mean by this that the more hours they work, the less they deserve to be paid.
So it seems to me that you haven’t really dealt with my first point, other than to repeat a part of it at the close of your comment. What I’m wondering is: How can Rawls assert both (i) that we should dismiss desert as a principle of distributive justice, since the fact some people act and decide in ways that are more productive than others is arbitrary from a moral point of view; and (ii) that we should affirm desert as a principle of retributive justice, since some actions and decisions by people should be punished.
(By the way, I think Rawls’ inheritance of (i) goes back much further than Frank Knight. I think he got it from Paul. Yes, that Paul.)
” the more hours they work, the less they deserve to be paid” If one of the main purposes of having rules about property and contract is to promote general prosperity, then that would be a bad rule. Also, the options it would present to people would be inconsistent with reciprocity; either one sacrifices to benefit others, or one benefits from the sacrifices of others without benefitting them. I don’t accept that desert is irrelevant in assessing social rules. But I also don’t think one can say that “the fact that some people act and decide in ways that are more productive than others is arbitrary from a moral point of view.” Choice within (approximately) just rules is not morally arbitrary, but generates legitimate entitlements.
The disanalogy Rawls drew between distributive and retributive justice had to do with whether the rules in question serve to specify and enforce duties that exist independently, or whether the rules are simply a social practice designed to achieve other purposes. So the crime of robbery presupposes the institution of property, whose purposes are to promote prosperity, and make possible individual autonomy (at least as far as personal property is concerned). But laws about assault aim to prevent conduct that is wrongful anyway, independent of the rules. That’s controversial, and probably too stark a contrast, because one can certainly imagine wrongs involving possessions in a state of nature. But I don’t think that distinction commits us to believing in desert as a principle of retributive justice. One could think that assault should be punished purely in order to deter such conduct. But I would assume punishment has an expressive dimension as well, and I think that would be bring in a notion of desert.
(I’m always thinking of desert as a particular kind of reason that someone should get something or be treated in a particular way; I don’t use it to refer to what someone should get all things considered. I’m defining desert in the way Feinberg did, I think: a noninstrumental relation of fittingness between a characteristic or activity that is substantially under an individual’s control, and a response that is called for on the part of others. The basic things people deserve are ‘responsive attitudes’, with modes of treatment being deserved only derivatively as expressions of these attitudes. So desert is fundamentally about relating to others in appropriate ways, assuming attitudes towards others are constitutive in part of relationships.)
I see that you do indeed conceive of desert as being about no more than having the appropriate “responsive attitude,” as you put it, e.g. respecting a just principle, whatever it happens to be. This would make it relevant to being a good or poor chooser in general, as discussed in the blog post by Heath that you cite. But that’s not how Rawls conceives of it. Rather, within the context of distributive justice, desert is one possible distributive principle among others. So, as Walzer emphasizes in Spheres of Justice, for example, it should be distinguished from, say, need, free exchange, or democracy. Because all distributive principles require that we exhibit the responsive attitude appropriate to them, I don’t think it’s helpful to conflate this with desert.
The responsive attitudes would be attitudes we have towards each other, primarily, rather than towards principles. There is a question about how ideas of desert that are appropriate in (or even constitutive of) interpersonal relations apply to the design of institutions, which I don’t think I’m fully clear about. Unfortunately Rawls didn’t define “desert” (though he did cite Feinberg’s “Doing and Deserving.”) When he discusses “the tendency for common sense to suppose that income and wealth… should be distributed according to moral desert,” Rawls focuses on the idea of reward by contribution as measured by marginal product, but that’s not the only possible basis of desert. That’s one of the problems with §48 of Theory. I summarize some of the other problems with §48 at p.57 of my PPE paper “Markets, desert, and reciprocity.”
Thanks. I’ll check it out!
Thank you for this, Andrew. I’m not sure about your claim that “Heath’s avoidance strategy […] obscures the basis of egalitarianism”. Even his understanding of wages as the result of supply and demand makes it clear that no one deserves his/her full wage and that there is room for taxation (to the extent that it’s not counter-productive). As I understand him, he just qualifies the claim that markets reward talent by pointing out that, more precisely, they reward demanded and scarce talents.
Hi Pierre-Etienne! Joe does make that point, which I agree with. People can develop abilities that are quite impressive, but if there isn’t much demand for the goods or services in question, the exercise of these abilities won’t be highly rewarded. Rawls’s main reference to Knight was on just this point. Reward by contribution as measured by marginal product depends on supply and demand, but “surely a person’s moral worth does not vary according to how many offer similar skills, or happen to want what he can produce.” Rawls cited Knight pp.54-7 of the book version of Knight’s article. Knight argued that productive contribution was not an “ethical measure of desert” because “the ownership of personal or material productive capacity is based upon a complex mixture of inheritance, luck, and effort, probably in that order of importance.” Knight thought that most people would agree that “inherited capacity represents an obligation to the world rather than a claim upon it.” It was a fallacy to draw an ethical distinction between income from labour and income from other factors, because “the capacity to labour productively derives from the same three sources as property ownership, namely, inheritance, luck and effort of acquisition, and with no obvious general difference from the case of property in their relative importance.” The value any service or product depends on demand, Knight continued, but “it is hard to see that… possession of the capacity to furnish services with are in demand, rather than other capacities, constitutes an ethical claim to a superior share of the social dividend, except to the extent that the capacity is itself the product of conscientious effort.” All this from 54-7, or 596-600 from the article version (The Quarterly Journal of Economics, Vol. 37, No. 4 (Aug., 1923)).
In these passages, I see two main ideas. One is that reward varies with supply and demand in ways that people can’t anticipate, introducing a strong element of luck into the allocation of reward for labour. We could try to reduce the influence of luck, but that would interfere with the price system’s main role, which is to aggregate dispersed information and send useful signals about what skills and resources ought to be developed and deployed where, in response to changing conditions.
The other is that people are not born with equal productive potential, and that a greater inheritance of scarce productive potential is not by itself a basis of desert of greater rewards. If people *were* born with equal productive potential, and conditions of development in childhood and adolescence were sufficiently equal, then it would not be necessary for inequalities between social positions to raise the lowest position. It’s this second point that I’m insisting on, and which I think Joe de-emphasizes or sidesteps. My worry is that it suggests that we don’t need to satisfy the difference principle, or something like it.
I understand your worry. These quotes from Knight’s article, which I didn’t know, are very interesting!