In Rescuing Justice and Equality, G.A. Cohen critiques what he calls the ‘incentives argument’ for inequality (Cohen Reference Cohen2008: 27). According to this argument, inequalities in income are just if they are necessary to render the badly off in society better off than they would otherwise be, for example, by motivating individuals to work harder than they otherwise would, or pursue lines of work that are particularly productive (Cohen Reference Cohen2008: 28). While, as Cohen points out, this argument is often made in the public sphere to justify a lower level of taxation than might otherwise be adopted, John Rawls also appeals to it to justify the difference principle (Rawls Reference Rawls1999a: 67–9, 130–31).
Cohen rejects the incentives argument on the grounds that people cannot employ it to justify their acceptance of incentive inequalities to their fellow citizens (Cohen Reference Cohen2008: 32). They cannot do so since it is within their control to decide how hard to work, or which occupation to work at. As Cohen puts it, people cannot claim,
in self-justification . . . that their high rewards are necessary to enhance the position of the worst off, since, in the standard case, it is they themselves who make those rewards necessary, through their own unwillingness to work for ordinary rewards as productively as they do for exceptionally high ones. (Cohen Reference Cohen2008: 122)Footnote 1
Cohen concludes that Rawls cannot appeal to the incentives argument to justify his favored understanding of the difference principle, which Cohen calls the ‘lax reading’ (Cohen Reference Cohen2008: 68–9). According to this reading, incentive inequalities are permissible if and only if they are necessary to improve the expectations of the worst off given the preferences and intentions people happen to have (Cohen Reference Cohen2008: 68–69). Instead, Cohen argues that Rawls is committed to the ‘strict reading’ of the difference principle, according to which inequalities are permissible ‘only when they are strictly, necessary, necessary, that is, apart from people's chosen intentions’ (Cohen Reference Cohen2008: 69). Cohen's critique therefore undermines the traditional interpretation of Rawls's difference principle as a promising compromise between the values of equality and efficiency, permitting inequalities that improve the expectations of the worst off. On Cohen's reading, Rawlsian justice condemns the use of incentive inequalities to bring about Pareto superior states of affairs, except where such inequalities are strictly necessary to do so.
In this paper, I argue that citizens of a Rawlsian society can justify their acceptance of a wide range of incentive inequalities to their fellow citizens. They can do so because they possess the right to freedom of occupational choice, and are permitted – as a matter of justice – to exercise this right by making occupational decisions on the basis of a wide range of values and preferences. I argue that this right is guaranteed by Rawls's first principle of justice, and entitles citizens – as matter of justice – to decide whether to accept or decline offers of employment free of state coercion. I argue further that the justification for this right, namely, that it promotes citizens’ full exercise of their capacity for a conception of the good, also entails that citizens are permitted to decide occupational questions on the basis of a wide range of preferences and values. Citizens can therefore justifiably accept those incentive inequalities that are necessary for them to perform a particular occupation, given their preferences and values. The correct interpretation of the difference principle is thus the lax reading, and incentive inequalities are a permissible means to realize Pareto superior states of affairs.Footnote 2
Because Cohen's critique is an internal one – his aim is to show that ‘when true to itself, Rawlsian justice condemns [incentive] inequalities’ (Cohen Reference Cohen2008: 68) – I will respond to it within the context of Rawls's theory of justice. I will therefore assume the truth of justice as fairness and will discuss Cohen's critique within the context of a just Rawlsian society, that is, a society in which the basic needs of citizens have been met, the principle of equal liberty has been satisfied, and fair equality of opportunity has been secured (Rawls Reference Rawls2005: 5–7). In part 1, I discuss freedom of occupational choice and argue that it ought to be understood as a liberty protected by the first principle of justice. I argue that the justification for this right also implies that citizens are permitted – as a matter of justice – to accept or decline employment offers on the basis of a wide range of preferences and values. In part 2, I consider three objections to my account. In part 3, I consider Cohen's claim in Rescuing Justice and Equality that freedom of occupational choice is consistent with both economic equality and the types of Pareto improvements permitted by the difference principle. I argue that if one adopts the Rawlsian understanding of freedom of occupational choice presented in part 1, Cohen's claim is false.
1. FREEDOM OF OCCUPATIONAL CHOICE
Rawls's first principle of justice states that:
Each person has an equal claim to a fully adequate scheme of equal basic rights and liberties, which scheme is compatible with the same scheme for all; and in this scheme the equal political liberties, and only those liberties, are to be guaranteed their fair value. (Rawls Reference Rawls2005: 5)
This principle governs the design of a society's political constitution, including the political process and structure of government, and the specification of limits to majority rule (Rawls Reference Rawls1999a: 53, 171–4; Reference Rawls2005: 336–8). Rawls's second principle of justice, which includes both fair equality of opportunity and the difference principle, governs the state's legislative enactment of social and economic policy (Rawls Reference Rawls1999a: 175). The first principle is lexically prior to the second such that the state may only enact social and economic policy that is consistent with it. Rawls's reasoning here is that the basic liberties protected by the first principle ‘have an absolute weight with respect to reasons of public good and of perfectionist values’ (Rawls Reference Rawls2005: 294). Infringements to the basic rights and liberties thus ‘cannot be justified, or compensated for, by greater social and economic advantages’ (Rawls Reference Rawls1999a: 53–4; Reference Rawls2005: 294–5).
This priority of the first principle of justice is important for our purposes. Because the first principle of justice is lexically prior to social and economic policy in the way that Rawls claims, citizens must presuppose it when deliberating about which social and economic policies to enact, and the nature of the claims they can legitimately make on each other. They must not therefore propose policies that violate this principle; and they may appeal to it, and its supporting arguments, as premises when justifying particular policies, or specifying the legitimate demands of the difference principle. If, as I shall argue below, freedom of occupational choice is best understood to be one of the liberties protected by the first principle of justice, citizens of a Rawlsian society can therefore appeal to it – and its supporting arguments – to justify their acceptance of incentive inequalities to their fellow citizens.
I carry out this argumentative strategy here in part 1. I first outline Rawls's case for giving the liberties priority over income and wealth and discuss Rawls's brief comments on freedom of occupational choice. Here, I make a textual case for the claim that Rawls's considered view is that this right is secured by the first principle of justice. I then argue on the basis of Rawls's account of the nature and priority of the liberties that freedom of occupational choice is also best understood as a liberty protected by the first principle of justice. I argue finally that this Rawlsian justification for understanding freedom of occupational choice as a basic liberty also implies a further claim, namely, that citizens are permitted – as a matter of justice – to exercise this right in a particular way, that is, to accept or decline offers of employment on the basis of a wide range of preferences and values.
1.1. The Priority of the Basic Liberties and the Place of Freedom of Occupational Choice
Rawls justifies the priority of the first principle by reference to a conception of the person as free and equal. According to this conception, persons possess two moral powers – a capacity for a conception of the good and a capacity for a sense of justice – and two corresponding highest-order interests in developing and exercising these powers (Rawls Reference Rawls and Freeman1999b: 312; Reference Rawls2005: 302). This conception of the person is modelled by the original position (Rawls Reference Rawls2005: 305). The nature of the parties models the capacity for a conception of the good; the parties therefore aim to choose those principles of justice that enable them to realize their highest-order interest in fully exercising their capacity to set, revise, and rationally pursue a plan of life, and to realize their higher-order interest in pursuing the determinate conception of the good that they happen to have (Rawls Reference Rawls and Freeman1999b: 312–13; Reference Rawls2005: 305). The veil of ignorance and the symmetrical situation of the parties model the capacity for a sense of justice (Rawls Reference Rawls2005: 305).
The parties to the original position choose the first principle of justice and assign it priority over the second because the basic rights and liberties provide the ‘background institutional conditions necessary for the development and the full and informed exercise of the two moral powers’ (Rawls Reference Rawls2005: 308). Rawls's idea is that citizens can only fully exercise their moral powers if they possess a sphere of action in which they, and not the state, are sovereign. Freedom of thought and the equal political liberties are necessary if citizens are to fully exercise their capacity for a sense of justice since they can only propose and apply the principles of justice if they are free to publicly evaluate government policy and participate in the political process (Rawls Reference Rawls2005: 335). Similarly, liberty of conscience is necessary if citizens are to fully exercise their capacity for a conception of the good (Rawls Reference Rawls2005: 335); only if they are free to decide whether to practice a certain religion or not is it up to them to decide which plan of life to set and pursue.
In contrast to the liberties, income and wealth are not necessary for realizing citizens’ highest-order interest in fully exercising their two moral powers. Instead, these social primary goods are necessary to realize citizens’ higher-order interest in pursuing a determinate conception of the good (Rawls Reference Rawls and Freeman1999b: 312–13; Reference Rawls2005: 308). As Rawls puts it, ‘income and wealth are needed to achieve directly or indirectly a wide range of ends’ (Rawls Reference Rawls2005: 308). Income and wealth are therefore important to the parties to the original position; however, they are less important than the basic liberties since citizens do not require them to realize their highest-order interest in fully exercising their two moral powers. The parties therefore have reason to choose the first principle of justice and assign it priority over the second (Rawls Reference Rawls2005: 305–10).
Rawls lists freedom of occupational choice – along with freedom of association – as a distinct social primary good (Rawls Reference Rawls2005: 308). He defines it as the negative right to accept or decline offers of employment, which may differ in terms of duties, location, hours of work and compensation (Rawls Reference Rawls1999a: 239–42; Reference Rawls and Kelly2003: 64; Reference Rawls2005: 228–32, 308–9, 335). It therefore demands that citizens be free of state coercion with respect to their choice of occupation, but does not entitle them to demand of either the state or private employers that they provide them with employment offers that are conducive to their plan of life. Rawls distinguishes this right from other economic liberties which are not secured by the two principles of justice, including the right to own the means of production, and freedom of contract – as understood by the doctrine of laissez-faire (Rawls Reference Rawls1999a: 54). For Rawls therefore, states may not force citizens to work at particular occupations; however, states are not prohibited by the two principles of justice from establishing public ownership of the means of production, or regulating the terms of employment that employers may offer to employees. For example, the two principles of justice do not forbid states from legislating minimum working conditions, imposing a 40-hour work week, or prohibiting the sale of certain types of services – e.g. sex. These questions are to be left to the political process.Footnote 3
Rawls is not entirely clear about which principle of justice freedom of occupational choice falls under (Rawls Reference Rawls2005: 308–9). At times, he claims that it falls under fair equality of opportunity (Rawls Reference Rawls1999a: 242–3; Reference Rawls2005: 76). At others, he understands it to be a liberty covered by the first principle of justice. For example, in Political Liberalism, he is clear that both freedom of occupational choice and freedom of association are ‘constitutional essentials’ covered by the first principle of justice rather than fair equality of opportunity (Rawls Reference Rawls2005: 228–32), and that both liberties are tightly connected to the more basic liberties of liberty of conscience and liberty of the person (Rawls Reference Rawls2005: 335). In Justice as Fairness: A Restatement, he puts the point even more explicitly:
The priority of liberty means that we cannot be forced to engage in work that is highly productive in terms of material goods. What kind of work people do, and how hard they do it, is up to them to decide in light of the various incentives society offers. (Rawls Reference Rawls and Kelly2003: 64, 157–8)
Despite these conflicting claims, I think that the textual evidence suggests that Rawls's considered view is that freedom of occupational choice is covered by the first principle of justice. The passages in which he takes this position are lengthier and more reflective on the matter (Rawls Reference Rawls and Kelly2003: 64, 157–8; Reference Rawls2005: 228–30, 335), whereas the passages in which he suggests that it is covered by fair equality of opportunity are much briefer (Rawls Reference Rawls1999a: 242–3; Reference Rawls2005: 76). More importantly, this interpretation fits better with Rawls's understanding of the nature and role of the two principles with respect to the different parts of the basic structure. The first principle of justice secures negative liberties such as liberty of conscience and freedom of thought, and applies to the constitution (Rawls Reference Rawls1999a: 53, 171–4; Reference Rawls2005: 336–8). Fair equality of opportunity, by contrast, governs the state's legislative enactment of social and economic policy (Rawls Reference Rawls1999a: 175). It requires that the state ensure equal chances of education and culture for all through the subsidization of private schools or the establishment of a public school system, and that the state regulate employers to ensure fair hiring practices (Rawls Reference Rawls1999a: 243). Fair equality of opportunity no doubt presupposes freedom of occupational choice in the sense that citizens may only have equal opportunities if they are free to pursue them. However, as the negative right to decide whether to accept or decline offers of employment, freedom of occupational choice fits better with Rawls's first principle of justice.
Of course, this textual evidence does not show that this is the best way to understand the place of freedom of occupational choice within Rawls's theory of justice. Nor is it clear how freedom of occupational choice enables citizens to justify their acceptance of incentive inequalities to their fellow citizens. To respond to Cohen, I must establish two further claims, namely, that (1) freedom of occupational choice is best understood as a liberty protected by the first principle of justice, given Rawls's account of the nature and priority of the liberties; and (2) citizens are permitted – as a matter of justice – to exercise this right by accepting or declining offers of employment on the basis of their preferences and values.
1.2. Freedom of Occupational Choice
Let me begin with (1). As I note above, the parties to the original position choose the basic liberties and assign them priority on the grounds that they are necessary if citizens are to fully exercise their two moral powers. Freedom of occupational choice is best understood as a basic liberty since it is necessary if citizens are to fully exercise their capacity for a conception of the good.
To see why this is so, consider first that the capacity for a conception of the good is the capacity to form, revise, and rationally pursue a plan of life (Rawls Reference Rawls2005: 19). Citizens fully exercise this capacity with respect to a sphere of their life by making decisions regarding it on the basis of their determinate conception of the good. Since exercising this capacity involves setting and pursuing that plan of life that one thinks best, fully exercising it involves governing the different aspects of one's life on the basis of one's values and preferences.
Consider second that if citizens are to fully exercise this capacity, they require rights against the state which secure for them a sphere of sovereignty in which they can govern their lives free of state coercion. Only if citizens possess such spheres are they free to make decisions regarding their lives on the basis of their values and preferences, rather than have their choices coercively directed by the state. Thus, liberty of conscience is necessary if citizens are to make decisions regarding their spiritual life free of state coercion. Similarly, freedom of the person is necessary if it is to be up to citizens to determine the purposes they will set and pursue with their body and mind, not the state. This point holds for occupational decisions. Citizens fully exercise their capacity for a conception of the good with respect to their occupational life by making occupational decisions on the basis of their values and preferences. Freedom of occupational choice, like liberty of conscience, is therefore necessary if citizens are to fully exercise this capacity, and so is best understood as a basic liberty.
Now, it does not follow from this claim that citizens are permitted, as a matter of justice, to make occupational decisions on the basis of their values and preferences. For this to be the case, it must be permissible for citizens to fully exercise their capacity for a conception of the good with respect to their occupational choices. This point is important for our purposes. Citizens cannot justify their acceptance of incentive inequalities to their fellow citizens simply by claiming that they possess freedom of occupational choice as a basic liberty; this liberty only prohibits the state from coercively directing their occupational choices. If citizens are to justify their acceptance of incentive inequalities to their fellow citizens, they must be permitted, as a matter of justice, to make occupational decisions on the basis of their preferences and values. That is, they must be permitted to exercise their right of freedom of occupational choice to realize the value that it secures, namely, the full exercise of their capacity for a conception of the good.
One might question whether Rawls is committed to this latter claim. Perhaps he thinks it unjust for the state to force citizens to take up particular occupations, but also thinks that justice places constraints on how citizens exercise their freedom of occupational choice. Cohen himself takes this position, arguing that citizens ought to be free of legal obligations with respect to their occupational choices, but not free of moral obligations (Cohen Reference Cohen2008: 183–96). Cohen argues further that citizens have a duty to exercise their freedom of occupational choice in accordance with an egalitarian ethos, which requires them to choose those occupations that promise the least advantaged the greatest benefit, and not require greater pay – i.e. incentive inequalities – to do so (Cohen Reference Cohen2008: 181–3). Importantly, this ethos does not entirely prohibit people from making occupational decisions on the basis of their values and preferences, since Cohen is also committed to a ‘reasonable personal prerogative’ (Cohen Reference Cohen2008: 1–11, 181).
Additionally, it seems reasonable to think that justice imposes constraints on how citizens may exercise their basic liberties. For example, although justice demands that citizens possess a legal right to freedom of speech, they are not permitted – as a matter of justice – to engage in hate speech. Why not think that Rawls is committed to a similar view with respect to freedom of occupational choice, that is, that justice demands that citizens be guaranteed freedom of occupational choice as a basic liberty, but that justice also imposes constraints on how they exercise it? More specifically, why not think that justice demands that they exercise their freedom of occupational choice in accordance with Cohen's egalitarian ethos?Footnote 4
In response, consider first that Rawls does not justify freedom of occupational choice or any of the other basic liberties simply on the grounds that state coercion is pro tanto wrong. Instead, as I argue above, he justifies these liberties on the grounds that they are necessary if citizens are to fully develop and exercise their two moral powers. As Michael Titelbaum puts it, although some political theorists justify citizens’ basic liberties by appeal to the need to prevent the state's coercive interference with aspects of citizens’ lives, Rawls does so in order to promote something, namely, the full development and exercise of citizens’ two moral powers (Titelbaum Reference Titelbaum2008: 309). The first principle of justice therefore does not only impose an obligation on the state to refrain from coercing its citizens, but also imposes an obligation on it to ensure that citizens possess a ‘fully adequate scheme of equal basic liberties’, that is, a scheme that allows ‘the adequate development and the full and informed exercise of both moral powers’ (Rawls Reference Rawls2005: 333). The rights protected by the first principle of justice, Rawls claims, are thus understood to have a ‘central range of application’ where this range is that sphere of action that citizens require for the ‘adequate development and full exercise of the two moral powers’ (Rawls Reference Rawls2005: 297).
Rawls's understanding of the basic liberties is thus different from Cohen’s, since for Cohen, the value of the freedom that is secured by the basic liberties ‘lies in the absence of coercion itself’ (Cohen Reference Cohen2008: 194–5). This point is important for our purposes. First, it explains why although for some theories of justice, including Cohen’s, the justification for a particular basic liberty need not have any implications for the permissible exercise of this liberty, Rawls's theory of justice is not one of them. Since Rawls justifies the basic liberties on the grounds that they are necessary if citizens are to fully exercise their two moral powers, it seems reasonable to conclude that Rawls thinks that it is permissible for citizens to employ their liberties to fully exercise their two moral powers. Second, it places a challenging argumentative burden on those who would deny that Rawls is committed to the claim that it is permissible for citizens to exercise their freedom of occupational choice so as to fully exercise their capacity for a conception of the good. They must provide an interpretation of Rawls according to which justice (1) demands that citizens possess the basic liberties in order that they may fully exercise their two moral powers; and (2) forbids citizens from employing their basic liberties to do so.
Consider second that Rawls can grant that there are constraints of justice regarding how citizens exercise their right to freedom of occupational choice without: (1) denying that it is permissible for citizens to decide occupational questions on the basis of a wide range of preferences and values; or (2) committing himself to Cohen's egalitarian ethos. Since Rawls is committed to the priority of the right over the good – the idea that principles of justice place limits on the conceptions of the good that are permissible (Rawls Reference Rawls2005: 176) – he might support the idea that citizens may not exercise their freedom of occupational choice to pursue racist conceptions of the good life, for example, by demanding greater pay as compensation for having to work with citizens of a different race. However, even if citizens may not make occupational decisions on the basis of impermissible conceptions of the good life – e.g. racist or sexist conceptions that deny the moral equality of persons – they may still make such decisions on the basis of a wide range of preferences and values, namely, permissible conceptions.
Moreover, these constraints that Rawls might endorse are far more limited than those required by Cohen's egalitarian ethos since this ethos prohibits people from deciding occupational questions solely on the basis of conceptions of the good life that are fully permissible in Rawls's sense. After all, Cohen's egalitarian ethos requires citizens to comply with the strict interpretation of the difference principle, according to which incentive inequalities are permissible only if they are strictly necessary to improve the expectations of the worse off, that is, irrespective of their preferences and values (Cohen Reference Cohen2008: 69–73). Even if we grant that Rawls would hold that people may not make occupational decisions on the basis of impermissible conceptions of the good life, he is still committed to a version of the lax interpretation of the difference principle, according to which incentive inequalities are permissible only if they are necessary to improve the expectations of the worse off, given the permissible preferences and values they happen to have. To see the distinction, consider Cohen's case of the doctor-gardener who is capable of both doctoring and gardening, prefers gardening at £20000 per year to doctoring at £20000 per year, but prefers doctoring at £50000 per year to both (Cohen Reference Cohen2008: 184). As Cohen notes, his egalitarian ethos requires the doctor-gardener to choose doctoring at £20000 per year (Cohen Reference Cohen2008: 184–5). By contrast, provided that the doctor-gardener does not prefer doctoring at £50000 per year to the alternatives for sexist, racist, or other reasons that deny the moral equality of persons, for Rawls, it is permissible for the doctor-gardener to choose doctoring at £50,000 per year.
One might object that I am supposing that there is far more daylight between Rawls's position and Cohen's position since Cohen is committed to a reasonable personal prerogative that permits citizens to make occupational decisions on the basis of their values and preferences, at least to some extent. Now, it is difficult to decide this question definitively since Cohen does not outline the precise scope of this prerogative. But, provided that Cohen's prerogative does not cover all preferences and values that are permissible in Rawls's sense – i.e. do not deny the moral equality of persons – there will be daylight between his and Rawls's positions on which incentive inequalities are permissible. Indeed, this difference is likely to be significant since Cohen emphasizes at a number of places that even taking his prerogative into consideration, justice permits very little income inequality (Cohen Reference Cohen2008: 61, 387–95). Cohen's egalitarian ethos, even when subject to a reasonable personal prerogative, places far greater constraints on the extent to which citizens may make occupational decisions on the basis of their preferences and values. Rawls can therefore grant that there are constraints of justice regarding citizens’ exercise of their right to freedom of occupational choice without thereby committing himself to Cohen's egalitarian ethos, or denying that it is permissible for citizens to decide occupational questions on the basis of a very wide range of preferences and values.
Finally, we can address the question of the permissibility of citizens making occupational decisions on the basis of their preferences and values more directly by considering Rawls's broader approach to questions of justice and asking what it implies here. Suppose that the parties to the original position turn their attention away from the basic structure of society and instead focus on the principles that should govern the way in which citizens exercise their basic liberties. In particular, with respect to the exercise of freedom of occupational choice, would the parties choose a principle that required citizens to exercise this liberty in accordance with Cohen's egalitarian ethos – subject to a reasonable personal prerogative? Or, would they choose a principle permitting them to make occupational choices solely on the basis of their permissible values and preferences?Footnote 5 I think it is clear that they would choose the latter. The parties are concerned to choose principles that permit them to fully exercise their capacity for a conception of the good. Since, as I note above, citizens fully exercise this capacity by making decisions regarding their life on the basis of their determinate conception of the good, they would opt for a principle of justice that permitted them to make their occupational choices solely on the basis of their permissible values and preferences, whatever they happen to be. Cohen's egalitarian ethos limits the extent to which they can realize this interest since it places restrictions on the extent to which they can make these decisions on the basis of their preferences and values, as opposed to considerations of what would be best for the least advantaged. The parties would only choose Cohen's egalitarian ethos if they privileged the second principle consideration of income over the first principle consideration of the full development and exercise of the two moral powers, thus reversing the priority ordering they attach to the two principles of justice as they apply to the basic structure. After all, Cohen's egalitarian ethos requires citizens to sacrifice the full exercise of their capacity for a conception of the good for the material benefit of the least advantaged.
Additionally, the right to freedom of occupational choice, as well as the permission to exercise it by making occupational decisions solely on the basis of one's permissible preferences and values, is particularly important for citizens given their highest-order interest in exercising their capacity for a conception of the good. Given the nature of labour market participation, citizens are likely to have strong preferences regarding the duties, hours of work, location and compensation specified by employment offers. If I prefer working outside with my hands, I have reason to pursue a position in construction or the forestry industry. If I love the ocean and hate big cities, I have reason to seek work in a small coastal town rather than a large city in the interior. If my projects are expensive, I have reason to seek employment that is highly paid. Depending on their preferences and values, individuals will also differ in terms of how they weigh all of these factors. If I value time with my friends and family but am also interested in pursuing an expensive project, I will only have reason take a position that involves long hours if I can secure a post-tax income that is high enough to fund my expensive project. If I cannot secure such a post-tax income, I should instead choose an occupation that leaves me with plenty of free time to spend with my friends and family. If I love big cities, I may only have reason to take a position in a small town if the position is ideally suited for me and promises high pay.
One might argue here that the parties to the original position would not select a principle that permits citizens to make occupational decisions on the basis of their preferences and values since such a principle is not necessary for the full exercise of their capacity for a conception of the good in the way that protections against state coercion are. Provided they possess freedom of occupational choice as a basic liberty, citizens are free to make their occupational choices on the basis of their values and preferences, even if they are wrong to do so.Footnote 6 But notice that the question before the parties concerns the principles of justice that should govern citizens’ exercise of their right to freedom of occupational choice. I have argued that the principle the parties would choose to govern citizens’ exercise of this right, given their highest order interest in fully exercising their capacity for a conception of the good, is one that permits citizens to fully exercise this capacity with respect to their occupational choices. This argument does not rely on the premise that this permission is necessary for citizens to fully exercise their capacity for a conception of the good, only that the parties would choose a principle of justice that entails it.
Furthermore, there is a sense in which this permission is necessary if citizens are to fully exercise their capacity for a conception of the good. A central feature of the original position is that the parties are choosing principles that will bind citizens’ behaviour (Rawls Reference Rawls1999a: 126). From the perspective of the parties therefore, the permission for citizens to make occupational decisions on the basis of their values and preferences is necessary if citizens are to fully exercise their capacity for a conception of the good since citizens will comply with the principles that are chosen.Footnote 7
To sum up our discussion thus far, I have argued that freedom of occupational choice deserves the status of a basic liberty because it is necessary if citizens are to fully exercise their capacity for a conception of the good. I have argued further that citizens are permitted, as a matter of justice, to exercise this right to realize the value it protects, namely, by making occupational decisions solely on the basis of their permissible preferences and values. This line of argument is important for our purposes. Citizens can appeal to it to justify their acceptance of a wide range of incentive inequalities to their fellow citizens. To see this, consider the following dialogue – first presented by Jan Narveson – which Cohen cites as a demonstration of his critique of the incentives argument:
Well-off: Look here, fellow citizen, I’ll work hard and make both you and me better off, provided I get a bigger share than you.
Worse-off: Well, that's rather good; but I thought you were agreeing that justice requires equality?
Well-off: Yes, but that's only as a benchmark, you see. To do still better, both of us, you understand, may require differential incentive payments to people like me.
Worse-off: Oh. Well, what makes them necessary?
Well-off: What makes them necessary is that I won't work as hard if I don't get more than you.
Worse-off: Well, why not?
Well-off: I dunno . . . I guess that's just the way I’m built.
Worse-off: Meaning, you don't really care all that much about justice, eh?
Well-off: Er, no, I guess not. (Narveson Reference Narveson1978: 287–8)
Armed with the line of argument I sketch above however, we can imagine a different exchange between Well-off and Worse-off:
Worse-off: I hear you took that consulting job . . .
Well-off: You heard right.
Worse-off: I also hear the pay is pretty good.
Well-off: It is, and really the big jump in salary was a necessary condition for me accepting the position.
Worse-off: What do you mean necessary? Surely you’re physically and psychologically capable of doing the job for the salary we ordinary folks can expect.
Well-off: Sure, but it was necessary given my preferences, and I’m permitted to make occupational choices solely on the basis of my preferences. Look, I really like spending time with my kids and it only makes sense to take on the extra hours and stress if the jump in pay is fairly large. I’ll have less time with them now, but I’ll also have the resources to do some pretty memorable things with them in the time we spend together.
The key difference between these dialogues is that in the latter Well-off is able to appeal to the line of argument I sketch above to justify – to her fellow citizens – her acceptance of an incentive inequality to perform the consulting position. Since her preferences and values provide a legitimate basis for making occupational decisions, she can justifiably refuse to take the consulting job without a pay raise. More broadly, because citizens are permitted to make occupational decisions solely on the basis of their permissible preferences and values, the difference principle should be understood on its lax reading: incentive inequalities are permissible if and only if they are necessary to improve the expectations of the worst off given the permissible preferences and intentions people happen to have.
Importantly, my account does not allow people to justify all incentive inequalities to their fellow citizens, or, for that matter, all manner of occupational decisions.Footnote 8 First, citizens can only justify those incentive inequalities that are in fact necessary to perform the job in question, given their preferences and values. Citizens cannot therefore justifiably accept or bargain for incentive inequalities that are greater than their reservation wage – i.e. the wage at which they are willing to take the job, given their preferences and the available alternatives. If I am offered a position at $50000/year, which also happens to be my reservation wage, I cannot justify the acceptance of a greater incentive inequality to my fellow citizens. I cannot say that I require $55000/year to perform the job, given my preferences and values, since I don't – I’m willing to perform the job for $50000/year. Similarly, if I receive offers for two positions between which I am indifferent, I cannot justifiably bargain for a post-tax income that is greater than my reservation wage. For example, suppose that I am a free agent point guard and receive offers to play for both the Los Angeles Lakers and the Los Angeles Clippers. If I am truly indifferent between these positions, I cannot justify the acceptance of an incentive inequality that is greater than my reservation wage to my fellow citizens. After all, I am willing, given my preferences and values, to play for either team at my reservation wage.
Second, my account also implies that in cases where citizens are indifferent between two possible positions, they should choose the one that is best for the least advantaged. Suppose I am indifferent between a position as a gardener at $20000/year, and as an NGO organizer at $20000/year, and that the latter position can be expected to raise the expectations of the least advantaged more than the former.Footnote 9 If I have no reason stemming from my preferences and values for preferring one to the other, the fact that the latter better promotes the expectations of the least advantaged is a reason in favour of choosing it. I cannot justify choosing the gardener position to the least advantaged since this choice seems to give their interests no weight in my decision.Footnote 10
Some Rawlsians might find these limitations of my account worrisome. After all, Rawls's theory of justice is often thought to permit citizens to do the best they can for themselves in the labour market. In response, consider first that these limitations arguably make Rawls's theory of justice more defensible. Many scholars have objected to the apparent inconsistency in Rawls's theory between the difference principle, which prohibits incentive inequalities that are not necessary to improve the expectations of the least advantaged, and the permission for private citizens to accept incentive inequalities that are not necessary to improve the expectations of the least advantaged (Narveson Reference Narveson1978: 284–91; Nagel Reference Nagel1991: 85–7; Cohen Reference Cohen2008: 121–4). My account partially resolves this inconsistency since it does not enable citizens to justify the acceptance of incentive inequalities greater than their reservation wage to their fellow citizens. Consider second that although my account does not provide citizens with the resources to justifiably accept incentive inequalities greater than their reservation wage, it does not claim that no such resources exist. It is therefore potentially consistent with accounts of Rawls's theory of justice that aim to defend the more ambitious claim that citizens can justify the acceptance of any incentive inequalities to their fellow citizens.
2. THREE OBJECTIONS
Cohen discusses freedom of occupational choice and its relation to incentive inequalities at length in Rescuing Justice and Equality. Before turning to this discussion, I consider three objections that one might raise against my argument in part 1. Because Cohen's critique of the incentives argument is supposed to be internal to Rawls's framework, the objections I consider are ones that can be reasonably said to be similarly internal. My aim in this paper, after all, is to refute Cohen's claim that ‘when true to itself, Rawlsian justice condemns [incentive] inequalities’ (Cohen Reference Cohen2008: 68).
For ease of expression, I shall use the abbreviation ‘RFOC’ (Rawlsian freedom of occupational choice) to refer to the two claims I argue for in part 1:
1. Freedom of occupational choice is a basic liberty.
2. Citizens are permitted, as a matter of justice, to accept or decline offers of employment on the basis of considerations stemming from their values and preferences.
2.1 RFOC: Different From Other Accounts?
First, one might question how novel my account is as a response to Cohen. Others have argued that there are limits to the demands that justice may make on people's occupational choices, and so have concluded that Rawls is correct to claim that incentive inequalities are permissible. For example, David Estlund argues that Rawlsians can defend the permissibility of incentive inequalities on the grounds that citizens of a just Rawlsian society possess certain personal prerogatives that impose limits on the demands of justice (Estlund Reference Estlund1998: 101–3, 112). These prerogatives include a prerogative to pursue self-interest, a prerogative to pursue the good of certain others, a prerogative to comply with one's moral requirements, and a prerogative to pursue morally significant purposes (Estlund Reference Estlund1998: 101–3, 112). Kok-Chor Tan argues similarly that people are permitted to make occupational choices on the basis of their values and preferences since the demands of justice must be balanced against the value of personal pursuits (Tan Reference Tan2004: 335–52). Tan therefore supports ‘institutional egalitarianism’ as a way of ‘balancing the demands of justice with the pursuit of personal ends’ (Tan Reference Tan2004: 334). On this view,
egalitarians are required as a matter of justice to be concerned with bringing about and maintaining an egalitarian basic structure; but they are not required as a matter of justice to be egalitarians in their interpersonal decisions and actions within the rules of the basic structure. (Tan Reference Tan2004: 334)
My response to Cohen is similar to these accounts since it too permits citizens to make occupational decisions on the basis of their values and preferences. However, it is also different in a number of ways. Furthermore, I think these differences show that my account offers a stronger defence of Rawls's position against Cohen's critique.
The most important difference between my response and those of Estlund and Tan concerns the ground of the permission for citizens to make occupational choices on the basis of their preferences and values. For Estlund and Tan, citizens possess this permission because there are limits to the demands of justice. On my account by contrast, justice demands that citizens be permitted to make occupational decisions on the basis of their preferences and values. On my account therefore, individuals can justify their acceptance of incentive inequalities to their fellow citizens by appealing to principles of justice; they need not appeal to the claim that there are limits on the demands of justice.
This difference provides two reasons to favour my account over those of Estlund and Tan, considered as responses to Cohen. First, a chief desideratum of a theory of justice is that it be able to adjudicate the claims that people make on each other, deciding which claims are legitimate, and which are illegitimate. This includes being able to distinguish between the private and the public spheres, namely, those spheres in which people are free to set and pursue their conception of the good life free of the claims of others, and those spheres in which they are subject to the claims of others. My account helps Rawls's theory of justice complete this task, clarifying that justice demands that citizens be permitted to make occupational decisions on the basis of their values and preferences. Estlund and Tan's accounts, by contrast, do not help Rawls's theory of justice complete this task. They do not do so since they draw the distinction between private and public spheres by imposing limits on the demands of justice, not by appealing directly to the central commitments of Rawls's theory of justice.
Second, because Estlund and Tan reply to Cohen's critique in this way, they leave him with a fairly strong line of response, namely, that Estlund and Tan have drawn the limits of justice too narrowly. For example, Cohen agrees with Estlund that the demands of justice must be limited by a personal prerogative (Cohen Reference Cohen2008: 181); he just disagrees with Estlund about the strength that this prerogative should have. In response to Estlund therefore, Cohen has argued that the prerogative Estlund proposes is too strong and that the incentive inequalities it allows are impermissible (Cohen Reference Cohen2008: 387–94). Cohen can respond to Tan's account in a similar fashion by arguing that Tan's institutional egalitarianism simply achieves the wrong balance between the value of personal pursuits and the demands of justice. Additionally, it is difficult to see how these disagreements can be resolved in a principled fashion. As Cohen himself puts it, the question of the degree of inequality permitted by a reasonable personal prerogative is a ‘difficult matter of judgment’ (Cohen Reference Cohen2008: 394). Cohen, Tan and Estlund simply seem to have differing intuitions about the appropriate balance between these competing values, and it is difficult to see how this difference can be resolved.
My account, by contrast, is not subject to this type of reply. It does not appeal to intuitive considerations regarding the appropriate balance between the demands of justice and the value of personal prerogatives. Instead, by showing how RFOC fits systematically within the broader framework of Rawls's theory of justice, I have shown that Rawls's theory of justice entails that a wide range of incentive inequalities are permissible. This is a particularly powerful line of response, given that Cohen's critique is intended to be internal to Rawls's framework.
A second difference between my account and Estlund's is that my account is grounded in a basic commitment of Rawls's theory of justice, namely, the idea that the correct principles of justice are those that are justifiable to persons concerned to fully develop and exercise their two moral powers. By appealing to this idea, I show that freedom of occupational choice is a basic liberty, and that citizens are permitted to make occupational decisions on the basis of their preferences and values. By contrast, Estlund responds to Cohen's critique by appealing to the idea of a personal prerogative that imposes limits on the demands of justice. This idea is foreign to Rawls's work, and so only provides a response to Cohen at the cost of substantially altering the view.Footnote 11 My account is therefore also superior to Estlund's as a Rawlsian response to Cohen.
2.2. Is RFOC Too Minimal?
One might object that RFOC has worrisome implications, for example, that it does not guarantee that people will be able to secure that occupation or post-tax income that is necessary to realize their conception of the good; or that it does not guarantee that those who do not possess marketable talents – relative to others – will have much choice about which occupation to pursue and so may be condemned to working menial or unfulfilling jobs. However, I don't think that these implications pose a problem for my account. First, even if both of these implications come to pass, it doesn't follow that citizens do not possess RFOC or that it has no value. The purpose of RFOC is not to ensure that people can realize their determinate conception of the good or ensure that they have an array of appealing employment offers. Instead, it is to ensure that it is up to them to decide, on the basis of their values and preferences, whether to accept or decline the offers that are made to them. Even in cases where citizens face limited options therefore, RFOC still has value since it entitles them, not the state, to make choices regarding their lives, and permits them to make these choices on the basis of their values and preferences, not the values and preferences of others.
Second, there is reason to think that RFOC should not grant citizens more than the permission to accept or decline offers of employment on the basis of their preferences and values. The fact that some won't be able to secure the position of employment or post-tax income that they want or that others will face limited options is an inevitable feature of reliance on the market to organize the economy. Because economic cooperation has the purpose of producing those goods and services that persons require to satisfy their preferences, the availability of occupations, and their accompanying salaries will always depend on the preferences that people have. If no one's conception of the good involves watching the films that I direct, I won't be able to realize my goal of being a successful film director and I won't be able to realize my expensive projects. Such an outcome might strike some as unjust since it seems to favour conceptions of the good that are market friendly (Van Parijs Reference Van Parijs1991).Footnote 12 However, in this case at least, the market ensures that no one has to make the ends of others their own (Dworkin Reference Dworkin2000: 66–71). A more substantive conception of RFOC, one that secured for individuals a certain array of employment offers, would require citizens to do so.
Third, a key feature of Rawls's conception of free persons is the idea that they possess the capacity to take responsibility for their ends, that is, to adjust them in light of the all-purpose means that they can reasonably expect in return for their participation in the system of economic cooperation (Rawls Reference Rawls2005: 34). Rawls's conception of justice thus includes a ‘social division of responsibility’:
society, the citizens as a collective body, accepts the responsibility for maintaining the equal basic liberties and fair equality of opportunity, and for providing a fair share of the other primary goods for everyone within this framework, while citizens (as individuals) and associations accept the responsibility for revising and adjusting their ends and aspirations in view of the all-purpose means they can expect, given their present and foreseeable situation. (Rawls Reference Rawls and Freeman1999c: 371)
Citizens are not therefore entitled to the means that are necessary to realize their determinate conceptions of the good. Instead, as free persons, they are responsible for revising their plans of life in light of the occupation and post-tax income that they can reasonably expect.
One might argue that this idea of the social division of responsibility provides an argument in favour of Cohen's egalitarian ethos as a requirement of justice. If revising one's conception of the good in light of the occupation and post-tax income one can reasonably expect is consistent with the full exercise of one's capacity for a conception of the good life, then why isn't compliance with Cohen's egalitarian ethos similarly consistent with the full exercise of one's capacity for a conception of the good? Why might citizens have a responsibility to revise their preferences and values in light of the all-purpose means they can reasonably expect, but not in light of the interests of the least advantaged in society?Footnote 13
In response, consider first that Rawls's idea of individual responsibility requires citizens to exercise their two moral powers in ways that are fully respectful of the claims of others to do the same. Citizens may not therefore expect their fellows to sacrifice their fair share of basic liberties or income so that they may satisfy some expensive set of preferences. However, although this idea of individual responsibility requires citizens to recognize the reasonable claims of their fellows, Rawls makes an important distinction in the types of claims that citizens may make on each other. Claims to the basic liberties take priority over claims to income and wealth such that citizens may not claim a greater share of income or wealth at the expense of a lesser set of basic liberties for their fellows; a basic liberty may only be limited ‘for the sake of one or more other basic liberties’ (Rawls Reference Rawls2005: 295). As I note in part 1, the basis for this distinction between types of claims is Rawls's distinction between citizens’ highest-order interest in fully exercising their capacity to set, revise and rationally pursue a plan of life, and their higher-order interest in pursuing the determinate conception of the good that they happen to have (Rawls Reference Rawls and Freeman1999b: 312–13; Reference Rawls2005: 308). Claims to the basic liberties take priority over claims to income and wealth since they advance the former interest (Rawls Reference Rawls and Freeman1999b: 312–13; Rawls Reference Rawls2005: 308); income and wealth advance the latter (Rawls Reference Rawls2005: 305). Thus, Rawls's first principle of justice is lexically prior to the second.
This distinction between the types of claims citizens may make on each other is important for our purposes. As I argue above, RFOC secures the highest-order interest of citizens in fully exercising their capacity for a conception of the good with respect to their occupational choices. Because all citizens possess this interest, individuals must temper the claims they make on their institutions and fellow citizens. For example, I cannot legitimately request that the state limit the freedom of occupational choice of those who might compete with me for a position that is conducive to my conception of the good; nor may I make a claim of justice on my fellow citizens to refrain from applying for it. More importantly, because the highest-order interests of citizens take precedence over their higher-order interests, citizens may not demand that their fellows limit the realization of their highest-order interests so that they may better realize their higher-order interests in realizing their determinate conception of the good life. Just as the lexical priority of Rawls's first principle of justice prohibits governments from limiting the basic liberties in order to realize greater economic growth, so too I may not demand of you that you limit your RFOC so that I might have a higher income. If you become a corporate lawyer instead of a stay-at-home-dad I may be better off materially; however, I may not demand that you consider my preferences to be binding on you when you decide what to do with your life.
This is why the full exercise of citizens’ capacity for a conception of the good is not consistent with Cohen's egalitarian ethos. Although Cohen endorses a personal prerogative that places limits on the demands of the egalitarian ethos, the ethos still requires citizens to make occupational decisions – in large part – on the basis of considerations of what would be best for the least advantaged. It requires citizens to limit the realization of their highest-order interest in fully exercising their two moral powers for the purpose of realizing the higher-order interest of others in realizing their determinate conception of the good. Citizens of a Rawlsian society must exercise their two moral powers in a way that is respectful of the interests of their fellows in doing the same. They must therefore exercise individual responsibility, tempering their claims to basic liberties against the claims of their fellow citizens to basic liberties, and tempering their claims to income and wealth against the claims of their fellow citizens to income and wealth. However, they need not temper their claims to basic liberties against the claims of their fellow citizens to income and wealth. For Rawls, claims to basic liberties, and the permissions citizens require to justly and fully exercise their two moral powers, simply take priority over claims to income and wealth.Footnote 14
Finally, even if one is unconvinced by these lines of argument, the worry that citizens lacking marketable talents may be condemned to work menial and unfulfilling jobs can arguably be allayed – at least to some extent – within Rawls's theory of justice in a way that is consistent with RFOC. Samuel Arnold develops this point in ‘The Difference Principle at Work’, arguing that the difference principle should not only be understood to govern inequalities in income, but also inequalities in the powers and prerogatives of office, that is, ‘occupational inequalities’ (Arnold Reference Arnold2012: 95). Occupational inequality, Arnold claims, ‘obtains between two jobs insofar as they convey unequal packages of the primary goods of authority, responsibility, and complexity’ (Arnold Reference Arnold2012: 106). Occupational inequalities can be reduced, Arnold claims, by flattening workplace hierarchies, democratizing workplace governance, and redesigning labour processes to reduce the division of labour (Arnold Reference Arnold2012: 107–8). Applying the difference principle to such inequalities, Arnold claims, involves maximizing the index of social primary goods of income and the powers and prerogatives of office of the least well-off in society (Arnold Reference Arnold2012: 114). The important point to note for our purposes is that the state can comply with this principle, for example, by requiring or incentivizing employers to reduce occupational inequalities, in a way that is consistent with RFOC.Footnote 15 RFOC, after all, only permits people to accept or decline offers of employment on the basis of their values and preferences; it does not entitle them to a particular occupation or a labour market free of all regulation.
2.3. RFOC and Productive Justice
One might argue that RFOC is in conflict with other requirements of justice that Rawls is committed to. Lucas Stanczyk develops this point in a recent article where he addresses the question of ‘productive justice’, that is, the question of how the goods whose distribution is governed by principles of distributive justice, should be produced in the first place (Stanczyk Reference Stanczyk2012: 144). Here, he criticizes a view of productive justice that he takes to be endorsed by Rawls, and which I defend above. On this view: (1) justice does not constrain occupational choice – as a matter of justice, people may choose whatever lawful jobs they want; and (2) justice forbids the state from forcibly assigning jobs (Stanczyk Reference Stanczyk2012: 145). Stanczyk argues that either (1) or (2) is false on the grounds that justice demands that the state secure more than (1) or (2). In cases where citizens cannot be persuaded or enticed by means of incentives to do the work necessary for the production of the goods necessary to satisfy the requirements of justice beyond (1) and (2), since justice requires the satisfaction of these requirements, then (1) or (2) must be false (Stanczyk Reference Stanczyk2012: 149). More formally, Stanczyk's claim is that the following three claims are jointly inconsistent:
(1) Justice does not directly constrain occupational choice.
(2) Justice forbids forcibly assigning jobs except where liberties are at risk.
(3) Justice requires society to ensure more than merely liberties (Stanczyk Reference Stanczyk2012: 152–3).
RFOC, as I define it above, covers (1) and (2).Footnote 16 One might therefore argue that Stanczyk's argument shows that Rawls ought not to be committed to RFOC since he is committed to (3). After all, Rawlsian justice requires the satisfaction of the second principle of justice, including fair equality of opportunity and the lax difference principle, and, one might argue, if Rawls must choose amongst these claims, he ought to reject either (1) or (2).
I think this conclusion is too hasty. First, it is important to note that what follows from Stanczyk's argument is not that citizens of a Rawlsian society do not possess RFOC, but only that RFOC is not absolute. At best, Stanczyk is entitled to conclude that in cases where citizens cannot be enticed to produce the goods necessary to satisfy the requirements of justice that go beyond the securing of the basic liberties, either citizens have a duty to take up the occupations necessary for the production of those goods, or states may force their citizens to do so. In cases where citizens can be so enticed (and so claims (1), (2) and (3) can be jointly satisfied), they possess full RFOC as I define it above. Moreover, as Stanczyk notes, there are a number of permissible ways in which the requirements of justice can be satisfied that are consistent with RFOC – i.e. the satisfaction of claims (1) and (2). In addition to the use of monetary incentives, states may also employ compulsory service systems. For example, the US National Health Service Corps provides medical, nursing and dentistry school students with publicly subsidized education in exchange for service in an underserved urban or rural area (Stanczyk Reference Stanczyk2012: 158). Students who default on their service obligation face serious financial penalties, meaning that they have no reasonable choice but to fulfil them. As Stanczyk notes, compulsory service regimes such as the US National Health Service Corps are compatible with RFOC since people have the initial choice whether to accept the subsidized education in the first place (Stanczyk Reference Stanczyk2012: 160–1).Footnote 17
Second, on a Rawlsian account of justice, the cases in which RFOC ought to be restricted in the way Stanczyk suggests will themselves be limited. On the one hand, I think that Rawls might grant Stanczyk's argument when claim (3) concerns the satisfaction of basic needs. For example, in A Theory of Justice, Rawls notes that it may be permissible for a society to adopt the ‘general conception’ of justice as fairness, which permits trade-offs between the basic liberties and other social primary goods, if the restriction of certain liberties is necessary to transform a less fortunate society into one that is capable of supporting just institutions (Rawls Reference Rawls1999a: 54, 217–18). Similarly, in Political Liberalism, Rawls notes that his first principle of justice ‘may easily be preceded by a lexically prior principle requiring that citizens’ basic needs be met’ and that this principle be given the status of a constitutional essential (Rawls Reference Rawls2005: 7, 166, 228–9). Either claim might lead Rawls to grant Stanczyk that in low income societies in which medical professionals are emigrating on mass to high income societies, either these professionals possess a duty of justice to practice in their home country, or they may be forcibly required to do so.
On the other hand, Rawls is very clear that the basic liberties be given priority over fair equality of opportunity and the difference principle (Rawls Reference Rawls1999a: 53–4; Rawls Reference Rawls2005: 294–5). Constraints of justice on the exercise of one's freedom of occupational choice may not therefore be justified by reference to either of these principles, even though they are also requirements of justice. For Rawls therefore, RFOC ought not to be restricted in cases where claim (3) concerns fair equality of opportunity or the difference principle. This conclusion is important because it follows that in societies that can meet the basic needs of their citizens through the use of monetary incentives or compulsory service schemes, citizens possess RFOC. Stanczyk's argument does not therefore lend any assistance to Cohen's critique of Rawls since the cases in which citizens’ RFOC is restricted will be limited. My reformulation of the incentives argument for inequality therefore holds for all societies in which the basic needs of citizens can be met without violating (1) or (2).
3. COHEN ON FREEDOM OF OCCUPATIONAL CHOICE
Cohen is not unaware of the general line of argument I present in part 1. In Rescuing Justice and Equality, he considers what he calls the freedom objection to his egalitarianism. According to this objection, equality, freedom of occupational choice and Pareto-optimality cannot be realized together (Cohen Reference Cohen2008: 181–3). Cohen's concern here is to allay worries that his commitment to equality is inconsistent with a commitment to freedom of occupational choice, which he defines as the legal right to decide ‘how much toil to apply . . . [and] what line to work at, at the given rates of pay’ (Cohen Reference Cohen2008: 182, 194). Cohen responds to this objection by claiming to show that freedom of occupational choice is consistent with his commitment to equality.
Cohen's response to the freedom objection, if successful, poses a serious problem for my Rawlsian response to Cohen's critique of the incentives argument. If Cohen is correct that equality, freedom of occupational choice and Pareto-optimality can be realized together, then it is unclear how citizens can appeal to RFOC to justify their acceptance of incentive inequalities to their fellows. Cohen's claim, after all, is that people can fully exercise their right to freedom of occupational choice while at the same time complying with a principle of equality that requires them to work at productive occupations without the provision of incentives.
The problem with Cohen's response to the freedom objection however, is that it does not apply to RFOC. It is simply not possible to render consistent freedom of occupational choice, equality and Pareto-optimality, if freedom of occupational choice is understood as RFOC, that is, to include not only the legal right to accept or decline an offer of employment, but also the permission of justice to do so on the basis of one's preferences and values.
To demonstrate the apparent inconsistency of equality, freedom of occupational choice and Pareto-optimality, Cohen considers the case of the doctor-gardener. The doctor-gardener is capable of both doctoring and gardening, prefers gardening at £20000/year to doctoring at £20000/year, but prefers doctoring at £50000/year to both (Cohen Reference Cohen2008: 184). If the egalitarian chooses equality and freedom of occupational choice, the doctor will choose gardening at £20000/year, thus violating Pareto-optimality (Cohen Reference Cohen2008: 185). If the egalitarian chooses freedom of occupational choice and Pareto-optimality, the doctor will choose doctoring at £50000/year, thus violating equality (Cohen Reference Cohen2008: 185). Finally, if the egalitarian chooses equality and Pareto-optimality, the doctor will be forced to doctor at £20000/year, thus violating freedom of occupational choice (Cohen Reference Cohen2008: 185).
Cohen's solution to this trilemma is to suggest that the egalitarian can have all three if the doctor-gardener chooses to doctor at £20000/year (Cohen Reference Cohen2008: 185). It is thus to suppose that the doctor is committed to equality and so will exercise her freedom of occupational choice to choose doctoring at £20000/year (Cohen Reference Cohen2008: 190). However, Cohen is only able to reconcile equality, Pareto-optimality and freedom of occupational choice in this way because his conception of freedom of occupational choice is much less robust than RFOC. On Cohen's account, freedom of occupational choice is the entitlement of justice to decide ‘what line to work at, at the given rates of pay’, free from state coercion (Cohen Reference Cohen2008: 182, 194). Freedom of occupational choice, on Cohen's account, is thus ‘secured by absence of legal obligation’ (Cohen Reference Cohen2008: 194). Freedom of occupational choice is not therefore violated by Cohen's requirement that citizens exercise this right in a way that promotes equality (Cohen Reference Cohen2008: 191). RFOC, by contrast, not only includes this legal right to accept or decline offers of employment, but also the permission, as a matter of justice, to do so on the basis of one's preferences and values. With respect to my argument in part 1 therefore, Cohen's response misses the mark. Cohen is right that if we understand freedom of occupational choice as he does, equality, freedom of occupational choice, and Pareto-optimality are consistent. If, however, we understand freedom of occupational choice as RFOC, they are not.
Cohen recognizes that others might understand freedom of occupational choice in a more robust way than he does. He therefore introduces a second trilemma that includes an alternative understanding of freedom of occupational choice, according to which it is not the freedom to choose one's occupation, but instead freedom in one's occupation: freedom as self-realization or flourishing (Cohen Reference Cohen2008: 205–6). On this new understanding of freedom of occupational choice therefore, equality and Pareto-optimality can only be secured at the expense of freedom of occupational choice; if the doctor-gardener is free to choose the occupation that best allows her powers to flourish, she will choose to garden at £20000/year (Cohen Reference Cohen2008: 205).
Cohen argues that this trilemma is also resolvable; however, there is an important difference between RFOC and this self-realization account. According to RFOC, freedom of occupational choice is not the freedom to realize oneself in an occupation; instead, it is the permission to accept or decline an offer of employment on the basis of one's conception of the good. As Cohen himself notes, these accounts come apart in an important way (Cohen Reference Cohen2008: 205). According to RFOC, the doctor-gardener exercises her freedom of occupational choice when she chooses to doctor at £50000/year instead of gardening at £20000/year, even if she could only flourish as a gardener. On the self-realization account however, she only exercises freedom of occupational choice when she chooses to garden.
Cohen briefly considers a position like RFOC, according to which persons have an ‘untrammeled moral right’ to choose their occupation. However, he dismisses it on the grounds that it presupposes self-ownership. As Cohen puts it:
Freedom is a universal value with which egalitarians must make their peace by providing a credible interpretation of it that is consonant with their views. But egalitarians are not similarly required to make their peace with the highly controversial thesis of self-ownership. (Cohen Reference Cohen2008: 214)Footnote 18
My argument in part 1 shows that self-ownership is not the only ground for such a right. Instead, one can justify a similar right by appeal to the idea that persons have a highest-order interest in exercising their capacity for a conception of the good. Since Cohen does not consider this position, his critique of the incentives argument – and his defence of egalitarianism – is incomplete.
4. CONCLUSION
Suppose that upon completion of college you are offered two jobs: (1) a position as a teacher in a science museum that pays a post-tax income of $30000/year; and (2) a position as a consultant that pays a post-tax income of $60000/year. You prefer to work as a teacher at $30000/year than to work as a consultant at anything less than $60000/year, but you prefer to work as a consultant at $60000/year than to work as a teacher at $30000/year. Rawls's difference principle, as it is traditionally understood, permits this incentive inequality since the higher post-tax income would seem to be necessary if you are to work as a consultant. But in what sense is it necessary? How, with a straight face, can you tell your fellow citizens that this incentive inequality is necessary since you are perfectly capable of working as a consultant for less than $60000/year? Are you not bound by justice to only accept those incentive inequalities that are strictly necessary for you to work as consultant, or even to decide this question in accordance with Cohen's egalitarian ethos, choosing to work as a consultant for no more than the average post-tax income?
I have argued here that you can accept the consultant position at $60000 and justify this decision to your fellow citizens. You can do so since you have a right to accept or decline offers of employment, and are permitted to make occupational choices on the basis of your preferences and values. In response to the challenge of your fellow citizens therefore, you may say the following: ‘Look, I prefer the work involved with the teacher position to that involved with the consultant position, and so it only makes sense for me to take the consultant position if it pays more, thereby allowing me to satisfy some of my more expensive preferences in my leisure time. Additionally, I am permitted – as a matter of justice – to make occupational decisions on the basis of my preferences and values.’
The principal implication of my account is that the correct understanding of Rawls's difference principle is the lax interpretation, not the strict interpretation as Cohen claims. More broadly, contra Cohen, Rawlsian justice permits the use of a wide range of incentive inequalities to realize Pareto superior states of affairs. Importantly, in drawing this conclusion, I do not, like Tan and Estlund, appeal to the idea that there are limits to the demands of justice. Instead, my claim is that justice demands that citizens be permitted to make occupational decisions on the basis of their preferences and values. My account therefore provides a novel defence of a central commitment of Rawls's theory of justice against Cohen's critical project.
ACKNOWLEDGEMENTS
I thank Joseph Heath, Louis-Philippe Hodgson, Chad Horne, Jan Narveson, Vida Panitch, Jonathan Peterson, Arthur Ripstein and Alan Wertheimer for comments on earlier drafts of this paper. For their helpful feedback, I also thank audiences at the G.A. Cohen Symposium at Concordia University, the Department of Bioethics, National Institutes of Health, and the Department of Philosophy, University of Cincinnati. Finally, I am very grateful to two anonymous reviewers for their challenging and insightful feedback, and to François Maniquet for his detailed editorial guidance and helpful comments.
