The same reasons which make it no longer necessary that the poor should depend on the rich, make it equally unnecessary that women should depend on men; and the least that justice requires is that law and custom should not enforce dependence.
J. S. MILLFootnote 1
I. INTRODUCTION
In 1848, reflecting on the attempts made in France ‘to carry Socialism into practical effect, by associations of workmen manufacturing on their own account’, J. S. Mill noted how they had begun paying equal wage rates per hour but then moved to payment ‘by the piece’.Footnote 2 He described this decision as ‘a compromise with the selfish type of character formed by the present standard of morality and fostered by the existing social institutions’.Footnote 3 For, as he explained,
the proportioning of remuneration to work done was really just, only in so far as the more or less of the work is a matter of choice: when it depends on natural difference of strength or capacity, this principle of remuneration is itself an injustice: it is giving to those who have; assigning most to those who are already most favoured by nature.Footnote 4
Mill, however, also noted that ‘until education shall have been entirely regenerated’, paying ‘by the piece’ was ‘far more likely to prove immediately successful, than an attempt at a higher ideal’.Footnote 5 Returning to this topic in 1861, Mill noted how little agreement there was regarding whether those who produce more because of their talents or skills should receive higher remuneration. People had different intuitions and appealed to conflicting, but not wholly implausible, principles. Consequently, Mill concluded that the only way in which the matter could be settled was by appealing to expediency rather than solely to intuitions about justice.Footnote 6 He remarked, however, on how quickly people invoke the need for incentives when it comes to objecting to more egalitarian arrangements, when the same objection applies to ‘nine-tenths of the business’ of our unequal society where most individuals do not ‘reap the benefits of their own exertions. . . .[Instead] from the Irish reaper or hodman to the chief justice or the minister of state, nearly all the work of society is remunerated by day wages or fixed salaries’.Footnote 7 Expediency thus had only a limited capacity to justify income inequality, which was, moreover, only a part of existing inequality, as the needy had to accept not only the worst paid, but also the most revolting occupations.Footnote 8
A century and a half later, whilst Mill's observations remain perspicacious,Footnote 9 the small pay differentials he begrudgingly accepted as temporary necessities have skyrocketed, and income inequality is at an all-time highFootnote 10 with inequalities in overall job desirability growing larger still.Footnote 11 Contemporary liberals like John Rawls, however, do not seem to share Mill's view regarding the justice of incentive payments. Instead, Rawls presents them as the corollary of his first principle of justice, the liberty principle, to which he accords lexical priority. ‘The priority of liberty’, Rawls explains, 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.Footnote 12
Rawls agrees with Mill that the distribution of natural talents is morally arbitraryFootnote 13 and yet argues that incentives are not only ‘expedient’, as Mill puts it, but ‘just’, if ‘the greater expectations allowed to entrepreneurs encourages them to do things which raise the prospects of the labouring class’.Footnote 14 If incentive payments for the best-off benefit the worst-off, perhaps because then ‘the economic process is more efficient, innovation proceeds at a faster rate, and so on’,Footnote 15 the resulting inequalities are acceptable in light of his difference principle, which permits inequalities that maximally benefit the worst-off. Rawlsians can thus defend incentive payments both by appeal to the priority of the first principle, the liberty principle, or by appeal to the second principle, the difference principle, which rejects inequalities detrimental to the worst-off, but permits non-detrimental inequalities, regardless of their magnitude.
G. A. Cohen's sustained criticism of Rawls's defence of economic incentivesFootnote 16 echoes some of Mill's insights. Some of Cohen's arguments, however, clash not only with Rawls's but also with each other, as well as with Mill's. A similar charge, moreover, can be made against Rawls. This article attempts to reconcile each author with himself, avoiding certain internal contradictions, and bring Mill, Rawls and Cohen closer to one another by defending a position that, I argue, they can all accept.
The article has four main sections. Section II explains both the initial plausibility of Cohen's campaign and how a closer examination reveals Cohen's reliance on unconvincing and contradictory arguments. Similarly, section III begins by explaining Rawls's plausible and attractive premises and then shows how these premises could lead to very different conclusions, and how conflict also emerges within the Rawlsian camp. Section IV then seeks a solution to both conflicts by reflecting on the concept of the worth of liberty, and occupational liberty in particular, showing its centrality to Rawls's theory. Section V, finally, focuses on the measures justice may require to ensure occupational liberty has worth for all. The article ends with a brief conclusion stressing the points of agreement rather than contention.
II. PROBLEMS WITH COHEN'S POSITION
A. The egalitarian ethos: three features
Cohen's critique of Rawls seems particularly plausible when considering cases involving highly paid professional groups engaging in industrial action. Take, for instance, the case of well-paid pilots or air-traffic controllers going on strike to demand even higher salaries, a recurrent situation in my native Spain. A pilots’ strike may inflict serious losses not only on company shareholders but also on passengers and company members, as well as on airport workers, such as cleaners and luggage handlers, who are far worse off than the pilots. Salary increases, moreover, are often financed either though reductions in the salaries of less essential staff, or on their number. Even individuals who have worked their whole lives for the company may be made redundant to cut down costs, while the salaries of the high-flyers sky-rocket. In Rawls's view, it appears, the fact that the pilots are worsening the situation of much poorer, more numerous and more vulnerable individuals, and inflicting significant losses on the wider society, appears not to matter, because the difference principle does not apply to particular workers’ decisions but only to the basic structure of society. So justice permits the pilots to ignore the worst-off in their decisions about industrial action.
Cohen begins his critique by noting that the pilots or other skilled groups do not become incapable of performing well without extra rewards.Footnote 17 Instead, they choose not to work, or not to work as productively, without their privileges. The need for incentives, then, arises only because of those individuals’ attitudes. Modifying those attitudes, then, could reduce the need for incentives.Footnote 18 Sharing Mill's optimism about the possibility of transforming market motivation, Cohen proposes that we adopt the difference principle also as a principle guiding personal conduct. This egalitarian ethos constrains the unlimited pursuit of financial self-interest in the labour market but includes compensation for special labour burdens, such as higher levels of risk or stress. This component is required by Cohen's defence of a partly welfarist, pluralist conception of what should be distributed equally in a just society, which he terms ‘equality of access to advantage’.Footnote 19
A second feature of this ethos is that it does not require individuals to ignore their own personal preferences and plans, for it grants an agent-centred prerogative to be guided by self-interest to some reasonable extent.Footnote 20 The third feature of Cohen's proposal is that it does not compromise liberty because it operates merely through workers’ own convictions rather than through legal coercion.Footnote 21 And in his view, ‘freedom is secured by the absence of legal obligation’.Footnote 22
Cohen also denies that the egalitarian ethos can be rejected on the grounds that, for Rawls, the principles of justice are supposed to apply only to the main institutions that constitute the basic structure of societyFootnote 23 because their ‘effects are so profound and present from the start’.Footnote 24 Cohen argues that an ethos, a sexist ethos, for example, can be sufficiently influential to be considered basic-structural, as it can drastically diminish women's prospects in life. The same applies to the choice between the market-maximizing ethos and the egalitarian ethos, which can also have profound effects on people's lives, and so can be considered basic-structural.Footnote 25
B. Moral coercion: Cohen's strategic contradiction
Cohen's arguments are powerful. Unfortunately, they are also in conflict with each other. Let me elaborate. Cohen argues that the ethos cannot conflict with liberty because the ethos operates through moral motivation rather than legal pressure, and ‘the absence of legal coercion’ suffices to secure freedom. However, the argument for including an employment ethos in the basic structure was that, like a sexist ethos, it could be influential enough to curtail individuals’ choices effectively, that is, it could conflict with liberty. Cohen is thus not only in conflict with Rawls but also with himself.Footnote 26 He needs to choose between the claim that an ethos cannot conflict with liberty and the claim that an ethos can conflict with liberty and indeed become so influential and oppressive that it must be included in the basic structure and subjected to the principles of justice. Cohen's version of the Millian insights regarding incentives, then, lacks the consistency those insights had in Mill.
C. Mill's consistency
J. S. Mill's writings on the ethics of rewarding productivity are scattered over different periods and works, where incentives are deemed both ‘expedient’ and ‘unjust’.Footnote 27 It is often said that Mill's utilitarianism and his liberalism were in tension.Footnote 28 However, Mill's 1848 and 1861 comments on incentives cited at the start of this article are perfectly consistent. First, regarding incentives, he deems them unjust because they involve ‘assigning most to those who are already favoured by nature’,Footnote 29 but we need to make a practical decision and it is more expedient to allow ‘payments by the piece’ than insist on payments per hour. Second, Mill's comments focus on the convenience of encouraging greater productivity, whilst Cohen aims to persuade the talented to choose useful professions, which is more likely to conflict with respecting their occupational liberty. Third, regarding possible conflicts with liberty, it was clear to Mill that legal coercion was not the only form of coercion, and not even the most effective form. For Mill famously wrote,
Society can and does execute its own mandates: and if it issues wrong mandates instead of right, or any mandates at all in things with which it ought not to meddle, it practices a social tyranny more formidable than many kinds of political oppression, since, though not usually upheld by such extreme penalties, it leaves fewer means of escape, penetrating much more deeply into the details of life, and enslaving the soul itself.Footnote 30
Cohen seems to agree that an ethos can coerce when he refers to the sexist ethos, and Rawls explains ‘the concept of liberty’, noting that ‘constraints may range from duties and prohibitions defined by law to the coercive influences arising from public opinion and social pressure’.Footnote 31 So perhaps both Cohen and Rawls can agree with Mill's view that not only legal prohibitions are coercive. Finally, it was also clear to Mill that even if legal coercion is not the only possible form of coercion, it does not follow that all social or moral pressure is objectionable. He would have found it bizarre if a theory of justice required protecting selfish high-flyers from public criticism. He wrote,
The acts of an individual may be hurtful to others, or wanting in due consideration for their welfare, without going to the length of violating any of their constituted rights. The offender may be then justly punished by opinion, though not by law.Footnote 32
For Mill, then, mobilizing public opinion against actions or agents which hurt the worst-off, or want in due consideration for their welfare, may be permissible, even if this effectively paralyses those actions or agents. So the discussion should not centre on whether an ethos can restrict liberty. For the fact that an ethos can coerce does not mean that Cohen cannot continue to defend the egalitarian ethos. It merely means that he cannot claim that nothing needs to give: achieving greater equality will have certain costs.
Now, whether the costs of reducing the need for economic incentives through moral persuasion and social pressure are best characterized in terms of a reduction in occupational freedom is another matter. I propose to describe them in terms of the worth of occupational liberty, a notion I shall now explain.
D. Rawls and the worth of liberty
Rawls contrasts the formal liberties, which are a ‘framework of legally protected paths and opportunities’,Footnote 33 with individuals’ real ‘capacity to advance their ends within the framework the system defines’ or ‘the ability to take advantage of one's rights and opportunities’.Footnote 34 We have a formal liberty when we count on a legal permission. The value of such permission, however, depends on whether we can employ that permission to some extent, and thus may be undermined by ‘poverty and ignorance’, or ‘a lack of means generally’.Footnote 35 Imagine, for example, a place where girls are legally permitted to attend school provided it is a single-sex school, but there is social pressure against girls attending school at all, and there is no single-sex school they can afford at a manageable distance. As a result, the merely legal liberty to attend school is of little worth to them.Footnote 36 Similarly, the legal right to occupational freedom may be of less worth in a society where the egalitarian ethos is widely affirmed and failure to conform with it is effectively enforced via informal sanctions.
Having noted a strategic contradiction in Cohen's position, and expressed a possible Rawlsian concern that Cohen's ethos could reduce the worth of legal occupational liberty, in the next section I discuss another possible strategic contradiction, this time in the Rawlsian camp.
III. PROBLEMS WITH RAWLS'S POSITION
A. Three plausible premises
Rawlsians may justify their position regarding incentives appealing to three plausible convictions:
-
(i) occupational choice has special importance;
-
(ii) once society has reached a certain level of affluence, occupational choice has greater importance than further material gains; and
-
(iii) occupational choice merits protection from legal coercion as well as from factors which may thwart de facto what de jure remains intact.
All three are powerful intuitions.Footnote 37 Occupational choice is special. We all have capacities and limitations that make us thrive in certain occupations and languish in others. Finding what fits us may make all the difference between a life of flourishing and self-fulfilment or one of failure and frustration. As Mill noted,
if there is anything vitally important to the happiness of human beings, it is that they should relish their habitual pursuit. This requisite of an enjoyable life is very imperfectly granted, or altogether denied, to a large part of mankind; and by its absence many a life is a failure, which is provided, in appearance, with every requisite of success.Footnote 38
Thus, ‘What do you want to be in life?’ is perhaps the most important question one can be asked. Answering it involves reflecting on almost everything we know about ourselves: our talents, flaws, values, commitments, and ambitions. Deciding what to be is intimately connected to the development of our potential, our identity, our intellectual skills, our personalityFootnote 39 and our ability to develop, revise and pursue our conception of the good − a claim the next section examines further. This lends plausibility to the claim that, like marriage, it should be chosen freely, without the interference of either legal or moral pressure. Finally, even the choice of partners and friends is often conditioned by our occupation. Having a choice over one's occupation is thus essential to the ability to shape one's life.Footnote 40
The second premise is also compelling. In societies where survival is not guaranteed, concern for material advancement may become an important consideration – perhaps even in respect to marriage. In developed societies, by contrast, social institutions should disregard such trade-offs.Footnote 41
Finally, the political community should be concerned not only with restrictions imposed by legal structure but also with those imposed by social structure. It is the real possibility of choosing one's vocation or husband that matters and not a mere legal permission that women are de facto unable to use.Footnote 42 Similarly, the legal right to occupational freedom may be of less worth in a society where the egalitarian ethos is widespread and informally but effectively enforced. If so, Cohen was overoptimistic about the possibility of leaving occupational freedom intact, while preserving efficiency and equality.
B. How the premises may lead elsewhere
The plausibility of the above premises, however, does not necessarily facilitate a defence of the market-maximizing ethos. In fact, those premises provide excellent arguments against so permissive an ethos. First, if occupational choice is so important for human beings, its distribution matters and it should not remain the exclusive privilege of a fortunate few. Instead, we should secure at least some valuable occupational freedom for those who have none even if this involves requiring the fortunate to collaborate in the efforts to secure such minimum by following some version of the egalitarian ethos. Second, if occupational choice matters so much more than further economic gains, we cannot cite the financial cost of granting certain occupational choice to those who lack it as a justification for not doing so. For the difference principle permits policies which involve less growth but greater benefits to the worst-off, and the plausible premises describe occupational freedom as even more important than growth and material gains.
Finally, since what really matters to individuals is not merely formal freedoms, there is no justification for granting lexical priority to maximally protecting the occupational freedom of the pilots at the expense of that of the cleaners. The latter are so deprived of choice that they may remain forever trapped in badly paid, exhausting and frustrating occupations. If we are concerned with restrictions caused by mere criticism, we should be even more concerned with the cleaners’ choices, since their constraints are much more inescapable and render them much more unfree.Footnote 43 Individuals in very low-paid occupations tend not to want to do what they do, but lack other options, have highly regimented and supervised routines, and lack escape routes as they are offered no in-job training or seniority routes to a different position.Footnote 44 The pilots, by contrast, have options and choices within their preferred occupation, and may well have fulfilled their childhood dreams about what they wanted to be, experienced flying to their heart's content, and achieved self-realization and self-respect, as well as wealth. The aircraft cleaners, however, have had neither occupational choice nor a financial compensation for its loss, which could eventually help them escape their far less freely chosen occupation. Perhaps if one of the cleaners was not underpaid and overworked, she would have completed her stained-glass evening course, set up a workshop in her garden shed and gradually made a living from this vocation; and a luggage handler might have turned his passion for dancing into a profitable occupation. But companies, whether airlines or universities, cannot always pass on the increased cost of higher salaries for their best-paid employees onto their customers, and so it is generally those with less bargaining power in the organization who pay the price. Very often, some will be forced to work longer hours, while others are made redundant to save costs. The higher the salary of the pilots, the less there will be available for the rest, and the narrower their escape options. Low taxes on high-flyers also shrink the budget to subsidize adult education or start-up business loans. Why should the pilots be granted freedom even from criticism, while others quietly accept life-sentences in back-breaking and soul-destroying occupations?
This argument is structurally similar to the feminist argument for restricting speech that silences women. More choice for the pilots means less for the cleaners. Advocates of the silencing speech argument claim that protecting the fair value of free speech for women should include a concern that female voices are not systematically drowned by male voices, whistles and a cultural environment that causes women to be seen, even stared at, but not heard.Footnote 45 The claim is not that some other value, such as equality, justifies restricting free speech or occupational choice. The claim, both in the silencing speech argument and here, is that the very same value, namely freedom, is at stake when some are being silenced or deprived of occupational choice so that others do not experience the slightest restriction. Rawlsians should be more permeable to this version of the critique of inequality-generating incentives, given that they allow that liberties may be limited ‘when they conflict with one another’,Footnote 46 and the priority of fair opportunity to be compromised for the sake of ‘chances given to those with the lesser opportunity’.Footnote 47
C. Economic coercion: a Rawlsian strategic contradiction
As noted above, Cohen's critics can challenge the sharp distinction between constraints imposed by legal and social structure to defend the talented from an ethos requiring them to moderate their demands. The baggage handlers and cleaners, however, can also oppose the pilots’ strike by arguing that the pilots’ increasing portion of a company's salary budget threatens their own occupational freedom by worsening their working conditions and escape options in the various ways just described. Rawlsians then also face a dilemma. Rejecting the egalitarian ethos on the grounds that its widespread existence will conflict with occupational liberty suggests the pilots’ choices should be protected not only from legal constraints but even from the more subtle pressure of moral criticism. But then the occupational choices of the cleaners, which are curtailed through legal factors as well as through less formal social factors affecting employment, should receive at least as much protection.
In order to reject an additional commitment to enhancing the occupational choices of the cleaners, Rawlsians may want to argue that the cleaners are legally free to leave, and that ‘freedom is secured by the absence of legal obligation’, as Cohen claims.Footnote 48 But this response is, first, incompatible with their rejecting the egalitarian ethos by claiming that even if the egalitarian ethos is not legally enforced, it still restricts occupational freedom in less formal ways. Second, advancing such a response clashes with Rawls's above-mentioned remarks that freedom may be curtailed by factors other than legal restrictions.Footnote 49 And finally, even if the cleaners are legally free to leave, such freedom may not be worth much to them, as Rawls states that the worth of liberty can be reduced by ‘a lack of means’.Footnote 50
As noted earlier (in section II.B), Cohen was downplaying the difference between legal and ethical pressure when discussing the sexist ethos, whilst when discussing the egalitarian ethos he was stressing the difference between legal restrictions which could be oppressive and an ethos which could not. By rejecting the cleaners’ complaint that their very limited choices receive insufficient protection, Rawlsians would also appear to be either stressing or downplaying at their convenience the distinction between constraints on occupational choice that have a different source. When trying to rule out the egalitarian ethos, they will be employing an expansive conception of what constitutes a constraint on occupational choice, so that even the risk of moral criticism counts as an intolerable violation. By contrast, when dismissing the complaints of the cleaners, they would be applying a much narrower view of what counts as a restriction on occupational freedom, so that so long as the cleaners are legally free to leave, their occupational choice is intact. This asymmetry seems questionable, particularly when we realize that the cleaners’ choices are not only restricted by ‘a lack of means’ and informal economic and social factors. They are also limited by the legal basic structure, which regulates pluri-employment, taxation, office hours, the working week and various factors effectively restricting the workers’ options. Such options are not facts of nature, but − as the last section shows − socially shaped and defined. There is, however, an attractive way out for Rawlsians, which involves focusing again on protecting the worth of liberty. This alternative approach is preferable to insisting that any version of the egalitarian ethos constitutes an intolerable violation of liberty, while the constraints on the cleaners involve no injustice. Such a claim is implausible, first, because there are both informal and basic structural constrains on the cleaners’ options, and second because reducing the worth of liberty of the cleaners can also be unjust. After all, Rawls states that the goal of social justice is to maximize the worth of liberty for the worst-off. And as the next section shows, this statement is at the heart of Rawls's thought.
IV. THE IMPORTANCE OF THE WORTH OF LIBERTY
Rawls states that ‘the basic structure is to be arranged as to maximize the worth to the least advantaged of the complete scheme of equal liberty shared by all. This defines the end of social justice.’Footnote 51 This section elaborates on how this general concern to enhance the worth of liberty fits with enhancing the worth of occupational liberty for the worst-off, and with central components of his theory, such as the list of primary goods, or the original position.
A. Protecting higher-order interests
As noted when defending the plausible premises, a central reason why the basic liberties are granted priority is their role in protecting our higher-order interests in developing and exercising our moral powers, and our capacity to form, revise and pursue a conception of the good.Footnote 52 Such interests cannot be served effectively by the existence of mere legal permissions to escape situations in which individuals are de facto trapped. Imagine, for example, a society dominated by a fundamentalist and superstitious creed, where there is a legal right to free speech but nobody dares to dissent or has the means to dissent publicly. In such an environment, the chances of effective rational revision would be severely curtailed. Thus, a concern with securing appropriate conditions for rational revision supports a concern with the worth of freedom, and occupational freedom in particular, as I now explain.
According to Rawls, in order develop their moral powers and pursue their life-plans, individuals require not only (a) basic rights and liberties, but also other primary goods.Footnote 53 These are:
-
(b) freedom of movement and free choice of occupation against a background of diverse opportunities;
-
(c) powers and prerogatives of offices and positions of responsibility in the political and economic institutions of the basic structure;
-
(d) income and wealth; and finally,
-
(e) the social basis of self-respect.Footnote 54
The development of the moral powers, then, requires access to certain kinds of employment. Otherwise, individuals will lack a ‘free choice of occupation’ among ‘diverse opportunities’; access to ‘offices and positions of responsibility’, and ‘income and wealth’, which in a Rawlsian society should all be acquired primarily through employment rather than through inheritance and class membership. Finally, whilst other social bases for self-respect have existed historically, in a Rawlsian society, employment will be key to what Rawls deems ‘perhaps the most important primary good’.Footnote 55 If the social bases of self-respect matter so deeply, it must matter that some face dirty, hazardous, insecure, repetitive, exhausting and badly paid jobs in which they cannot learn or make decisions, have somebody constantly looking over their shoulders, lack privacy, and can be made redundant at any time. As Rawls himself observes,
lacking a sense of long-term security and the opportunity for meaningful work and occupation is not only destructive of citizens’ self-respect but of their sense that they are members of society and not simply caught in it. This leads to self-hatred, bitterness, and resentment.Footnote 56
Promoting meaningful work and protecting the worth of occupational liberty is thus essential to protecting the development of citizens’ moral powers and willing allegiance to a social order.
B. The original position
Rawls famously claims that sound principles of justice are the objects of rational choice from behind a veil of ignorance in a hypothetical situation termed the original position.Footnote 57 The agents facing such a choice assume that it is possible that the individuals they represent turn out to occupy the least advantaged social position, and so choose the option where the worst outcome is the least bad for their representatives. Placed in such a position, the parties to the original position would not select rules conducive to individuals living in conditions in which liberty has no value, particularly to the worst-off, and exists only formally. For this is precisely the sort of disaster that the appeal to the original position is meant to avert. We cannot be expected to be willing to play our part in the social production of goods regardless of the conditions of health, safety, dignity or remuneration we are offered. This would violate ‘the strains of commitment’ by imposing an excessive burden on some of us.Footnote 58 In the original position, the parties will select instead rules that allow individuals to have more than one acceptable occupational option, and the chance to escape occupations that nobody would freely chose.
C. Beyond absence of legal coercion
Finally, as suggested earlier, focusing on the worth of liberty also allows us to address the problem of protecting occupational liberty from factors other than legal restrictions. Granting maximal occupational freedom for those who have most of it may come at the expense of those who have least of it. For example, maximal occupational liberty for men has traditionally been achieved at women's expense, confining women to housework and childcare, so that men could pursue their careers unencumbered by domestic responsibilities. Women were meanwhile left with nothing but what Mill described as ‘the feeling of a wasted life’.Footnote 59
Mill criticized a woman's reduced occupational choice as ‘a flagrant social injustice’,Footnote 60 and it is unlikely that Rawlsians would want to deny its injustice provided women are not legally bound to sacrifice for their husbands’ careers, and can instead divorce them and relinquish their children to pursue their own vocation. To assert such a position would be to rely on Cohen's implausible claim that ‘freedom is secured by the absence of legal obligation’.Footnote 61 As indicated, this is not Mill's view, or Rawls's explicit view, and Cohen himself denies this view when he discusses a sexist ethos. None can thus claim that only inequalities in occupational freedom which derive from direct legal prohibitions on women (or cleaners) constitute an injustice. All can instead adopt the view that other constraints that reduce the worth of occupational freedom of women (or cleaners) may cause injustice. These other constraints could include indirect legal constraints such as those regulating office hours and the length of the working week and informal factors such as an ethos discouraging pilots and other high-flyers from insisting on their privileges, or discouraging women from insisting on their rights. Much the same applies to the variety of factors which explain the limited occupational choices of racial minorities.Footnote 62 Once it is settled that major inequalities in occupational choice and its worth, caused by a mixture of such factors, can constitute an injustice, it is easier to reconcile Mill, Rawls and Cohen. The next step would be to grant that when such major and unjust inequalities emerge, we should seek to balance the worth of the occupational choice of those who have most (such as men or pilots) with those who have least (such as women and cleaners), and that we ought to take certain measures to reduce these inequalities. I devote the next and final section to this issue.
V. PROTECTING THE WORTH OF OCCUPATIONAL LIBERTY
A. Implementation
Some may argue that although Rawls granted enormous importance to the worth of occupational liberty and the existence of so many meaningless and precarious occupations, he did not advocate specific remedies because he deemed occupational inequalities either inevitable or impossible to measure. Neither option is convincing, however. First, Rawls was optimistic about the possibility of reforming the division of labour. He wrote:
[T]he worst aspects of this division can be surmounted. No one need be servilely dependent on others and be made to choose between monotonous and routine occupations which are deadening to human thought and sensibility. Each can be offered a variety of tasks so that the different elements of his nature can find a suitable expression.Footnote 63
Second, although Rawlsians often appeal to measuring difficulties,Footnote 64 non-pecuniary indicators of desirability, such as hours, holidays, job security, and safety, are often easier to measure than net income, as work expenses or indirect or in-kind earnings, for example, are often hard to pinpoint or measure. It would therefore be implausible to invoke practical difficulties to veto the inclusion of all non-income inequalities. Finally, although Rawls did not make more explicit recommendations, there are other examples of his commitment to an egalitarian conception of the fair value of liberty. For example, he argues that ‘a wide dispersal of property’ is ‘a necessary condition. . .if the fair value of the equal liberties is to be maintained’.Footnote 65 He also notes that protecting ‘the fair value of political liberty and fair equality of opportunity’ requires correcting ‘gradually and continually. . .the distribution of wealth’ not only though income redistribution, but through ‘a number of inheritance and gift taxes, and. . .restrictions on the rights of bequest. . .[for] the unequal inheritance of wealth is not less inherently unjust than an unequal inheritance of intelligence’.Footnote 66 In addition, Rawls proposes concrete measures to protect the equal worth of particular liberties, such as election campaign regulations and subsidies, and suggests society may have to ‘bear at least a large part of the cost of organizing and carrying out the political process’.Footnote 67
Some of what Rawls says about political freedom can be said about occupational freedom. Society may have to bear at least part of the cost of containing the growth of inequalities in the worth of occupational liberty. For example, enhancing the worth of the occupational freedom of those who lack it may require different types of basic structural reform. First, it may involve earmarked subsidies, for example, for childcare or adult education. Second, it may require legal restriction, for example, on the length of the working day or week or the number of positions one may simultaneously occupy, in order to reduce both inequality and unemployment. Third, as Philippe Van Parijs argues in Real Freedom for All, it may require an unconditional basic income that can offer individuals, whether housewives or cleaners, the opportunity to break free from oppressive situations, and eliminate not merely the poverty trap, but also the existence of occupations so dreadful that only a desperate person will accept them.Footnote 68 Regarding this suggestion, we may recall that Rawls himself favoured a social minimum and argued that any reasonable liberal conception of justice will require ‘measures ensuring for all citizens adequate all-purpose means to make effective use of their freedoms’.Footnote 69 As noted earlier, even if such measures were costly, or dampened economic prosperity, Rawlsians could not reject them on this ground. For the plausible Rawlsian premises affirm that once a society has reached a certain level of development, occupational choice has priority over further material gains. Finally, protecting the worth of occupational liberty may also require regulating the social shaping of jobs, a less familiar idea I shall now explain.
B. Poverty and occupational unfreedom
In the absence of the relevant legal or ethical restrictions, task bundles can be arranged to maximally benefit the employer, often at great cost to vulnerable employees. As Mill noted,
The really exhausting and the really repulsive labours, instead of being better paid than others, are almost invariably paid the worst of all, because performed by those who have no choice. . . .The more revolting the occupation, the more certain it is to receive the minimum of remuneration, because it devolves on the most helpless and degraded, on those who from squalid poverty, or from want of skill and education, are rejected from all other employments. . .The inequalities of wages are generally in an opposite direction to the equitable principle of compensation erroneously represented by Adam Smith as the general law of the remuneration of labour. The hardships, and the earnings, instead of being directly proportional, as in any just arrangement of society they would be, are generally in an inverse ratio to one another.Footnote 70
Rawls sided with Smith,Footnote 71 but contemporary research proves Mill right. For example, firms using hazardous technologies possess strong incentives to reduce skill and training requirements to minimize the level of compensating wage differentials they must pay. They can respond to those incentives by defining and grouping the tasks attached to a position and redrawing the boundaries of a job, minimizing the training requirements of the most hazardous occupations. Firms also possess incentives to avoid continued employment or insurance commitments to workers at high risk of injury.Footnote 72 Hazardous jobs are not only badly paid but associated with frequent unemployment because they tend to concentrate in secondary labour-market firms which are prone to bankruptcy. They also offer poor worker autonomy, on-the-job training, promotion possibilities, and greater risk of temporary and permanent lay-off.Footnote 73 Predictably, ethnic minorities, the young and the unskilled are over-representedFootnote 74 in ‘hazardous. . .uncreative, rigid, insecure and low-paid’ positions.Footnote 75 All this makes inequalities in overall job desirability phenomenally large, and even greater than income inequalities.Footnote 76 In fact, although wages are the most important single determinant of a job's desirability, combinations of non-monetary features can matter twice as much.
Thus, taking into account a job's non-monetary characteristics – such as hazards, hours, repetitiveness, worker autonomy and authority – more than doubles the estimated level of labour-market inequality.Footnote 77 It also tracks better the impact of factors like race, class, or gender, which matter two to five times more when one considers also non-monetary factors than when one counts only earnings. Women's jobs, for example, are inferior to men's on twelve of the fourteen non-monetary measures used by Christopher Jencks and his co-authors.Footnote 78 This challenges Randall Filer's theory that the reason women receive less pay is that they value a job's non-monetary features far above money and so pick the best rather than the best-paid jobs.Footnote 79 If it is true that women value non-monetary factors more than men, what this shows is that women fare worse than men even in what matters to women most.
Reducing the resulting occupational stratification could require a number of measures. One possibility would be to introduce income tax deductions for occupations with undesirable features such as high-risk levels.Footnote 80 Another option would be to require all jobs to have some redeeming features, so that not all the advantages accumulate in some jobs whilst being entirely absent from others. We are familiar with the concept of a minimum wage and we could become equally familiar with the concept of a ‘minimally desirable job’. We could do something about firms that make savings in compensatory payments by asking their worst-paid workers to perform the most dangerous or stressful tasks, whilst offering them no training or promotional opportunities. We could employ a consensual list of desirable factors and require all jobs to include some percentage of them. And we could think of many other solutions. What we first need to settle is whether an extremely stratified society with massive inequalities in job desirability should be considered merely because it includes some formal basic liberties and income inequalities which are better for the worst-off than some feasible alternatives. I think that Rawls and Cohen will agree with Mill that it should not, and that building on this agreement can help resolve their dispute, as I explain below.
C. Reconciliation
Brian Barry was right to criticize James Fiskin's assumption that equality of opportunity was satisfied in an extremely unequal society where babies were randomly switched at birth.Footnote 81 Inequality in opportunity for jobs and training is only one side of the problem; the existence of high occupational stratification is also a matter of justice.Footnote 82 If the worth of occupational freedom is to be protected, individuals must have a plurality of acceptable options, not only a formal permission to take dangerous, revolting, repetitive, and dehumanizing occupations. As I noted, this is a position behind which Mill, Rawls, and Cohen can all rally and which offers a way out of the impasse in the Rawls/Cohen debate, with each of them being trapped into their own internal contradiction. As explained at the start, in the course of his debate with Rawls, Cohen ended up inconsistently claiming that (i) an egalitarian ethos could not reduce individuals’ liberty because ‘freedom is secured by the absence of legal obligation’Footnote 83 and that (ii) freedom is not secured by the absence of legal obligation, as women's occupational freedom could be greatly reduced by a sexist ethos. Cohen may continue to claim that ‘freedom is secured by the absence of legal obligation’ providing he accepts that the worth of liberty can be reduced by non-legal factors, and that such reductions can be unjust. This is also what Rawlsians need to accept, instead of using an expansive conception of what can reduce liberty to protect the pilots from egalitarian criticism, and a narrow conception of what can reduce liberty to avoid describing the cleaners as lacking occupational liberty. Both inconsistencies can be avoided by granting that it is the worth of liberty that is limited and that such limitations can be unjust. It would be unjust to reduce the worth of liberty for the pilots excessively or when such reduction is not required to increase the worth of liberty for the least advantaged. And it would also be unjust not to increase the worth of the occupational freedom for those with least of it, if this could be achieved with the permissible measures listed earlier and by promoting some version of the egalitarian ethos.
One may now argue that this middle-ground position is further from Cohen than from Rawls, given its emphasis on occupational choice over economic equality and basic structural reforms over a personal ethos. So, let me add the following observations in response. Cohen's egalitarian ethos could enhance the occupational choice of the worst-off by reducing poverty, which, as Mill notes, is the main cause of occupational unfreedom. Also, given that some jobs are unavoidably unappealing, income is an essential instrument to diminish inequality in overall job desirability. Income, moreover, allows access to numerous after-work occupations, like stained-glass making or dancing, which provide individuals – like the protagonist of Saturday Night Fever – with many of the goods others obtain through work: self-realization, recognition, self-respect, and the chance to learn, improve, be creative and be part of a team. It would be arbitrary to say that those important aspects of human flourishing only matter when they are obtained in nine-to-five occupations.
A moderate egalitarian ethos which weighs the worth of occupational choice heavilyFootnote 84 seems both unobjectionable and perhaps essential to rally the required support for the proposed reforms.Footnote 85 Emphasizing basic structural changes and mutually binding public rules may be a good way to make the most of available altruism, by concentrating the human capacity for self-denial where it is likely to be more cost-effective. Relying on both institutional means and an ethos, however, may be better still. The absence of the right ethos can cause a whole system to collapse, as in work-to-rule strikes, or be counterproductive, as when gender egalitarian policies backfire in a sexist context.Footnote 86 An ethos may also be essential for the successful adoption of superior organizational structures, and so may remain an important part of the solution to the existence of major inequalities in real occupational choice merging from a mixture of formal and informal factors.Footnote 87
VI. CONCLUSION
Cohen claims that ‘freedom is secured by the absence of legal obligation’.Footnote 88 But he also expresses the contrary opinion when discussing a sexist ethos. This inconsistency could be eliminated if he grants that, even if liberty can be reduced only by legal coercion, the worth of liberty can be reduced in other ways. Rawls, for his part, cannot consistently claim that a non-legally backed egalitarian ethos represents an intolerable violation of the pilots’ occupational liberty, and then claim that the cleaners’ legal freedom to leave suffices to establish their occupational freedom is intact. Both can instead speak of losses in the worth of liberty.
Thus, despite their apparent disagreements, there is a position that Mill, Rawls and Cohen can all endorse. Both Mill and Rawls emphasize the importance of meaningful work and occupational freedom, and although Cohen's campaign is focused on income inequality, he does deem non-monetary inequalities, such as uncompensated differential labour burdens, unjust. In addition, his conception of equality of access to advantage would also favour reducing inequalities in overall job desirability and the worth of occupational freedom. All three could also come to agree that inequalities in the worth of occupational freedom can be unjust even if they do not derive solely from direct legal prohibitions, as has traditionally been the case with men maximizing their occupational choice at women's expense.
Finally, whilst both Cohen and Mill find incentives expedient but unjust, Rawls does not say as much. He does, however, discuss the moral arbitrariness of talent and the importance of meaningful work, and should, in accordance with key tenets of his doctrine, be concerned with improving the worth of occupational liberty for those who have the least of it. He defines the goal of social justice as the maximization of the worth of liberty for the least advantaged, and even if he does not condemn inequality-generating incentives as unjust, he could still deem unjust a society which does not balance the use of incentives with measures to enhance the worth of occupational liberty for those for whom it has least worth. Since this will require a number of basic-structural changes, which will in turn require the presence of at least a moderately egalitarian ethos, the distance between these authors can be reduced. Both can converge on the Millian position advocated here which avoids the pitfalls and inconsistencies of either Cohen's or Rawls's views, whilst retaining their most appealing insights.Footnote 89
