The first of these sections, section 41, describes the general role principles of justice have in the assessment of political economy intending to show only that doctrines of political economy have to include a notion of the public good which ensures that there has to be a role within them for principles of justice. Since the 'basic structure' is the key notion of Rawls' conception of justice it follows that political economy is seen by him as part of what creates and fashions wants for all. Hence political economy is not viewed as a self-standing discipline but as part of the consideration of moral and political matters.
In making this point Rawls considers the objection that political economy is concerned only with existing wants whereas the contract view describes an ideal conception, an argument which understands the two to be in necessary opposition to each other. However, such a stark way of presenting the relationship between the contract view and political economy is understood by Rawls to rest upon seeing the contract view as incorporating perfectionist constraints. By contrast, Rawls' own view of the contract position involves only a commitment to primary social goods albeit in so doing does make anthropological commitments of a certain sort. But the point of the theory of primary goods is that they are meant only to apply to the most general rational wants and not to include specifics that are contentiously ideal in form. So the "theory of the good" assumed within the contract view is not taken by Rawls to define an ideal of the person.
In some respects this point is, however, odd given the conclusion of the previous chapter which described a 'Kantian interpretation' of 'justice as fairness'. Taking this interpretation to require viewing the original position as incorporating conditions of autonomy does seem to require an ideal view of the contracting parties. It is likely to be Rawls' contention that this view involves no substantive commitment to the sense of the 'good' of these parties and thus to include no more than the formal constraints of right. But it is still the case that the view of autonomy that structures the original position is one that not all would accept as an ideal in the initial situation so is a move 'from' the initial situation. Precisely the move between the initial situation and the original position was, indeed, a central part (on my interpretation at least) of the argument of Chapter III.
Rawls articulates in this section further reasons for taking the notion of 'stability' very seriously as it is a prime source of the way that the formal constraints of right are manifested. This does lead Rawls to admit that the principles of justice do define at least 'a partial ideal of the person' and that this ideal is one that social and economic arrangements have to be bound by. Indeed, it is part of the point of this to resist the other pole of the stark contrast between political economy and the contract doctrine. Just as Rawls wishes to mitigate this contrast from one side by down-playing to a certain extent the ideal character of the contract view by denying that it is perfectionist so also he has to prevent the contract view merely being seen to submit (as utilitarianism is often thought to) to existing wants. It is the ideal of the person that is involved in the contract view that prevents it from simply accepting existing wants.
It is due to this ideal of the person that the contract view represents that it requires certain institutional forms to be embedded in its conception of justice. In this respect, as Rawls himself argues, there is a clear sense in which 'justice as fairness' is in alliance with perfectionism against utilitarianism. Whilst utilitarianism does not have to meekly accept existing wants either, given its commitment to optimific states of affairs, it is still the case, on Rawls' view, that its choice of ideals depends importantly upon how existing desires continue into the future. Another way of putting this is that utilitarianism does not include an ideal of the person in its first principle and in this respect is quite unlike both 'justice as fairness' and perfectionism.
The two principles of justice are not intended to be dependent upon the contingencies of existing desires but, unlike perfectionism, it is also not the case that these principles are supposed to arise a priori. The point is only that the principles are meant to arise from the construction of the original position but this construction is not one in which all conceptions would have equal chances given the way that the construction is constrained by the forms of right.
One of the oddities of how Rawls reflects back on the achievements of Chapter III at the beginning of Chapter V is that he now presents the methodology of the original position, in apparent accord with the idea of the 'Kantian interpretation' as, in an important sense, arising from a certain kind of individualism. The individualism in question is theoretical in the sense that the rationality of the contracting parties is modelled on the sense of what any one given reasoner would take to be rationally mandated. It is therefore meant to lead to a valuation of communal conditions but is not first framed in terms of how these conditions are meant to proceed. In presenting the issue in this way, Rawls may well mis-characterise his procedure since it rather appears that the appeal in the 'Kantian interpretation' to autonomy requires a sense of the mutuality of respect, something that cannot be modelled simply on the reasoning of one party.
In section 42 Rawls turns next to the question of distinct economic systems. However, as always, such systems are not considered from within economics but only from the principles of justice. Political economy is understood by Rawls to be concerned importantly with the public sector and with the regulation of property, taxation and the structure of markets. In presenting this view of political economy Rawls distinguishes between different elements of the public sector. The first element of it he considers concerns the ownership of the means of production. The classic distinction between socialistic and private-property arrangements concerns precisely this with the former viewing the public sector as the main or sole place in which ownership of the means of production takes place. By contrast, an ideally pictured private property system would appear to give the public sector a much smaller role in the ownership of means of production and perhaps circumscribe such public ownership only to generally agreed public goods such as apply to utilities and certain forms of transport.
If the ownership of the means of production is one element of the public sector the other that Rawls differentiates from it is the proportion of social resources devoted to public goods. In turning to this second element of the public sector Rawls concentrates on how such public goods have distinguishing characteristics that separates them from private goods properly considered. The characteristics in question are that public goods are indivisible in nature and they are public in quality. Indivisibility applies to the need that all elements of the public have to their supply. Since the need for them is thus general they cannot easily be parcelled into separable units of consumption.
Public goods can, on this conception of indivisibility, be understood to be have more or less of the characteristic of resistance to private consumption. In the case that is at the extreme end of the spectrum a public good has to be made completely and equally available to all. This would be 'full indivisibility' for the good over the society. The example Rawls provides of this is national defence against attack from outside. No part of the society can be denied provision of this good so it has the extreme characteristic of 'full indivisibility'. Since public goods have, to varying degrees, this characteristic of indivisibility, they are not provided by means of market arrangements but require provision by means of politics. In the extreme case of 'full indivisibility' the distribution of the public good is also not something that is considered in the sense that it is supplied to all.
However the characteristics that are specific to public goods carry with them certain special problems. One is the problem of 'free-riders' that afflicts taxation. Essentially if a good is made generally available then the calculations of some is that it matters not if they pay for its up-keep since provision will not be affected thereby. It is precisely due to this problem that coercive force is required to ensure that payments for indivisible public goods is provided. This does, however, also point to the need for such coercion to be built into the social contract. It is interesting and instructive that it is only at this stage of his theory that Rawls arrives at the requirement for coercion whereas Kant, by contrast, explicated authorisation to use coercion as the second step in his theory of right immediately after the statement of the universal principle of right. This was done even before such questions as provision of public goods was considered. Rawls' own example of national defence shows that authorisation of use of coercion in the articulation of right must be foundational for a society to be said to exist at all and so its late arrival in Rawls' theory does have the peculiar consequence of almost leading to be overlooked something that is key to political theory.
If the response to the 'free-rider' problem is expressed by Rawls by appeal to the coercive need that arises from provision of indivisible public goods the next point he raises concerning them is the notion of externality. Indivisible goods have wide effects and even if the financial basis of their provision was limited within a society its formation could be a matter of indifference to no one. Such public harms as effect the environment also have to be understood as part of the costs of the provision of the public good that produces this effect. Indeed, market mechanisms are alone incapable of responding to such externalities which is why governments have to attend to them.
Rawls' reference to the 'state of nature' is also made in this context, a notion otherwise rarely referred to in his theory and which some commentators argue has been banished from it in favour of the construction of the original position. In truth the original position cannot be seen to have simply replaced the reference to the 'state of nature'. Consideration of externalities of harm is essentially calculation of what 'naturally' arises and has to be regulated. Part of the case for the contract is made through the regulation in question. In this sense Rawls' contractarianism is far from having dispensed with the reference to the 'state of nature'.
Modern theories of rationality in terms of games theories replicate the problems of the 'state of nature' as Rawls' original accounting of the prisoner's dilemma testifies. In this situation the parties have been artificially isolated from each other and the problem is precisely how to ensure contractarian formations that promote the public good are to arise and this is nothing other than a restatement of the problem of the state of nature. Rawls terms this an 'isolation' problem and he distinguishes it from an 'assurance' one in which the latter is formulated through the question of how to effectively eliminate the free-rider question. Free-riders in a general sense are partaking of the 'benefits' of natural positioning whilst dependent upon social protection and the diminishment of the possibility of the former is required by others if they are to be assured of the point of their cooperation. But this is nothing other than the case for a sovereign power restated as an 'assurance' problem. What the distinction between 'isolation' and 'assurance' does point to is that the case for the contractual agreement is not equivalent to the case for the sovereign as the authority executive in relation to it and in this respect Rawls has added something at this point to the consideration of state of nature theory.
Having separated out the two elements of the public sector it becomes clear that the proportion of resources devoted to public goods is an analytically separable question from the ownership of the means of production. A private-property economy could allocate more or less resources to the purposes of public goods as could a socialist one. Further, the provision of public goods is not itself necessarily one that requires public providers. These are questions that Rawls leaves to 'political sociology' however and he does not consider them therefore foundational for the theory of justice.
This does not mean that Rawls has little to say about the role of market mechanisms. Rather, Rawls assumes that all regimes will use the market to ration out consumption goods since any other mechanism would be unnecessarily cumbersome. The difference between a market economy and a socialist one is not taken by him to reside there. Rather, in the free market the output of commodities is understood to be guided by 'kind and quantity' according to household preferences. By contrast the 'kind and quantity' of such output would be guided, in a socialist economy, by either 'collective decisions' or the preferences of planners.
Rawls thus concedes to 'market socialists' the view that a market economy could be utilised in important respects within a socialist system, not least due to the greater efficiency of the market in allocation of resources. However, the failures of markets are also well noted by Rawls, not least with regard to the provision of public goods. However, market systems are consistent with equal liberties and fair equality of opportunity and do not require forced direction of labour. Further, they decentralise how economic power is manifested.
The case for the 'consistency' of market arrangements with socialist institutions is made through a distinction between two separable functions of markets considered as operative in the determination of price. Price is taken by Rawls to have two functions. One is allocative and this he relates to economic efficiency. The other is distributive and Rawls connects this with how income is received by individuals. The allocative function of prices could be maintained through artificial rent attaching to the use of resources by state institutions and this is required if there is to be any efficiency in their use. But this function of pricing does not require distribution of the price charged to any owners. Distributive pricing is thus restricted under socialism but not allocative pricing.