© 2011

Social Contracts and Economic Justice

Albert Weale*

Paper prepared for Colloquium Centre for Ethics and Global Politics, LUISS University,

Rome 15 April 2011.

Address for correspondence:

Department of Political Science

School of Public Policy

University College London

29/30 Tavistock Square

London

WC1H 9QU

Tel: 020 7679 4993

Email: a.weale@ucl.ac.uk

* This paper incorporates work from a manuscript currently entitled Social Contract, Democracy,

Justice under an ESRC Professorial Fellowship on ‘Social Contract, Deliberative Democracy and

Public Policy’ (RES-051-27-0264). I am grateful to the ESRC, and so ultimately the UK tax-payer, for

such support.

2

Social Contracts and Economic Justice

Albert Weale

Within all modern democracies persistent questions for public policy arise from concerns

about property and social justice. To what degree and under what circumstances should

people be allowed to amass large amounts of wealth? How far should people be required to

show willingness to work in order to be entitled to unemployment and other social security

benefits? Does the community at large have a responsibility for the financial costs of health

care, even when that care is extremely expensive? To what degree should individuals be held

personally responsible for their choices about savings, investment in education and health-

related behaviour? Do deep and long-standing inequalities in wealth and income undermine

the common citizenship that is an essential element of a democratic political system? Does

freedom of contract extend to allowing organisations to exclude people from goods and

services because they do not like their skin colour, their mode of dress or their religious

opinions? Should the economically productive be allowed to garner the full fruits of their

labour or should they be expected to share their good fortune with fellow citizens or even

humanity at large?

Political questions about property are not new. Discussing the source of political factions in

constitutions, Aristotle remarked that economic inequality was a source of political

controversy, for those ‘that desire equality enter on party strife if they think that they have too

little although they are the equals of those that have more, while those that desire inequality or

superiority do so if they suppose that although they are unequal they have not got more but an

equal amount or less’ (Aristotle, Politics, V. ii. 2, 1302a). Rousseau (1755: 170) famously

said that the ‘first man who, having enclosed a piece of ground, to whom it occurred to say

this is mine and found people sufficiently simple to believe him, was the true founder of civil

society’, going on to lament that no one had challenged the impostor and so no one had saved

numerous crimes, wars, murders, miseries and horrors. Madison (1787) noted that ‘the most

common and durable sources of factions, has been the various and unequal distribution of

property’ such that those ‘who hold, and those who are without property, have ever formed

distinct interests in society’. Sidgwick (1891: chapter V), having accepted that private

ownership and thereby the right to exclude others from use was essential to the productive use

of land, wrestled with the problem of the extent to which uncultivated land ought to be

allowed to be taken into private ownership and in what ways the determination of the limits of

private and common ownership should be settled.

3

One answer with an ancient lineage to the question of how to resolve disagreements about the

control of material resources is to appeal to the claim that the social order should be regarded

as a form of social contract in which the norms and principles that govern the distribution of

goods and services are to be regarded as a set of conventions adopted as the outcome of an

agreement that all individuals in society have reason to accept. In Book II of The Republic

Plato has Glaucon explain this theory in extended terms:

‘What people say is that to do wrong is, in itself, a desirable thing; on the other hand,

it is not at all desirable to suffer wrong, and the harm to the sufferer outweighs the

advantage to the doer. Consequently, when men have a taste of both, those who have

not the power to seize the advantage and escape the harm decide that they would be

better off if the made a compact neither to do wrong nor to suffer it.’ (Plato, The

Republic, Book II, 358-9).

Although there are not extensive references to this view in classical political thought (see

Guthrie, 1969, chapter 5), Plato has Glaucon say that a contractual account of justice is ‘what

people say’, suggesting that he is reporting a commonplace. To the extent to which Glaucon

can be placed in a tradition, he is joined by Hobbes, Locke, Rousseau and their inheritors.

The second half of the twentieth century witnessed a striking revival of social contract theory.

In some political cultures, it became an essential reference point for to the intellectual and

ideological discussion of justice within democracies. In its modern revival, social contract

theory took many forms (notably Barry, 1989, 1995; Gauthier, 1986; Grice, 1967; Harsanyi,

1976; Rawls, 1996, 1999: Scanlon, 1982, 1998), but the various accounts were united by the

thought that that just terms of social co-operation are to be understood on a model of the

contract that each individual would have reason to make with other individuals, with all

taking due regard for their own prudential interests. From this perspective, the fundamental

question individuals have to face when confronting questions of social justice is not ‘what

arrangements conduce to my interests?’ but ‘what arrangement would I have reason to

propose as the elements of a social contract that others also have reason to accept?’. The

differing forms of social contract theory can be seen as addressing the conceptual and logical

issues implicit in rendering this question meaningful and significant. What is involved in

agents having a reason for proposing and accepting a particular set of contractual terms?

How much information about their own personal circumstances and abilities is it allowable to

ascribe to the contracting parties consistent with our being able to say that an agreement

among agents was one that was just? What is the currency of justice in terms of which

4

contracting parties are supposed to conduct their discussions: is it individual welfare,

objective interests or ‘primary goods’ understood as those goods it would be rational to want

if one wanted anything at all?

Some approaches (Harsanyi and Gauthier) have taken a rigorously reductionist form. They

have sought to characterise the rationality of the parties to a social contract in terms that were

non-ethical and, by considering choice under appropriate circumstances, derive the content of

political morality from non-ethical reasoning. As such they were able to give determinate

answers to the question of the content of principles of justice. In the case of Harsanyi this

was the principle of average utilitarianism and in the case of Gauthier the principle of

minimax relative concession. Despite the impressive quality of these constructions, each has

encountered various difficulties. Even if the premisses of Harsanyi's argument are taken as

self-evident, in the sense that the von Neumann-Morgenstern axioms were understood as

essential requirements of rational behaviour, the derivation of average utility does not pass the

test of reflective equilibrium for many people. Moreover, many have questioned the

assumption that the von Neumann-Morgenstern axioms do characterise rational behaviour, let

alone whether other aspects of the set-up, in particular the specification of the contract as the

choice of a single individual. In the case of Gauthier, the redistributive implications of

minimax relative concession will be attractive to many peoples' first-order judgements of

social morality, but serious questions have been raised as to whether the rationality of

bargaining can be characterised in the way Gauthier has done, and a more conventional

expected utility model has simply driven the problem of determining content back from the

specification of first-order moral principles to the problem of how to specify the constructivist

model. Contestability over model specification may be as essential as contestability of

political principles.

The alternative has been to move away from a formally specified account of rationality to an

appeal to reasonableness, understood as a desire to be able to justify one's behaviour to other

as well as a willingness to respect decency, as in the case of Barry (1989; 1995). Thus, for

Barry, individuals will be able to veto proposed social arrangements in the light of their

impact on their absolute level of well-being, their relative level of well-being or the need for

public goods. In these formulations the strict reductionism of the formal axiomatic approach

is absent, but the hope is still to construct a model of contractual reasoning providing an

account of content, even if the content is vague or indeterminate to some degree. Yet this

move prompts the obvious question: once one moves away from an account of rationality

specified in purely formal terms, how far is one able strictly to derive the content of a

5

justifiable political morality or how far is one simply expressing a particular morality in a

metaphorical contractual form?

Both approaches, one in terms of formally specifiable rationality and the other in terms of

reasonableness, are a priori in the sense that they are developed within a hypothetical form of

contract theory. As Brian Barry noted, the a priori method consists of asking whether there

are things that nobody could reasonably accept in the original position in the absence of

coercion or misinformation. However, Barry accepted that, though not toothless, the a priori

method may not always get us very far. Hence, an alternative, empirical, method is called for,

which Barry described in a passage worth quoting at length:

'The empirical method starts from observation rather than pure thought. It is

animated by the consideration that actual societies approximate more or less

closely the conditions … that I shall refer to for convenience as “the

circumstances of impartiality”. Thus, a society in which each section of the

population has its own organizations and organs of communication to

articulate its interests and aspirations is closer to the circumstances of

impartiality than one in which, say, business is well organized but labor is

not, and in which almost all the organs of mass communication are owned

and controlled by the rich. Similarly, a political system in which parties

represent the distinctive interests and aspirations of different groups is closer

to the circumstances of impartiality than one in which all successful

candidates have either to have money or to be acceptable to those who have

it. Again, a society in which there is a good deal of fellow feeling for other

citizens will be closer to the circumstances of impartiality than one in which

many people are unmoved by the lot of sections of the population with which

they do not identify. And, finally, a culture in which politics is widely

regarded as a matter of debate rather than as a game - where arguments are

thought of as more than the window-dressing for self-interest - will obviously

be closer to the circumstances of impartiality.' (Barry, 1989: 347-8).

The same idea is picked up in a more extended discussion in section 16 of Justice as

Impartiality where Barry develops the notion of the circumstances of impartiality as an

empirical counterpart to the Scanlonian original position (Barry, 1995: 104) and where the

guiding idea is ‘that just laws and policies are more likely to arise in actual societies the closer

they come to instantiating these hypothetical conditions’ (Barry, 1995: 100). What then

might these conditions of a culture of reasonableness be? For Barry policy making in

6

reasonable societies needs to be transparent and open so that measures are consulted on

openly and with a real chance that the consultation will make a difference. Citizens need to

be well informed and educated, with no monopoly of the means of communication and with

public funding of political parties. Policy makers need to respect expert opinion. Committees

in the legislature need to be strong relative to the executive, with weakly disciplined political

parties so that legislators can follow the logic of the argument, a pattern that will be facilitated

by multi-member constituencies using PR with a low threshold of representation. Empirically

the conditions under which political decision-making is governed by a norm of

reasonableness are to be found ‘in some of the smaller Western European countries such as

the Netherlands and Scandinavia’ (Barry, 1995: 106). Indeed, Barry is even prepared

advocate that citizens of larger countries should consider breaking them up into several

independent countries or at least strengthen the autonomy of their regional governments.

For Barry the examples of societies that approximate the circumstances of impartiality are

empirical counterparts to a purely hypothetical social contract defined in an a priori way.

Because we cannot be certain in the hypothetical model how the parties would reason, we use

the counterpart as a control on our a priori theorising. It would thus be an implication of this

approach to say that if we fail to find empirical examples of societies that embody the

circumstances of impartiality using the principle of average utility, say, in their decision

making, then we should have less confidence in any a priori model of contractualist social

morality that yielded average utility as a uniquely justifiable principle. This is not to say that

the empirical counterpart societies have a contractual basis. Barry did not assume that either

the Netherlands or Scandinavia was actually founded on a social contract negotiated in

conditions of impartiality. Rather the argument is that, however it has come about

historically, their political practices resemble the conditions that we would want to specify in

a contract for justice. In the remainder of this paper I take up this research programme

outlined by Barry to explore what implications it has for our theory of economic justice.

Social Contracts as Resource Regimes

Consider communities of the sort studied by Elinor Ostrom (1990) in which common pool

resources have to be managed. Common pool resources are a species of public good. Pure

public goods (in the economist’s sense) are characterised by non-rivalness and non-

excludability. In the case of common pool resources, there is some rivalness, since, with a

common pool resource, use by others, in particular use by a sufficient number of others, will

deplete the value of the resource for any particular individual. Examples of common pool

7

resources of this type cited by Ostrom include fishing grounds, groundwater basis, grazing

areas, irrigation canals, bridges, parking garages, mainframe computers and streams, lakes,

oceans and other bodies of water. These common pool resources are typically large enough

to make it difficult, if not impossible, to exclude potential beneficiaries from use, but they

also have the feature that the actions of these potential beneficiaries can spill over onto the

use by others, as the example of over-fishing illustrates only too well, or as can be

experienced by anybody who has tried to park their car at a busy time of the day in a public

car park.

Problems of common pool resources are often treated by social scientists as instances of the

prisoners’ dilemma, the classic example being Hardin’s discussion of the ‘tragedy of the

commons’. Ostrom’s approach, however, is not to look at these issues in an a priori way, but

instead to examine empirically cases in which such common pool problems have been

successfully managed (sometimes over centuries) by contrast with those cases in which they

have not been managed successfully. One of Ostrom’s examples will give the flavour of what

successful management involves (see Ostrom, 1990: 18-21). In Alanya in southern Turkey

coastal waters were being over-fished as a result of the over-capitalisation of the fleet and the

competition for increased yields. In the early 1970s the local co-operative in Alanya began

experimenting with allocating fish sites to local fishers, which consisted of the following

system:

1. Each September a list of eligible fishers was prepared.

2. Within the area normally used, all fishers and all usable fishing locations were named

and listed. The sites were so spaced that the nets in one site would not block the fish

in an adjacent site.

3. These named locations were in effect from September to May.

4. In September the named fishers drew lots and were assigned to the named fishing

locations.

5. From September to January each day each fisher moved east to the next location;

after January each fisher moved west one place.

Note that we have here an example of what Young called an institution: '... identifiable

practices consisting of recognised roles linked by clusters of rules or conventions governing

relations among the occupants of those roles' (Young, 1989: 5). One of the ways that such

institutions operate is by providing norms that govern actions, thus enabling coordinated

human activity to take place. Ostrom argues that the evolution of institutions, of the sort of

which Alanya provides an example, solve the collective action problems associated with the

8

management of common-pool resources. In essence, what is happening is that the institution

is providing a norm of behaviour for individuals, and action in accordance with that norm is

sufficient to prevent the free-riding that undermines collective action.

Common pool resource regimes will emerge in situations where the physical environment and

mode of production create the inter-dependence in which short-term individual maximising

behaviour will create adverse effects for others. The examples of successful regimes cited by

Ostrom include mountain grazing and crop harvesting in Switzerland and Japan and water

management in Spain and the Philippines. These are cases where agreement on common

rules is necessary if the underlying resource is to be conserved for the future. The success of

regimes is evidenced by the length of time that they have performed the function of

conserving resources. Among the oldest are the regime that governs grazing and foraging

rights in Törbel in Switzerland, where the articles of association were signed in 1483, the

management of common lands in Japan during the Tokugawa period (1600-1867) and beyond

as well as the water irrigation regime in Valencia, Spain going back to 1435 (Ostrom, 1990:

62, 65 and 69 respectively). These and other regimes have survived fluctuating pressures in

fragile environments. Ostrom does not claim that these regimes are optimal in any sense, the

uncertainties and variable climatic conditions would make this difficult, but she does claim

that they are successful given the challenges they confront and the period of time over which

they have functioned (Ostrom, 1990: 59-60). The communities in which the resource regimes

are established are stable and persist over time, so that costly investment at one time carries

the expectation of return at a later time for the members of that community.

Can we really model the idea of a just social contract on the pattern of common pool resource

regimes? In answer to this question, it is worth noting that common pool resource regimes we

are considering involve situations where joint action is required by agents in order to achieve

their individual good. Ostrom (1990: 31) highlights the extent to which we need to

distinguish between the act of appropriation of a resource, for example catching fish or

abstracting water from a river basin, where producers are in competition with one another,

from the act of maintaining the resource system, for example stocks of fish or river basin

management, where producers have an interest in common in maintaining the integrity and

well-functioning of the system. Producers are in a situation of scarcity not abundance, and so

they are in competition with one another for the harvest that comes from the joint resource

system, but they need to co-operate with one another to maintain that system in being. Given

this situation they cannot avoid questions over the terms of their co-operation.

9

The second feature of common pool resource institutions that can make them a model for just

social contracts is that they arise in situations of rough equality of power. All participants in

the resource regime have some ability to impose harm on others (Ostrom, 1990: 26). To be

sure the equality of power is only likely to be rough in the sense that some individuals may be

in a better position to impose harm than others. For example, those upstream in a river basin

may be able to abstract more than those downstream in such a way that there is less than is

required for those farming downstream. Similarly those with better capitalised boats may be

able to harvest more fish from a given stock of water than others less well capitalised.

However, the general situation with common pool resource regimes is that each participant

has some ability to impose harm on the rest. This element in the conditions of a social

contract was identified in Glaucon’s account of justice as contract and it is of course a marked

feature of social contract theory in the modern era from Hobbes onward. If there were

marked inequalities of power, then a hegemon could establish the conditions under which

some forms of production could take place, and we would not have a social contract.

Putting these elements of the requirement of joint action and rough equality of power together

shows how common pool resource regimes satisfy Hume’s three conditions for the

circumstances of justice. Given that participants in common pool resource regimes are in

competition for scarce resources, they are not in a situation of superabundance in which there

is no competition between the different members of society. However, if the resource system

is well maintained, neither are they in a position of harsh scarcity in which time horizons

become short and rules of controlled behaviour break down. Hence none of these societies

are in the words of Hume (1751: 182) in ‘such want of all common necessaries, that the

utmost frugality and industry cannot preserve the greater number from perishing, and the

whole from extreme misery’. In fact Hume’s summary of the conditions of justice could well

serve as a general summary of the circumstances that we find in common pool resource

regimes:

‘The common situation of society is a medium amidst all these extremes. We are

naturally partial to ourselves, and to our friends: but we are capable of learning the

advantage resulting from more equitable conduct. Few enjoyments are given us from

the open and liberal hand of nature; but by art, labour, and industry, we can extract

them in great abundance. Hence the ideas of property become necessary in all civil

society: Hence justice derives is usefulness to the public: And hence alone arises its

merit and moral obligation.’ (Hume, 1751: 183).

10

Thus, if we take Hume's account of the circumstances of justice as a guide, we have reason to

regard resource regimes as forms of social contract.

Hume's thought that the ideas associated with property become necessary in civil society also

highlights another reason why we might regard resource regimes as forms of social contract.

The regimes have at their centre restrictive rules to constrain appropriation (Ostrom, 1990:

59). We thus have some situation that embodies Hart's (1955) natural right to freedom, a

situation in which the principle of fairness can be applied as a condition for constraining

freedom. If we think of the question of justice as being one in which individuals have to

exercise restraint on their freedom, then analysing the principles that emerge from CPRs

provides us with one such model. But what principles of economic justice does the model

suggest?

A Constrained Full Fruits Principle of Justice

In the common pool resource regimes identified by Ostrom, a general principle is that

participants receive the full fruits of their labour. The fish they catch, the wood and plants

they harvest or the cows they graze belong to them. Strictly speaking, since work is

organised around the household, it is the household unit that receives the full fruits of the

labour of its members. This is an important qualification to which we shall come back later,

but for now we may say that the product of labour is owned by those who undertake the work

involved in securing the product. Many theorists and others regard the logic of the full fruits

principle as inegalitarian. It allows those with more productive power to reap more than those

with less. Yet the communities in which common pool resource regimes operate are

egalitarian communities in the sense that producers enjoy an equality of status in determining

the functioning of the regime. Democratic participation produces an outcome that rewards

according to relative productivity. How is it possible to combine these two elements? To

answer this question, we need to remember that in the common pool regimes, the simple rule

of productivity is qualified in various ways.

To understand the extent to which there is a general problem here, it is necessary to

characterise the social enterprise in which participants are engaged. Within common pool

resource regimes, we are not considering a Lockean state of nature in which individuals can

appropriate what they will. Rather we are considering a situation in which participants

become entitled to appropriate natural resources for themselves, provided that they conform

to the commonly laid down rules. This situation goes beyond a Hohfeldian liberty, in which

11

each is free to appropriate but none is under any obligation to allow the other to appropriate.

In a (null) regime governed by Hohfeldian liberties, participants are free to appropriate, but

are under no obligation to allow others to appropriate. For example, two salvage vessels may

compete with one another to be first to the wreck. Under common pool resources regimes, by

contrast, there are duties to refrain from interfering with others appropriation, for example by

waiting one's turn in a rotation system.

However, with CPRs this limitation on a Hohfeldian liberty-right is a modest one. When

participants secure the full fruits of their labour, they are under no obligation to share their

product with others. Why should this be? The contract abridging the Hohfeldian liberty is

intended to be one to mutual advantage. That is to say, each participant intends to be better

off as a result of the contract than without the mutual restraint, because the contract promises

an end to the mutually destructive exercise of Hohfeldian liberties. No one would agree to

give up their Hohfeldian liberty unless it was advantageous to do so. However, although this

argument has some force, it does not establish the full fruits principle by itself. All this

establishes is that without some gain, it would not make sense of an appropriator to sign a

contract that involved a restriction upon the freedom to appropriate. It does not show that in a

contract from which each gains, some gainers might have to share the reward of their work

with others. After all, one can gain and still share some of the fruits of that gain with others.

Indeed, this principle – that in gaining one is under an obligation to share the fruits of one's

labour with others, is a central part of Rawls's rationale for the difference principle, as we

shall see later. So the logic of contractual advantage, by itself, will not justify the full fruits

principle. What is the logic of saying that the mutual gain is accompanied by the right to

retain the full fruits of one's labour than by mutual gain accompanied by pooling or

redistributing the fruits of labour?

Firstly, although the agreement is one that allows participants to keep the full fruits of their

labour, the mode of production is such that variation in the ability to harvest reflects primarily

skill differences and luck rather than position within an organisation. Members of the

community are harvesting from nature and therefore have a limited capacity to increase their

returns above a certain level. Because successful common pool resource regimes last a long

time, skill acquisition is through custom and practice, and though there are likely to be some

differences in this regard, the span of taught ability in relevant respects will be low, as skills

and techniques are diffused throughout the community. In short, differences in productivity

are not related to the emergence of a separate class of persons who can control significant

elements of the rate of return to effort.

12

Secondly, across common pool resource regimes, there is a random, and therefore equalising,

element in the assignment of rights to harvest or to use resources. In Törbel, for example, the

trees used for heating and construction are marked to be felled, and assigned by lot to a group

of households. In Hirano, Nagaike and Yamanoka each kumi (collection of households) was

assigned a zone by rotation for harvesting winter fodder, and the crop was harvested

collectively for technical reasons to do with the minimisation of accidents. In Valencia a

rotation scheme is used for access to water via the communal canals, from which each farmer

can extract as much as possible in turn, whereas in Murcia each farmer is assigned a fixed

time period for water use. In the zanjera communities in The Philippines, when water is

scarce, rotation systems are established. To be sure, straight randomisation (and therefore

equalisation over time) is not the only rule that is used for allocating rights. In Törbel, a

proportionality rule is used for grazing rights, with each farmer allowed rights in proportion to

the number of cows owned and a proportionality rule is used in the lower zanjera

communities for the allocation of water. However, such proportionality, whilst a

modification to a strict egalitarianism is not a gross departure from an egalitarian arrangement

given its limited extent.

The rule of appropriation from labour is not one that is based upon a substantive conception

of what participants are entitled to, for example the rule that they are entitled either to a

minimum share of produce or to an equal share or even to a share that though unequal

maximises their expectations, as would happen under the difference principle. In fact, the

practice of appropriation gives considerable play to procedural elements in the sense that

appropriation is nested in agreed procedures that leave flexibility in outcomes. From one

point of view the fundamental principle of appropriation can be regarded as one of equality of

opportunity, when the relevant opportunities are made up of the physical features of the

harvesting regime with suitably randomised allocation of appropriation rights. However, it is

also possible to argue that there is such equality of opportunity, involving the opportunity to

become unequal, because implicitly it is understood that there is simply a lack of capacity on

the part of different participants to accumulate so much property that they would fall into a

different economic class from their peers. The unequal outcomes that arise as a result of the

equal opportunities are thereby limited in span.

Since the returns to individuals require there to be agreement on collective appropriation

rules, we can see the rules of allocation as taking place in a context in which there is a

contract to maximise the long-term total appropriation from the resource regime. If there

were not this commitment, then it would always be possible for a participant to propose

changing the rules so as to increase yield without detriment to future stock. With a fixed

13

sized population, this is equivalent to maximising the average yield of participants. The

constraints imposed on individual appropriation that typically emerge are ones that can be

seen as ones that have the effect of constraining variance around the mean. With their

essential procedural element, this is not to say that the rules of justice require appropriation to

conform to some profile of return. In some years, under some circumstances, variance will

increase or decrease purely by chance or as the result of variations of effort among

participants. Nonetheless, although an average return maximizing rule, it is so with an

implicit constraint on the span of variation around the mean.

Variations in private appropriation are also off-set by the fact that all participants have access

to the common resources. In many cases one could imagine that there is a private ownership

solution, according to which some individuals could command the use of the previously

common resource. Enclosure has after all been a frequent phenomenon in agricultural

communities where labour is necessary to secure product. However, it is a significant feature

of common pool resource regimes, noted by Ostrom herself, that the commons remain

common, with the implication of collective control – in which participants have a share – over

the use of the commons. The maintenance of resources as commons thus plays a part in

limiting the extent to which individuals can accumulate advantages over others.

In noting the constraints that are implicit elements in the appropriation rule of the common

pool resources, I do not mean to suggest that the formal rule is the adoption of a maximum

average returns principle plus a constraint relating to the variance around the mean. We

should still conceptualise the rule as allowing those working to receive the full fruits of their

labour, but acknowledge that this rule is acceptable because the social and economic

conditions in which it is adopted are ones in which the variance of returns will not be high.

However, for clarity, it is best to refer to the relevant principle as the 'constrained full fruits'

principle, rather than a simple principle of allowing people to take what they can get.

Can we give a formal account of why the constrained full fruits principle would be adopted as

the result of a social contract? To do so would be to locate the principle as the upshot of the

bargaining arising from the calculus of those individuals who have to form the social contract.

Given the situation of a Hohfeldian liberty, in which no one owns the whole of the commons'

natural resources, can we identify the calculus of choice that would lead to a convergence on

the constrained full fruits principle within the limits of the collectively agreed harvesting

rules? The social contract is an agreement to abridge the individual liberty to harvest by

participants. In this situation, we might think that those who can anticipate being more

productive or luckier in respect of their labours will not accept a contract that gives them less

14

than their anticipated marginal product. Those who anticipate being less fortunate cannot

force the more productive to do so. So perhaps the constrained full fruits principle has its

origins in this logic? However, those who anticipate being less productive cannot be forced

into an arrangement in which there is no sharing at all, since they could always threaten to

continue acting under the freedom to appropriate what they can, thus undermining the basis of

the social agreement. Of course, this threat could mean that there is less available for all,

since over-harvesting of a resource will lead to the depletion of that resource, but that is

simply the nature of the threat-advantage that the less productive have.

The difficulty of identifying how an agreement might emerge from the calculus of individual

choice is, I suggest, general and reflects an underlying weakness in our analytical tools. For

example, as an alternative to the log-jam given above, we could imagine coalitions of actors

coming together to agree the rules of a regime that they could impose upon the all members of

the community. However, if coalitions are allowed to shift, so that the members of a proto-

winning coalition can be induced to defect to another, the core of the multi-lateral bargaining

game will be empty. In short, in the absence of a theory of multi-lateral bargaining (Muthoo,

1999: 336-38), reliable predictions about the likely outcome of the multi-lateral bargaining

game are impossible to find. Instead we need to look at what alternative rationales might lead

to a situation in which the constrained full fruits principle emerges as the solution to the

contractual problem.

The Logic of the Full Fruits Principle

Before setting out a possible rationale for the full fruits principle in terms of which we can

understand its role within a social contract theory of economic justice, let me indicate one

way in which one might think about this problem that seem to me to be misleading, although

it has figured strongly in some accounts of economic justice.

Consider, in the first instance, Locke's account of primary acquisition, which has some

similarities to appropriation under common pool resource regimes, although there are

significant differences. Locke formulated a right to property as a result of the effects of

labouring on the natural world:

'He that is nourished by the Acorns he pickt up under an Oak, or the Apples he

gathered from the Trees in the Wood, has certainly appropriated them to himself. No

Body can deny but the nourishment is his. I ask then, When did they begin to be his?

15

When he digested? Or when he eat? Or when he boiled? Or when he bought them

home? Or when he pickt them up? And 'tis plain, if the first gathering made them

made them not his, nothing else could. That labour put a distinction between them

and common.' (Locke, 1690: 330).

The argument here appears to be similar to one of backward induction. Suppose we have a

chain of necessary conditions from A to D, such that: only if A, then B; and only B, then C;

and only if C, then D. Then if D is permissible, so must be A, B and C by the principle of

those who will the end, also will the means. If we believe that it is right for people to have

appropriated, then we are committed to thinking that the sequence of acts that led to the

appropriation was permissible.

Locke construed the right to property as a natural right. That is why he thought that when

individuals appropriated resources from the commons, they are free to do so without the

consent of others (Locke, 1690: 330). On this account, two people may both labour for the

same acorns without either infringing the property rights of the others. Of course, when some

are successful and others are not, then the losers have not been able to exercise their rights to

advantage, but for Locke this is no injustice since all, winners and losers, are in a situation in

which there is enough and as good left for others by those who have successfully

appropriated. Locke intended his argument to be a refutation of Filmer's claim that God gave

the earth to Adam, and so to his successors in common, so that any individual appropriation

would require the consent of the community (Ryan, 1984: 16-17). Locke hopes to show by

appeal to the labour theory of property that the consent of the community is not necessary.

However, the move from 'no injustice' to 'no need to require permission of others' is rather too

quick. All may have an interest in conserving their labour and not wasting it in a fruitless

search for what others are going to appropriate, even when they recognise that others will

have the right to appropriate should they get to the resource first. If there really is enough and

as good for all to appropriate sufficient to meet their needs, then often all will benefit from

co-ordination in matters where the exercise of rights are concerned. If labouring is a

necessary condition for the acquisition of property, it is seldom sufficient, since preventing

wasted labour involves co-ordination even where there is as much and as good left for others.

Considerations other than justice may imply a right for the community to control the

appropriation of individuals.

Locke thought that there would always be as much and as good for others, because the same

law of nature that allowed the appropriation of the fruits of the earth also forbade

appropriators to take more than they could enjoy by themselves: 'Nothing was made by God

16

for Man to spoil or destroy.' (Locke, 1690: 332). By the same token, people are entitled to

acquire land, so long as there is enough and as good for others: 'For he that leaves as much as

another can make use of, does as good as take nothing at all.' (Locke, 1690: 333). It is this

condition that is not met in the case of common pool resource regimes. Such regimes are

necessary in order to control the adverse effects that one appropriator can have on others.

Common pool resources are not pure public goods in which the consumption of one person

does not detract from the consumption of another, although there may be threshold effects, by

which it takes a certain number of appropriators to have an effect, so that it will be true in

some cases that what one person does is immaterial.

The Ricardian socialist, Thomas Hodgskin (1832: 24-6) endorsed the Lockean right to

property through labour as a claim of natural right, citing the passage that I have quoted, as

part of a general argument denying that capitalists and landlords had any right to the product

of those who laboured. Hodgskin contrasts this natural rights position explicitly with the

Benthamite doctrine that all property is a consequence of the law and that the law-maker

establishes the rules of property. Instead, the right of workers to the full fruits of their labour

stems from the combination of the power to labour and the natural wants that stimulate work.

Laws may give landlords and capitalists the right to a return on their assets, but in doing so

they are establishing an artificial, rather than a natural, system. Moreover, rising productivity,

as a result of the application of human skill and ingenuity, will enable workers to live on

smaller and small plots of land (Hodgskin, 1832: 61-2). From this point of view, a natural

right to property might underlay a common pool resource regime. It would allow inequalities

in returns to labour, as Hodgskin (1832: 41) allowed, noting that there was no complete

sharing of goods even within a family. But it would deny a share in the product of labour to

those who were not involved in producing it.

Hodgskin's position seems to be the right one for someone to take up who is persuaded by

Locke's labour theory of property as a natural right arising as a result of mixing one's labour

with nature. However, the natural right is one that is hard to defend. To be sure Locke's

backward induction is correct if one is concerned with a theory of possession. Once you have

eaten the acorn, there is no doubt that you possess it. But why should possession in this sense

be construed as a right to property, rather than as merely appropriation? Moreover, once we

move beyond the realm of immediate consumption to the ownership of a stock of capital

resources, the claims that others may make upon one's possessions are not limited by the

physical fact of possession. If I borrow without permission your coat to go out on a cold

night when I know that you are staying at home, I have infringed your property right, but I

have not deprived you of your possession. The two concepts are simply not coextensive. In

17

the absence of common rules that are legislatively agreed in some way, there is no

satisfactory way to define one's right to possess certain objects.

I suggest that to see an alternative way of identifying the special significance of one's own

labour and ts relationship to ownership, we need to consider what values might be lost if the

connection were not maintained. In particular, we need to consider what values would be lost

from the point of view of a social contract construction, in which deliberative rationality, the

equality of standing of the participants and common interests were central ideas.

To illustrate this, consider a situation in which each individual was entitled to the full fruits

not of their own labour but of someone else's labour. Let us imagine a community confronted

by a common pool resources problem that constrained the appropriation of each individual

producer, but then randomly assigned the product of each person's effort to someone else.

Each person is matched with one other person. No person can keep his or her own product,

but no person is left without someone else's product upon which they could call. Of course,

such an arrangement would increase transactions costs compared to a situation in which

producers keep their own product. Transactions costs here are defined as in standard theory

as the costs of running a particular economic system. Under the one-to-one assignment

regime, records of transactions would be needed, products might have to be stored waiting

collection and transport would have to be provided. However, quite apart from these

transactions costs issues, there are other serious issues.

The first such issue concerns the loss of incentive to production that such a system would

entail. Where individuals are allowed to keep the fruits of their labour, they have greater

incentive to put in more effort and so raise productivity. Even a modest bias to self under

one-to-one assignment would reduce total product, since at the margin there would be no

reason to extend effort as much for others as for oneself. Of course, paired individuals would

seek to bind themselves in various ways, making the transfer of their product conditional

upon certain minimum requirements forthcoming from the other side. Here again there would

be increased transaction costs, since agreements would need to be drawn up and compliance

monitored. So, from the point of view of common interests, there is a reason to allow

individuals the full fruits of their labour as a way of ensuring that resources are not wasted.

However, questions of productivity cannot be the whole reason favouring the own fruits

principle over someone else's fruits principle. A scheme of lump-sum ability taxes as well as

the experience under Soviet planning, in which incentives were structured so as dramatically

to increase productivity (Olson, 2000: 114-44) show that increasing productivity has no

18

essential connection to keeping the fruits of one's own labour. In principle, it is possible for

central planners so to structure incentives within an economy so as to increase productivity

even when workers do not keep the fruits of their own labour. Here again informational

requirements will increase transactions costs, but this is a distinct matter from saying that

productivity would fall if some other principle than that of keeping one's own fruits were

adopted. What this suggests is that, if one is to explicate the special significance of one's own

labour, then it is necessary to supplement productivity considerations with considerations of

another type.

Deviations from effort for others compared to oneself also risks problems of fairness in

exchange. Under the principle of one-to-one assignment two parties exchange their products.

But the exchange might not be fair in the sense of being an exchange of equivalents. Within

the standard construction of an exchange economy, participants exchange goods at their

equivalent value, given by the price where the economy is large enough. However, under

one-to-one assignment, we do not assume that goods have a price. Instead there is just the

stock of goods your counter-part has produced in a given period, which is assigned to you,

and the stock of goods you have produced in a given period assigned to your counter-part.

There is no reason why such an exchange should lead to a transfer of equivalents. Perhaps for

example your fishing or your crop growing has been blessed with good fortune during the

course of time, whereas your counter-part has been unlucky. In terms of exchange value,

price, you would not be getting the full value of what you have given up, whereas your

counter-part would be getting more by way of exchange value than what he or she was giving

up. In that sense, exchange would be unfair. It could be, of course, that these differences

would be evened out over time if the pairing went on long enough, but this just highlights the

increased transactions costs that are associated with such a regime.

Yet again, although the reassignment regime creates problems of fairness, it is possible to

imagine situations in which these problems are dealt with. We could suppose that instead of

being assigned the product of one other person, you were assigned the average product of

everyone else but yourself, via a mechanism that first pooled product and then redistributed it.

However, then the question arises as to how the balance is struck between what one puts into

the common pot for others and what one is allowed to draw out of the common pot for

oneself. If one put into the pot as much as the average of everyone else, then there is no

problem, but this cannot be guaranteed.

So far we have not considered the value of freedom, but this consideration seems to be

indifferent between a full fruits of one's own labour regime and a regime in which one

19

acquires an entitlement to the fruits of others' labour. Under both regimes monitoring and

penalties are required to maintain the protection of common pool resources and ensure that

the right to harvest is fairly upheld. How far such monitoring and enforcement infringes on

personal liberty by comparison in the two regimes is an open question. The extent to which

extra monitoring and enforcement is required in one rather than the other does not turn on the

principles that are embodied in each but on the willingness or otherwise of the participants to

follow the principles of their own free will. It may be, of course, that there is a principle of

human nature by which one is more inclined to abide by rules that permit one to enjoy the full

fruits of one's own labour than by a rule in which one is entitled to the fruits of others' labour

conditional upon their being entitled to the fruits of your labour. Certainly, such a rule has the

virtue of simplicity and once adopted by a segment of the population may even have a self-

reinforcing quality to it. Also, it is significant that it is the rule that is followed in the

empirical examples of successful common pool resource regimes that we have available. If

functional necessity is interpreted as what arises from a situation in which agents are seeking

an optimal response to the conditions and constraints that confront them, then the functional

necessity of the full fruits principle under observed empirical conditions seems strong.

Having acknowledged all this, there does not appear to be a general reason why the full fruits

principle should dominate in collective choice a principle of sharing. It may simply depend

upon preferences. It is at least conceivable that there are some societies in which able

novelists would prefer to write chain novels with others than to compose themselves. If such

preferences prevailed, then more coercion would be necessary to enforce the full fruits

principle than to enforce the swapped fruits or the shared fruits principle. In short,

considerations of freedom cannot help us decide the intrinsic merits of the different rules,

though in a context of practical decision making they may provide a reason of enforcement

for favouring one over the other.

Comparisons with Other Principles of Justice

The full fruits principle, suitably constrained, contrasts with a number of other principles of

economic justice. We can highlight the distinctiveness of the empirical approach to contract

theory by contrasting its principles with other principles frequently advanced in contemporary

debates.

The Difference Principle. For Rawls 'the higher expectations of those better situated are just

if and only if they work as part of a scheme which improves the expectations of the least

20

advantaged members of society' (Rawls, 1999: 65). The canonical statement of the principles

governing social and economic goods runs as follows (Rawls, 1999: 266):

'Social and economic inequalities are to be arranged so that they are both:

(a) to the greatest benefit of the least advantaged, consistent with the just

savings principle, and

(b) attached to offices and positions open to all under conditions of fair

equality of opportunity.'

Thus, according to the difference principle, no inequality is just unless it works to the

advantage of the least well off.

The difference principle contrasts with the full fruits principle, even in its constrained form.

The constraints imposed on producers securing the full fruits of their labour consist in

limitations on their freedom to use resources beyond a point at which the cumulative effect of

such use would degrade or destroy those resources as well as any other principles of

assignment, usually reflecting some random element that evens out access to those resources.

The constraints have the effect of imposing limits on the span of inequality that might

otherwise be generated by a simple application of the full fruits principle. (These limitations

reflect the rough equality of power assumption that is a feature of justice-conforming social

contracts.) However, since there is no redistribution of the fruits of labour from those whose

efforts secure more to those whose efforts secure less, inequalities will sometimes (perhaps

even typically) arise when the position of the least well-off could be improved by

redistribution from the better off to the worse off. To understand the significance of this

contrast, it is useful to compare the difference principle with the full fruits principle both in

terms of their respective logical structures and in terms of the values and conceptions of

society that they presuppose.

In logical terms, the difference principle is at first sight what Nozick (1974: 155) called an

'end-state' principle of justice, that is to say a principle that judges the justice of a distribution

by its correspondence with a particular pattern, in this case the conformity of the distribution

with some specified pattern of outcome in the economy. An end-state principle of justice

places an obligation on those agents for whom the duties of justice are relevant an obligation

to bring about that end-state. Thus, we may suppose that an end-state conception of justice

would oblige the political authorities to bring about that end-state, or at least establish

institutions the general tendency of which was to produce that end-state, and require members

21

of society to cooperate in that goal. However, although this is one possible reading of the role

of the difference principle, it does not seem to be the way that Rawls conceived matters.

Properly understood the difference principle does not place on members of a society a duty to

ensure that the position of the least well-off is as high as it can be. Rather it says that when

individuals seek advantages for themselves, they can only do so in justice so long as they

secure the maximum benefits possible to the least advantaged. Thus, Rawls says:

'… the more advantaged have a right to their natural assets, as does everyone else;

this right is covered by the first principle under the basic liberty protecting the

integrity of the person. And so the more advantaged are entitled to whatever they can

acquire in accordance with the rules of a fair system of social cooperation.' (Rawls,

1999: 89, italics added)

On this way of reading the difference principle, the right to one's natural assets is prior to any

obligations one may have to promote the well-being of the least advantaged. That this

became an important element of justice in Rawls's account of the difference principle is

highlighted by the fact that this reference to the right to one's natural assets and the

entitlement to whatever one can acquire in accordance with the rules of a fair system (which I

have italicised) is added to the corresponding statement in the first edition and strengthens the

claims made therein. Here is the original passage:

'… the more advantaged representative man cannot say that he deserves and therefore

has a right to a scheme of cooperation in which he is permitted to acquire benefits in

ways that do not contribute to the welfare of others.' (Rawls, 1971: 104)

In other words, there is a move in Rawls's rationale for the difference principle from the

earlier claim that one has no liberty to acquire advantages unless one thereby benefits others

to the later claim that one has a liberty to acquire advantages provided that others are

benefited. Although people do not deserve their natural talents, principles of freedom and the

basic integrity of the person mean that people do have a right to their natural assets.

How does this matter to our interpretation of the difference principle? Nozick said that it was

an end-state principle that made justice a matter of a pattern in the outcome of economic

transactions. However, this is potentially misleading since it might seem to imply that the

members of a society should act so as to maximise the well-being of the least well off. On

this view of the difference principle, Cohen (2000: 126-7; compare Cohen, 2008: chapter 1) is

22

right to say that Rawls's acceptance of economic inequality is problematic, because if the

more productive members of society affirm the underlying values of the difference principle,

they would see that they had no right to secure more by way of incentives when they could

make the same contribution for the same final income as everyone else. However, on the

view being offered here, the difference principle does not impose an obligation upon all to

raise the well-being of the least well-off to the highest level possible, but rather a condition

under which when the more productive use assets that are rightfully theirs, they should also

accept an obligation to pool the benefits of those assets. The difference principle says that

when seeking to advantage themselves all members of society may only take those benefits

that arise from a system in which the well-being of the least well-off is maximised.

The distinction is important because it has implications about the extent to which the more

productive are required to exercise their abilities. Taken as a goal of public policy, a goal that

had to be internalised by the better off, the difference principle might require the more

productive to work beyond any freely chosen level of effort in order to raise the well-being of

the least well-off. To illustrate the issue, consider how the balance between work and leisure

might be structured. Suppose a society in which the more productive chose to work only

three days a week. In such a society, the least well-off might be better off were the more

productive to work for five days a week. Yet, if working the extra two days is not valued

those with more productive assets, the situation of the worst-off is not improved. The more

productive may choose not work the extra two days, because they do not value the gain they

receive in material goods as highly as they value the loss of leisure they incur from the extra

days of work. Under a scheme in which the combination of work and leisure were freely

chosen, the gains to the least well-off would not be secured.

We can strengthen the contrast even further by noting that the lot of the least well-off could

be improved, without any loss of economic efficiency, by tilting the incentives of the better

off toward longer working hours by imposing a lump-sum tax on ability that the well off were

required to pay. In these circumstances, the better off would work longer hours and a portion

of the material benefits they produced would be available to the least well-off to improve their

position in accordance with the difference principle interpreted as a social goal. Standard

economic theory shows that a lump-sum tax does not distort incentives at the margin and so

will not reduce work effort below that which is maximally attainable. Were such lump-sum

taxes to be imposed, then there would be grounds for Nozick's (174: 169) otherwise

implausible claim that taxation is 'on a par with forced labour'. Another way of putting the

point is to say that such taxation is inconsistent with recognising the separateness of persons

(Weale, 1993: 81-2). If we see the issues from this point of view, it is possible to see why

23

Rawls revised his earlier formulation of the difference principle by making it clear that the

freedom and integrity of the person need to be preserved in any account of a just distribution.

The fruits of the labour of the better off are to be redistributed to the least well-off, so far as

possible, but conditional upon a free choice about work effort made by the better off.

At the root of the difference principle in Rawls's formulation, there is a specific conception of

society, the economy and the part that individuals severally and collectively play in

production. Here is a passage it is worth quoting at some length (differences from the first

edition are stylistic rather than substantive):

'The difference principle represents, in effect, an agreement to regard the distribution

of natural talents as in some respects a common asset and to share in the greater

social and economic benefits made possible by the complementarities of this

distribution. Those who have been favored by nature, whoever they are, may gain

from their good fortune only on terms that improve the situation of those who have

lost out. The naturally advantaged are not to gain merely because they are more

gifted, but only to cover the costs of training and education and for using their

endowments in ways that help the less fortunate as well. No one deserves his greater

natural capacity nor merits a more favorable starting place in society. But, of course,

this is no reason to ignore, much less to eliminate these distinctions. Instead, the

basic structure can be arranged so that these contingencies work for the good of the

least fortunate. Thus we are led to the difference principle if we wish to set up the

social system so that no one gains or loses from his arbitrary place in the distribution

of natural assets or his initial position in society without giving or receiving

compensating advantages in return.' (Rawls, 1999: 87)

It is this commitment to the idea of natural talents as a common asset that formed the moral

core of Rawls's proposal for the difference principle. (Later we shall consider the claims of

some luck egalitarians that the difference principle does a poor job of capturing the full force

of the common assets principle.)

If it is right to identify the difference principle with this conception of natural talents as

forming a common asset, then we have a clear contrast with the constrained full fruits

principle. In the Locke/Hodgskin interpretation of the full fruits principle, the exercise of

labour power was the means by which goods and products are removed from the common

store and reserved entirely to the use of particular individuals. Locke's example of the acorns

and oaks, which merely replicates the practice of those involved in common pool resource

24

regimes, is intended to establish that it is labour that enables individuals to appropriate. Thus,

rather than regarding labour power as a common asset, and therefore under the control of

those who manage common assets, the full fruits principle rests on the assumption that labour

distinguishes what is common from what is private. Under common pool resources, no one

owns the labour of another, though each recognises that in order to enjoy the returns to their

own labour they must grant a similar right to others.

What this contrast with the difference principle suggests is that in order to show the moral

kernel of the full fruits principle, we need to show what might be the special connection

between the free exercise of labour power and justice in distribution. How could it be that a

society of free producers, capturing within agreed constraints the full fruits of their labour,

embodied an ideal of economic justice? In answer to this question I submit that my earlier

rationale was the right one. Empirical examples of common pool resources reflect the

Humean circumstances of justice, and in these circumstances participants will seek for

principles that are to the common advantage under conditions of approximate equality of

power. Herein lies the rationale of the full fruits principle.

Luck Egalitarianism. The luck egalitarian principle is that people should be able to benefit

from their option luck, but should not be able to benefit from brute luck. The standard case of

option luck is that of a gamble that is voluntarily entered into. The argument in this case is

that one cannot object to the inequalities of option luck, because the individuals involved

knew the range of outcomes that confronted them. Logically, this can be contrasted with

three positions, the last of which – though logically interesting – is not of much substantive

importance. The first position, which libertarian, is that one should be able to benefit from

both option luck and brute luck. The second, the position of the pure egalitarian, is the one

should benefit from neither option luck nor brute luck. All benefits should go into the social

pot for redistribution. The third and last position is that one should be able to benefit from

brute luck but not option luck. No one, so far as I know, defends this position in the current

debate, but it is a position that could be held from a puritan point of view of a Weberian kind.

In this case gambling is thought of as morally wrong, but providence has distributed different

capacities and gifts to different individuals. Though interesting, I leave this position aside in

the present context.

Theoretically, it is possible to challenge the conceptual distinction between brute luck and

option luck. Dowding (2010) has some good examples in which brute luck can be transferred

into option luck. In general, brute luck can distinguish people into those that have better

25

skills to appraise situations and decide which options they wish to take a punt on. The luck

egalitarian, however, can reply by saying that these are difficulties in operationalising the

principled distinction, they are not as such objections to there being such a distinction.

Maybe one day social science will advance to the point where it can, for example, identify the

influence that good parenting has on the ability of children to make prudent choices, and will

find some way of extending the practices of good parenting to all children. For example, the

ability to make good option choices requires various computational skills in probability, and

we may be able to understand the conditions under which calculation skills are nurtured and

enhanced.

Let us leave this question of whether the distinction between brute and option luck is as

significant as some have claimed. How does CPR justice stand with respect to option luck?

How far is option luck the right principle to apply?

Notice that in the CPR case, the use of randomisation is common and this might lead us to

think that we were concerned with a species of option luck. However, it is socially

constructed option luck. In this sense, it is not like a lottery in which people can choose to

participate or not. The need to preserve the commons means that rules are needed to control

access to the resource. As an individual, one does not have a choice to participate or not in

the lottery. Suppose one was of a superstitious turn of mind. One might feel that things

would go badly for you in the lottery. Still, it would be fair to require you to enter, since that

is the price for all attaining the highest good that they can. So it is not right to say that one

should be free to benefit from a willingness to take a risk, because you have no choice as to

whether to take the risk or not.

Both brute and option luck seemed to be involved in determining the harvested returns that

people get in CPR regimes. From one point of view, it is a matter of option luck as to which

slots one is randomly assigned to. However, brute luck may be involved in determining one's

ability to take advantage of the options. One may work harder or more efficiently through

having been well trained when young by parents or teachers, or perhaps one just has more

strength. The full fruits principle will entitle one to keep the product of one's labour, however

it arose through the combination of luck and skill. Can we find a basis for such a principle?

One consideration that applies here is that the span of difference is not enormous. That is to

say, although some people benefit more than others as a result of good fortune, no one is in a

position to benefit to the extent that their share becomes dominating. The equality of social

relations is preserved to that extent. The second point is closely related. The ability to benefit

26

is limited by the raw exercise of human skill in fishing and farming. The benefit is not gained

as a result of one's place in a certain sort of organisational hierarchy, in which one is in a

position to cream off some of the benefit of the labours of others.

There is some parallel here between the justice of this example, and the conditions that

Gauthier (1986: chapter VII) identifies for saying that people are not entitled to keep the full

fruits of their labour. Notice, however, that we cannot go as far as Gauthier and say that there

is a fundamental distinction between income in the form of rent and income that is a

compensation for time or effort. Rent will exist in the well-functioning CPR economy

because some people would undertake the same activity for a smaller product, depending on

their attitude to work, leisure, consumption and so on. Removing rent is not only impossible

in practice, it is exploitative in theory, because it would require the imposition of a lump-sum

tax on ability to avoid disincentives.

Redistribution?

Contractualist reasoning in the empirical mode suggests that at the core of economic justice is

the idea that those who work are entitled to the full fruits of their labour provided that the

labour takes place in conditions that are suitably constrained, both with respect to the

aggregate burden that is imposed on productive resources and with respect to ability to secure

on a regular basis advantageous productive positions. The chief reason for thinking that this

account provides us with an account of justice is that the Humean circumstances of justice are

met in the relevant communities: members of the community have modest altruism and an

equality of power. Whatever they decide will be just in these circumstances.

However, there is one obvious problem with this approach. An important element in our

thinking about justice is that goods – or at least some goods – should be distributed according

to need. The point is not simply that a good society is one that ensures that the needs of its

members are met. Rather, to claim that distribution according to need embodies justice is to

say that those in need have claims that they can assert as a matter of right against the fellow

members of their society. Where the relief of poverty or destitution is not associated with

claims based on right, then we may have a regime of humanity or compassion, but we do not

have a regime of justice. Yet, redistribution according to need seems incompatible with the

principles that those who work are entitled to the full fruits of their labour.

27

This dilemma was put very well by Cohen from within his own socialist theory of justice,

although he notes that the same point was made by Anton Menger nearly a hundred years

before. Cohen begins by noting that the traditional communist theory of the working class

was that it combined four distinct elements: it constituted the majority of society; it produced

the wealth of society; its members were the exploited in society; and its members were the

needy people in society. For Cohen, a significant problem facing any theory of justice is the

way in which under contemporary class conditions, these elements come apart:

'Particularly problematic, from the point of view of a socialist political philosopher, is

the coming apart of the exploitation and need features. It forces a choice between the

principle of a right to the product of one's labor embedded in the doctrine of

exploitation and a principle of an equality of benefits and burdens which negates the

right to the product of one's labor and which is required to defend support for very

needy people who are not producers and who are, a fortiori, not exploited.' (Cohen,

2000: 108)

The implication of this disjuncture of circumstances is that a theory of justice can no longer

be based upon a claim by workers to the full fruits of their labour. How can one respond to

this difficulty?

The first point to note is the empirical social contract is being used as a model of reasoning

about principles of justice, not a model of actual social organisation. Because the full fruits

principle is typically adopted for empirical social contracts that the same principle can be

applied to social and economic conditions in the real world. Perhaps the most important

feature of the empirically modelled social contract is a rough equality of power among

participants. However, in the absence of that condition being satisfied, application of the full

fruits principle would allow actual inequalities of power to influence the fate of people. If

real world conditions are unjust, then a simple application of the full fruits principle might

compound that injustice rather than alleviate it. For example, the ability to labour is

dependent upon training and the imparting of skills. Applying the full fruits principle in the

absence of such training would not mean greater justice but less. Similarly, an important

element on the applicability of the full fruits principle is that individuals labour in the context

of a randomising assignment of opportunities, thus off-setting positional advantages that some

might gain over others. Where this condition is not satisfied, then it would be wrong to apply

the full fruits principle without modification.

28

We also have to consider how need arises before considering the role that it has a theory of

justice. Any human being is going to be in need at some point in their life, if only when an

infant. More generally, periods of need typically coincide with the life-course – infancy and

old age – or with contingent events that create financial dependence, whether that be illness or

interruption of work. Much of what we observe as redistribution to the needy is not in fact a

form of vertical redistribution from the well-off to the poor, but a horizontal form of

redistribution from periods in peoples' lives when there is a surplus to periods in their lives

when there is a deficit. Institutionalised redistribution can therefore be conceptualised as a

form of collective insurance rather than a simple transfer from one group to another. It

certainly ought not to be thought of as a transfer from those who work to those who do not

work. When such collective insurance is provided, members of society are not being

deprived of the full fruits of their labour. Rather they are choosing to consume the fruits of

their labour in ways that even out the fluctuations of consumption over the life-cycle. To the

extent to which this situation obtains, there is no conflict between redistribution according to

need and a fair return to labour.

Even for schemes of redistribution that are built upon the principle of collective insurance, we

cannot simply assume that there is a balance between contribution made and return secured.

In many cases this is simply a consequence of the nature of insurance. The pooling of risks

that insurance involves can only work if those who need to draw down on the insurance are

able to rely upon the surplus generated by those who do not. If those who are unlucky enough

to need the insurance are poor relative to those who are fortunate enough not to need it, then

there will be redistribution of income from rich to poor. But this is an incidental by-product

of there being a collective scheme of insurance in place, not a policy to deprive people of the

fruits of their own labour. The labour of the fortunate has yielded them enough fruit to enable

them to obtain the security that comes with the insurance.

However, it may be argued that even horizontal forms of redistribution are unfair in the sense

that implicitly the insurance pool is defined as one that is society-wide. Suppose a sub-set of

those who are members of society could collectively insure with one another to their mutual

advantage, because, for example, they constituted 'better risks', just as those who live high

outside the flood plains of a river can obtain better house insurance than those who live on the

flood plain itself (other things being equal). If we think about collective schemes of

redistribution as being forms of insurance, do we not imply that were some sub-group to be

able to secure better terms for itself than the rest of society, there would be a violation of the

full fruits principle? The more advantaged could obtain the security of insurance at lower

cost thus absorbing less of the fruits of their labour. If they are compelled to contribute to a

29

society-wide scheme, then they have unnecessarily expended some portion of the full fruits of

their labour that was not for their own benefit. We seem to have returned to the Rawlsian

assumption that the social contract is to be regarded as 'an agreement to regard the distribution

of natural talents as in some respects a common asset'. To that extent labour does not enable

individuals to take from the common stock of assets but is made to contribute to the common

stock of assets.

However, the full fruits principle is not a cab one can get on and off at will. In particular, it

enables us to ask how any group is in a position to be able to secure for themselves better

terms and conditions than the common run of members in society. If that advantage is purely

as a result of their labour – say because they have worked hard, saved when they could have

consumed and been more fortunate in the returns on their efforts – then compulsory

enrolment in a society-wide scheme of insurance would be hard to justify. It would certainly

not be in accordance with the full fruits principle. However, suppose it is due to inheritance

or some other form of income unearned without labour, then the situation is more

complicated. The full fruits principle says only that people are entitled to at least the full

fruits of their labour. It does not say that they are only entitled just to the fruits of their

labour. Other sorts of income, derived from gift or other good fortune, can be justly enjoyed,

without compromising the full fruits principle. However, if exemption from the collective

insurance is made possible by the unearned income in such a way as to worsen the situation of

those who have only their earned income, then there is a problem, because exercising the

rights of unearned income has in effect deprived those with only earned income of the value

of their resources. Those with more than earned income have in effect obtained a positional

advantage. In practice, of course, it may be difficult or impossible to tell when this has

occurred, but conceptually the point is valid.

Collective schemes of insurance are like flood defences in that in providing protection to

some they provide protection for all on terms that do not respond to positional advantage.

However protection for a sub-set can only be undertaken in a market in which the relatively

fortunate trade with some others who are willing to underwrite whatever risks are at issue. In

many cases, such market based risk redistribution raises no issues of justice. There are

differences in attitudes to risk just as there are differences in consumer preference for other

goods and commodities. However, with some insurance events – ill health being the most

obvious example – there are well-known market failures in health insurance stemming from

the fact that third-arty payments are involved. Full development of this argument must await

The Political Theory of the Property-Owning Democracy, the companion volume to this

work, but here it is important to record that, to the extent to which market-failure arguments

30

can be employed, we have no violation of the full fruits principle if we insist that the scheme

of insurance in collective – even though it will produce its own form of redistribution.

Special vulnerabilities arise in respect not of production but of reproduction, vulnerabilities

that extend to dealing with stages in the life-course. The empirical model of contractual

reasoning appeared to have the defect that it was premissed on households, rather than

individuals, as the unit of account for considering issues of justice. However, when

considering the issue of redistribution, this bracketing of internal relations in the household

and the domestic division of labour turns out to be an advantage, because it means that we

have to consider explicitly what is involved in moving away from the household as the unit of

analysis. From the point of view of individuals, the household is the location for the

exchange of services, and potentially contains the most extensive form of redistribution.

Indeed, so important is the topic that it is given a chapter of its own in the larger work of

which this paper is a part.

The Hayekian Challenge

It can be argued that the principles of justice developed so far suffer one major weakness,

namely that they are suitable for small scale societies, but are not suitable for what Hayek

(1973) called the 'Great Society', that is a society that arises as a spontaneous order out of the

multiplicity of interactions among a large number of agents rather than as the product of a

single organising intelligence. Hayek (1976) thinks that social justice is a mirage in such

societies, because in seeking for some overall ordering of social affairs it is ignoring the

limitations on organisation that exist in the Great Society.

Properly to answer this challenge would take up more space than is available here, but there

are two points that need to be made. The first is that Hayek's position seems too strong since

it would cast doubt on any attempt to impose a conception of justice upon a society, and

would therefore seem to undermine any conception of justice that was not libertarian. In fact,

Hayek resiles from this implication in his discussion of Rawls, noting that Rawls is concerned

to define principles that cover the general tendency of social and economic institutions to

produce certain sorts of results rather than an attempt to impose a particular type of

distribution in society. Quoting Rawls (1963: 102) who says that 'the principle of justice

define the crucial constraints which institutions and joint activities must satisfy if persons

engaging in them are to have no complaints against them' and that whatever distribution

emerges in such a situation 'may be accepted as just', Hayek claims that this is similar to his

31

own position (Hayek, 1976: 100). Whether there is the consistency that Hayek claims is a

moot point, but the discussion does at least show that whatever constraints of justice might be

inferred from a model of decision making in a face to face society, that they are so derived is

not an objection to those constraints merely by virtue of the source of that derivation.

However, this is not the end of the reply to Hayek. For we need to note that 'great societies'

could only be such with a high degree of social coordination allowing the creation of

corporate actors, who could enter into legal and economic arrangements with others. No

Great Society' is a result of spontaneous order. Principles of corporate ownership and liability

have typically been the result of explicit legislation. The most obvious example of this is

legislation to the creation of the limited liability company. In the UK for example in the

twenty years before the initial legislation in 1844, there was a vigorous debate among

economists as to whether limited liability was compatible with the principles of freedom of

contract (Atiyah, 1979: 364-68), with some arguing that there should be no restrictions on the

freedom of persons to enter into an exchange with any group of people formed on whatever

basis and others arguing that limited liability would encourage persons to enter into

speculative and risky enterprises because they themselves had nothing to lose. As Atiyah

(1979: 566) notes, there was a certain unreality to the discussion, because 'limited liability

was only part of a broad process … to limit the responsibility of contracting parties.'

However, we understand these discussions, it is clear that since limited liability is the result of

a deliberate legislative act, no Great Society is purely the result of a spontaneous order, which

is the condition that Hayek's critique needs to satisfy.

Moreover, 'he who says organisation says oligarchy'. Organisation allows for the detachment

of personal effort from a share of organisational returns. Asset specificity and opportunistic

behaviour in a context of asymmetric information mean that some individuals will derive

more benefit than their marginal contribution and others less. Compounded by barriers to

occupational entry and corporate control of natural and social resources, we cannot read off

from the returns that individuals derive what their labour has involved. When these features

of economic life are influenced by the disproportionate influence of a social class or group,

then injustice arises. One response to this has been to seek to alter the forms of economic

organisation within which people work in large scale societies, but proposals to create a

situation in which workers hire capital, rather than capital hiring labour, have founded on a

number of problems. The alternative solution, arrived at when the balance of power among

social classes has been more or less favourable, has been to insitutionalise some form of

redistributive social policy.

32

These points are only a sketch of an answer to the Hayekian challenge, but I hope they go

some way to showing that a principle of economic justice built upon the principle of workers

enjoying the full fruits of their labour, suitably constrained, is not only an outcome of an

empirical social contract, but could have wider application beyond the types of society in

which that principle is predominant.

References

Aristotle, The Politics, translated by H. Rackham (Cambridge, Mass. and London: Harvard

University Press).

Atiyah, P. S. (1979) The Rise and Fall of Freedom of Contract (Oxford: Clarendon Press).

Barry, Brian (1989) Theories of Justice (London: Harvester-Wheatsheaf).

Barry, Brian (1995) Justice as Impartiality (Oxford: Oxford University Press).

Cohen, G.A. (2000) If You're an Egalitarian, How Come You're So Rich? (Cambridge, Mass.

and London: Harvard University Press).

Cohen, G.A. (2008) Rescuing Justice and Equality (Cambridge, Mass. and London: Harvard

University Press).

Dowding, Keith (2010) 'Luck, Equality and Responsibility', Critical Review of International

Social and Political Philosophy, 13: 1, pp. 71-92.

Gauthier, David (1986) Morals by Agreement (Oxford: Clarendon Press).

Grice, Russell (1967) The Grounds of Moral Judgement (Cambridge: Cambridge University

Press).

Guthrie, W.K.C. (1969) History of Greek Philosophy III (Cambridge: Cambridge University

Press).

Harsanyi, John C. (1976) Essays on Ethics, Social Behavior, and Scientific Explanation

(Dordrecht: D. Reidel).

Hart, H.L.A. (1955) ‘Are There Any Natural Rights?’, Philosophical Review, 64, pp.175-91.

Hayek, F.A. (1973) Law, Legislation and Liberty, Volume 1, Rules and Order (London and

Henley: Routledge and Kegan Paul).

Hayek, F.A. (1976) Law, Legislation and Liberty, Volume 2, The Mirage of Social Justice

(London and Henley: Routledge and Kegan Paul).

Hodgskin, Thomas (1832) The Natural and Artificial Right of Property Contrasted (Clifton:

Augutus M. Kelley, 1973 edition).

Hume, David (1751) Enquiry Concerning the Principles of Morals, reprinted in Essays,

Moral, Political, and Literary by David Hume, edited by T. H. Green and T. H. Grose

(London: Longmans, Green, and Co., 1889), pp. 167-287.

Locke, John (1690) Two Treatises of Government, edited P. Laslett (New York: Mentor,

1965).

Madison, J. (1787) ‘Federalist Paper No. 10’ in The Federalist Papers, edited with an

introduction by C. Rossiter (New York: The New American Library Inc., 1961 edition).

Muthoo, Abhinay (1999) Bargaining Theory with Applications (Cambridge: Cambridge

University Press).

Mueller, Dennis C. (2003) Public Choice III (Cambridge: Cambridge University Press).

Nozick, Robert (1974) Anarchy, the State and Utopia (Oxford: Basil Blackwell).

Olson, Mancur (2000) Power and Prosperity: Outgrowing Communist and Capitalist

Dictatorships(New York: Basic Books).

Ostrom, Elinor (1990) Governing the Commons: The Evolution of Institutions for Collective

Action (Cambridge: Cambridge University Press).

Plato, The Republic translated F. M. Cornford (Oxford: Clarendon Press).

Rawls, John (1963) 'Constitutional Liberty and the Concept of Justice', reprinted in

Rawls, John (1971) A Theory of Justice (Oxford: Oxford University Press).

33

Rawls, John (1999) A Theory of Justice (Oxford: Oxford University Press, revised edition).

Rousseau, Jean-Jacques (1755) Discourse on the Origin and the Foundations of Inequality

among Men in Jean-Jacques Rousseau: The First and Second Discourses, edited, translated

and annotated by Victor Gourevitch (New York: Harper & Row, 1986).

Ryan, Alan (1984) Property and Political Theory (Oxford: Basil Blackwell).

Scanlon, T.M. (1982) ‘Contractualism and Utilitarianism’ in A. Sen and B. Williams (eds),

Utilitarianism and Beyond (Cambridge: Cambridge University Press), pp.101-28.

Scanlon, T.M. (1998) What We Owe to Each Other (Cambridge, Mass.: Belknap Press).

Sidgwick Henry (1891) The Elements of Politics (London: Macmillan).

Weale, Albert (1993) ‘Justice, Social Union and the Separateness of Persons' in David

Gauthier and Robert Sugden (eds) Rationality, Justice and the Social Contract (New York

and London: Harvester- Wheatsheaf) pp. 75-94.

Young, Oran (1989) International Cooperation (Itahca, NY: Cornell University Press).