Introduction

Let us imagine a world in which all institutions satisfy the minimal conditions of social justice so that basic human rights are secured for everyone as far as it is reasonably possible. Let us further assume that in this world, all injustices perpetrated in the past have been adequately rectified so that no one can complain that his situation was wrongfully caused by others. And last but not the least, let us stipulate that the domestic societies found in this world are integrated in a global economic order, which regulates their trade and capital flows. The question I wish to address in this article is the following: Should we, in the name of justice, object to this world if equally talented and motivated persons still have dramatically unequal opportunities because of their place of residence?¹ Of course, protection of basic human rights and rectification of past injustices already go to some length in reducing inequalities between persons, since the former involves a minimal threshold under which nobody is allowed to fall, and the latter includes compensation for harms wrongfully caused by others. But is this sufficient or does justice also require that opportunities — or at least 'some' opportunity — be equalized worldwide?

The purpose of this article is twofold: (1) to show that in the presence of global market competitions, gross global inequalities of resources are likely to translate into objectionable inequalities of opportunity and (2) to show that a plausible version of global equality of opportunity can be constructed, which demands that equally talented and motivated persons who participate in the global economic order should have a roughly equal chance to benefit from this order if they so choose, irrespective of the society to which they belong. This undertaking is not without difficulties, all the more because the domestic ideal of equality of opportunity has itself been subject to vehement criticism. Some have pointed to internal inconsistency as it can only be achieved at the expense of the nuclear family, and more broadly, of any form of social relationships that can affect things such as the development of talent and the willingness to make an effort (Charvet 1969; Lloyd Thomas 1977; Fishkin 1983: 44–82; Rawls 1999a: 64). If equally talented and motivated persons are to have an equal chance of attaining given advantages, it indeed seems to follow that all social factors that can influence their capacity to develop and exercise the talents and motivation necessary for obtaining these advantages must be neutralized, including the passing on of an affective, cognitive, cultural or material heritage through the family. In recent years, the ideal of equality of opportunity has also been repeatedly denounced as morally unacceptable by the so-called 'responsibility-sensitive egalitarians' (Arneson 1989; Cohen 1989: 916–17; Roemer 1998).² Although these authors may differ in what they consider to be morally justified and unjustified inequalities, they all agree that persons cannot be held responsible for the possession of their natural talents and cannot therefore be entitled to the benefits flowing from the exercise of their natural talents. Yet, by levelling off social circumstances, proponents of equality of opportunity mean precisely to reward persons for their aptitudes and to some degree these will be attributable to brute luck. This, according to responsibility-sensitive egalitarians, amounts to arbitrarily favouring the talented over the non-talented. What should be levelled off, they argue, are all differential circumstances of individuals that (a) lie beyond their control and (b) affect their ability to attain advantages.

Obviously, one should not expect the global version of the principle of equality of opportunity to be more popular than its domestic counterpart. For to the difficulties already mentioned should be added those related to its global scope. So it has been objected that place of residence is not a morally arbitrary characteristic of the person, but should be allowed to influence the opportunities open to him/her and may thus justify the existence of global inequalities of opportunity (Miller 2005: 67–70). Another difficulty concerns the absence of global shared understandings: it has been claimed that the magnitude of cultural diversity at the global level makes it impossible to implement a global principle of equality of opportunity (Boxill 1987; Miller 2005: 59–64).³ The underlying idea is that equality of opportunity cannot make sense unless people agree on the meaning and relative importance of various social goods, and that forcing such agreement upon them would be unacceptable because it would fail 'to express due toleration for other acceptable ways [...] of ordering society' (Rawls 1999b: 59), or because cultural diversity is a valuable good that must somehow be preserved (Boxill 1987: 152).

The combination of these difficulties explains that the ideal of global equality of opportunity has not been the subject of much discussion. Few attempts made to defend it have been extremely cautious, indeed sometimes very vague (Pogge 1994: 196, 198, 221–2; Caney 2001; Moellendorf 2002).⁴ And more fundamentally, to my knowledge, no decisive counterargument has been offered to defeat the criticisms of its most radical opponents (Boxill 1987; Brock 2005; Miller 2005).

The first part of this article presents the two predominant interpretations of the idea of global equality of opportunity — namely, that all persons of equal talent and motivation in the world should have, respectively, 'identical' or 'equivalent' opportunity sets — and shows that they are both subject to serious objections. It then argues that this should not lead us to abandon egalitarian commitments at the global level, for doing so would amount to leaving the potential unfairness of global market competitions (as well as of the processes through which their rules are defined) unaddressed. It also notes that the fairness of competitions is at the heart of a specific kind of equality of opportunity — namely, 'competitive equality of opportunity' — which is immune (at least to a certain extent) to the charge that no standard for comparing opportunities across cultures can be found.

The second part elaborates on the idea of competitive equality of opportunity and applies it to the present global economic order. It argues that a conception of global equality of opportunity cannot make sense unless it is confined to one specific opportunity understood as some specified agent(s)'s chance to attain some specified goal(s) if he so chooses, without the hindrance of some specified obstacle(s). It then proposes the following specification of global equality of opportunity: all equally talented and motivated persons who participate in the present global economic order should have a roughly equal chance to benefit from this order if they so choose, irrespective of the society to which they belong. Correlatively, it suggests that efforts should be made by the international community to enable persons who participate in the global economic order to fairly represent their interests in global economic negotiations and to acquire the qualifications they need to carve for themselves a place in the global economy. The final section examines the idea of 'morally arbitrary factors' more closely and argues that, as far as the ideal of equality of opportunity is concerned, a 'morally arbitrary feature of a person' is not necessarily a feature for which this person cannot be held responsible, but is first and foremost a feature that is of no relevance given the nature of the goal of the stipulated opportunity. It then suggests that as far as the goal under consideration is to benefit from the global economic order, it is reasonable to hold that agents' talents may, but their place of residence should not, be allowed to affect their chance of success.

The Question of Cross-Cultural Metric

The standard definition of global equality of opportunity holds that persons of similar talent and motivation should enjoy equal opportunities, no matter to which society they belong. As Simon Caney puts it, 'it is unfair if some have worse opportunities because of their national or civic identity;' global equality of opportunity requires 'that persons do not have worse opportunities because of their nationality' or, positively, that persons 'have the same opportunity to achieve a position independently of what nation or state or class or religion or ethnic group they belong to' (Caney 2001: 113–14).

But in spite of its apparent simplicity, this basic statement raises a number of difficult questions. To begin with, the requirement of equality with regard to opportunity sets can be interpreted in several ways: it might be taken to mean that all equally talented and motivated persons in the world should have exactly the same opportunities, or that they should all have equally valuable opportunities, which in turn raises the question of how the value of different opportunity sets can be measured and compared. It might further be questioned whether egalitarian principles of justice may properly be applied to all persons in the world: given that not all societies endorse a conception of the person as equal, implementing a principle of global equality of opportunity might amount to illegitimately imposing a Western conception of justice on all societies. And finally, it might not be obvious that a person's place of residence should not be allowed to affect the nature and the extent of her opportunities: since the policy decisions and the social practices adopted within a domestic society must inevitably affect the opportunities available to its members, it seems that global equality of opportunity cannot but undermine the principles of political responsibility and self-determination. This section concentrates on the first issue and aims at clarifying how the phrase 'global equality of opportunity' is best understood. The second and third questions will be addressed in the next sections.

A first way of interpreting the principle of equal opportunity is to say that persons of similar talent and motivation should have identical opportunity sets, that is, that they should have a roughly equal chance to attain exactly the same set of advantages. In this vein, Darrell Moellendorf tells us that '[i]f equality of opportunity were realized, a child growing up in rural Mozambique would be statistically as likely as the child of a senior executive at a Swiss bank to reach the position of the latter's parent' (Moellendorf 2002: 49).

It has been objected to Moellendorf that his benchmark is too culturally specific, indeed ethnocentric (Brock 2005: 347). He would choose the positions valued in his preferred society, and then argue that all equally talented and motivated persons should have an equal chance of attaining them. In so doing, he would illegitimately favour the values esteemed in his own society over those esteemed in other societies.

Yet, it is questionable whether Moellendorf's approach to global equality of opportunity can be accused of cultural insensitivity, at least in the way just described. If Moellendorf indeed conceives of equal opportunities as identical opportunities, then he could just as well modify his example as follows: 'If equality of opportunity were realized, a child growing up in Switzerland would be statistically as likely as the child of a tribe's chief in rural Mozambique to reach the position of the latter's parent'. For to say that members of different societies should have identical opportunity sets is to say that they should all have an equal chance to enjoy exactly the same advantages, no matter of how these advantages are valued in their respective society.

This clarification should by the same token make us aware of the unrealistic and impracticable character of Moellendorf's approach. The real problem with it is that it is simply impossible to provide all persons with identical opportunity sets. And there are several explanations for this state of affairs. There can be biological reasons why a person is denied a chance to be, to have or to do something: not all persons can, for instance, biologically bear children. There can also be logical reasons for this: the opportunity to become a protestant minister demands, for instance, the surrender of the opportunity to become a catholic priest (Green 1989: 12). Some positions presuppose a particular citizenship (e.g. President of the United States) or, as becomes apparent in the reformulation of Moellendorf's example, a particular ethnic belonging (e.g. tribe chief). And finally, because different societies hold different cultural values, it is unlikely that all equally talented and motivated persons will find it equally easy — and thus have a roughly equal chance — to attain the same advantages. As Bernard Boxill (1987: 148–9) nicely illustrates: The bright girl in New York is inevitably going to have better prospects of becoming a businesswoman in New York than the equally bright boy in Hindu society; and he will have better prospects than she of becoming a priest in his society. This is only partly because the education needed to be a success in business in New York is different from, and probably incompatible with, the education needed to be a priest in Hindu society. Even if schools offering both educations existed in both societies, the ethos of each society would ensure that the inequality remained.

In view of these difficulties, it may be tempting to interpret the principle of equal opportunity as requiring that persons of similar talent and motivation be given equivalent opportunity sets, that is, equally valuable (though different) clusters of opportunities. This is the option favoured by Simon Caney (2001: 121): 'global equality of opportunity requires that people of equal talent have access to positions of an equal standard of living [...] equality of opportunity does not require equal opportunity to identical tokens; it can be met where people have equal opportunity to positions of the same type'. Capturing the equality of opportunity sets in terms of the standard of living they enable to attain allows Caney to avoid Moellendorf's problem, since it is in principle possible for people of different cultures, races, ages, genders, societies, nationalities, and so forth to enjoy equivalent standards of living without their having a roughly equal chance, or even a chance, to achieve the same particular advantages.

But in order to decide whether different opportunity sets are equivalent, we must at least be able to compare the standards of living to which they give access. What Caney must show, therefore, is that the vast cultural diversity across domestic societies (and in particular, the different values and scales of values conveyed by the cultural traditions of each domestic society) does not prevent us from constructing a shared standard for assessing and comparing the standards of living attainable by members of different societies.

Caney does not elude this issue and proposes to solve this problem of metric by interpreting persons' standard of living in terms of Martha Nussbaum and Amartya Sen's 'capabilities', that is, in terms of capacities to function in various ways deemed to be objectively important (Nussbaum 1992; Sen 1993). The underlying idea is that some human functionings (e.g. being well-fed, being free from avoidable disease, taking part in the life of the community) are valued in all cultures — even if their concrete realization may vary from one culture to another — and that reducing opportunities to a common denominator of capabilities to enjoy these functionings enables us to make intercultural comparisons. More specifically, persons of similar talent and motivation are said to enjoy equal opportunity sets when they have an equal chance to attain positions enabling a similar achievement of functionings.

However, things are maybe not so straightforward. To begin with, the fact that a same functioning (e.g. being educated) will be realized differently in different societies can make it difficult to decide whether members of one society enjoy a worse, equal, or better opportunity to achieve it than members of another society. Furthermore, given that societies vary not only in respect to the meaning they give to functionings, but also in respect to the hierarchy they establish among functionings, it may be still more difficult to decide whether members of different societies enjoy better, equal, or worse opportunities in an overall sense. As Miller (2005: 64) emphasizes, the problem identified here is not a technical problem of measurement: it is not that we lack the data that would enable us to compare societies in terms of the opportunities they provide for work, leisure, mobility, and so forth. It is essentially the problem of saying what equality of opportunity means in a culturally plural world in which different societies will construct goods in different ways and also rank them in different ways.

This scepticism is echoed by Boxill (1987: 148), who argues that the main obstacle to global equality of opportunity is 'that the world is made up of different societies with different cultures and different standards of success'.

Miller and Boxill's rejection of the ideal of global equality of opportunity relies on what they take to be a crucial disanalogy between the domestic and the global cases. As just suggested, equality of opportunity is assumed to make sense only between persons who share enough cultural understandings with respect to the meaning and relative importance of social goods. And according to them, while this condition can be met within a domestic society, even within a modern pluralistic society, cultural differences are too great at the global level to permit the comparison and a fortiori the equalization of individual opportunity sets. Hence their conclusion that global equality of opportunity cannot be implemented unless cultural diversity is abolished worldwide.⁵

In view of these difficulties, a common strategy has been to focus the attention on the provision of what can be called a 'decent set of opportunities' (Brock 2005: 350) or on the satisfaction of an 'international decency standard' (Miller 2004: 130–1). The leading idea has been that instead of trying to provide equally talented and motivated persons with equal opportunity sets (whatever is meant by 'equal'), we should ensure that some standard opportunities or capabilities — that is, opportunities or capabilities everyone can be assumed to have a reason to want whatever his goal(s) in life — are present in everybody's set. So Brock focuses on the removal of 'barriers to developing a range of skills and capacities that would be useful no matter what goals people set themselves' (e.g. secure access to food, clean water, basic health care, work, and elementary education), while Miller acknowledges an obligation to help people who cannot engage in the range of core human activities that we find recurring across culturally varied societies, given the conditions prevailing in their society (e.g. working, playing, raising families, etc.). The point is that by giving up global egalitarian commitments, both positions can avoid the above-mentioned problem of metric, since there is no need to compare different opportunity sets anymore.

But the price to pay is that profound inequalities are allowed to persist that can in some circumstances translate into objectionable disadvantages. In fact, this holds true not only for Brock and Miller's decency views, but also for Caney's conception of global equality of opportunity. To see this, a first point to note is that even if it could be shown that two persons have access to positions of an equivalent standard of living within their respective society (which is contested by Miller), this would still not establish that they are equal from the point of view of the resources that they have at their disposal. As Sen himself points out, achieving a same functioning may require different amounts of resources depending on the society to which one belongs: 'For instance, to be able to "appear in public without shame" may require higher standards of clothing and other visible consumption in a richer society than in a poorer one' (Sen 1999: 71).

By themselves, inequalities of resources may be unobjectionable. But a serious problem must arise once members of different societies start interacting with one another, and more particularly, once they start competing against one another for the same advantages, be it as a result of migration or of trade liberalization. For, from then on, persons who initially enjoyed access to positions of an equivalent standard of living may end up with very unequal opportunities, depending on the amount and kind of resources needed and made available in the society from which they come. To take an example, consider the kind of educational resources available in Western societies (e.g. compulsory secondary education, broad access to books, to computers, and to the media): it may be expected that members of these societies will come to enjoy a better opportunity to attain positions that require superior training and intellectual skills than members of developing societies. The same considerations apply to the present global economic order as a whole. Global markets exist, which function as sites of competition for social and economic advantages that most people are presumed to want, but that cannot be attained by all those who desire them (e.g. labour, goods and services, and technology and capital). As in any competition, there are winners and losers: some will enjoy the advantages at stake, others will either not enjoy them at all or only in a much more limited way than others. To quote the Human Development Report 2005: Success in global markets depends increasingly on the development of industrial capabilities. In a knowledge-based global economy cheap labour and exports of primary commodities or simple assembled goods are insufficient to support rising living standards. Climbing the value chain depends on managing the processes of adapting and of improving new technologies (United Nations Development Programme 2005: 120–1).

Here too, it is to be expected that members of wealthy societies will have a better opportunity to benefit from global market competitions than members of poor societies. Likewise with negotiations setting the rules of the global economic game: gross inequalities in resources are likely to translate into inequalities in bargaining power, which cause the interests of the stronger parties to be weighed much more heavily than those of weaker parties, thereby reinforcing initial inequalities.

If so, there is an important sense in which gross inequalities of resources worldwide may appear objectionable: they are likely to undermine the fairness of global market competitions and economic negotiations.⁶ Now, the fairness of competitions (including the processes by which their rules are defined) is the subject of a very specific kind of equality of opportunity: from the perspective of 'competitive equality of opportunity', it is indeed unfair if some competitors are prevented from winning the prize at stake in the competition because of morally arbitrary factors. Put broadly, competitive equality of opportunity demands that equally talented and motivated competitors should have a roughly equal chance to attain the advantage(s) at stake in the competition, and is said to obtain when all manipulable factors that affect their chance of success are equalized.⁷ This mainly includes the removal of unjustified formal barriers and the redress of background disadvantages that impede their capacity to acquire the talents that predict success in the competition.⁸

To close this section, it is important to note that that the adoption of a competitive version of equality of opportunity allows us (at least to a certain extent) to avoid the problem of metric raised by Miller and by Boxill, without having to abandon the kind of egalitarian commitments just mentioned and so without having to leave the potential unfairness of our 'globalizing world' completely unaddressed. Since competitive equality of opportunity is not concerned with persons' prospects for success in life as human beings (e.g. their well-being, their standard of living), but only with their chance of success in a given competition as competitors, there is no need to equalize and thus to find a cross-cultural standard of comparison for their complete opportunity sets or life chances either. What must be done is only to ensure that all persons who participate in the global marketplace have a fair chance to attain the advantages at stake, without the hindrance of morally arbitrary factors. What this might entail at the global level is the topic of the next section.

Global Equality of Opportunity Specified

To better understand what a competitive version of global equality of opportunity might represent, we need to return to the concept of equality of opportunity itself. What do people have in mind when they talk about an opportunity? Following Peter Westen (1985: 842), I suggest that every opportunity has a triadic structure: it is a relationship of some agent or class of agents X to some goal or set of goals Y with respect to some obstacle or set of obstacles Z. More specifically, a person possesses an opportunity when she has a chance — that is, less than a guarantee, but more than a mere possibility⁹ — to attain some specified goal(s) if he/she so chooses, without the hindrance of some specified obstacle(s).

To say further that two persons have 'equal' opportunities presupposes that their respective opportunities have been measured, compared with each other and on this basis found equivalent. Here again, we encounter the difficulty of finding an appropriate standard of comparison. But I think that Westen, by narrowing the range and the content of the opportunities to be compared, manages to overcome this problem: 'the standard [of measure] can be any stipulated opportunity — any "specification" of the three variable terms of which opportunities consists' (Westen 1985: 844). More specifically, different persons enjoy an equal opportunity when they all have a chance to attain the same specified goal(s) without the hindrance of the same specified obstacle(s).

On this view then, people can face different obstacles and still enjoy an equal opportunity to attain a specified goal, provided only that they are free from the same specified obstacle(s). And conversely, people can be free from the same obstacles and yet lack an equal opportunity to attain a specified goal, if the obstacles they are free from are not specified as relevant obstacles. Although Westen does not discuss that point, it seems that the same line of reasoning can be applied to the goal(s) of the opportunity under consideration. So, it seems plausible to say that people can have a chance to attain different sets of goals without the hindrance of the same specified obstacles and still enjoy equality of opportunity, provided at least that some of the goal(s) they have a chance to attain are specified as relevant goal(s). Here again, we find the idea that in order for equality of opportunity to obtain between two persons, it is not necessary that they both have equal opportunity sets, that is, a roughly equal chance to attain an identical or an equivalent set of goals irrespective of all the same obstacles. It suffices that they have a chance to attain some goal(s) that we specify as relevant, without the hindrance of some obstacle(s) that we specify as relevant.

Perhaps the most important lesson to be drawn from Westen's account is that statements of global equality of opportunity are incomplete if one of the terms of the above-described triadic relationship is left indeterminate. A statement of the form 'everyone everywhere must enjoy the same bundle of opportunities' (Miller 2004: 125) might suggest a wide range of possible prescriptions — from the most acceptable to the most controversial — because it specifies neither the goal(s) of the prescribed opportunities (everyone everywhere must have a chance to do, to have or to be what?) nor the obstacle(s) from which the specified agents should be free (no one must be hindered by the colour of her skin, or by her language, or by her place of birth?).

In order for a conception of global equality of opportunity to make sense, it must specify (a) to what agent(s) the opportunity belongs; (b) toward what goal(s) the opportunity is directed; and (c) what obstacle(s) the opportunity is intended to remove. Each of these dimensions is shaped by an independent set of normative considerations, which may of course be subject to dispute. To begin with, we can think of many kinds of goals (or advantages) whose attainment seems desirable in most parts of the world: health, jobs, knowledge, political influence, income, etc. Note, however, that a competitive version of equality of opportunity will admit only those desirable goals that can be regarded as proper objects of competitions. This rules out the sorts of goals that we think ought to be available to anybody who wishes to achieve them and which may therefore not be striven for in competition with others (e.g. being well-nourished). Similarly, the fact that equality of opportunity is extended globally does not necessarily mean that it applies to any agent whatever. Since the proposed conception of global equality of opportunity relies on the notion of 'participation in the global economic order', non-participants (if any) are excluded from its scope. Now, different classes of participants might be identified as appropriate agents of global equality of opportunity. The literature on global justice has concentrated on the dichotomy between societies (whether states, nations, or peoples) and individuals, but we could as well highlight the condition of groups like children, women, steel workers, farmers, students, trade unions, firms, etc. And finally, obstacles can range from a pure and simple exclusion from competition to the fact of having to bear extra burdens compared to other competitors. In that respect, the crucial question will be to determine what burden is to be considered an obstacle that 'must' be removed, that is, an obstacle that is not allowed to affect the chance that the specified agent(s) of the opportunity reach the specified goal(s) of the opportunity.

Let us now examine more closely how a particular version of global equality of opportunity can be constructed — that is, how a specific content can be given to the agent(s), the goal(s), and the obstacle(s) of global equality of opportunity — taking into account the existence of a global economic order and reflecting on the conditions of fairness it must satisfy in order for all the persons who participate in it to be treated as equal participants. As might be expected, the critical issue will be to determine what may properly be regarded as 'relevant' goal(s), agent(s), and obstacle(s) of a global version of equality of opportunity.

Goal(s)

Certainly, one's life cannot be assimilated to a competitive game. As Nozick (1974: 235) famously put it: 'life is not a race in which we all compete for a prize which someone has established; there is no unified race, with some person judging swiftness'. But as I have already indicated, competitive equality of opportunity is not concerned with a person's life as a whole, but only with her participation in a given competition. And nobody will contest that today there is an increasingly global market for labour, goods, ideas, and capital, which creates valuable economic opportunities for millions of people who sell their labour and their products, develop ideas and raise capital. Nor will it be contested that the existence of global market competitions has a deep impact on individual life prospects: the vast majority of persons can be considered 'participants' in the global economic order, even though their participation may be more 'passive' (in the sense of 'being affected by' this order) than 'active' (in the sense of 'contributing to' and 'gaining from' this order). What is more, there is a system of rules and practices governing the global economic game, and these rules are the product of human decisions.

As Sen (2002) points out, the central question is not whether to use global markets, but rather how to use global markets in a way that is fair. There is no doubt that persons, whatever the society to which they belong, have much to gain from global trade, the diffusion of knowledge and technology, the international division of labour, foreign direct investments, and so forth. Global economic and scientific interactions have always been and will continue to be an important source of economic growth, and incidentally, a major driving force for the advancement of human welfare. Yet, the benefits and costs of the global economic integration are not necessarily fairly distributed among its participants, be it across or within societies.

In its most general sense, a competition is fair when no competitor is barred from winning the prize at stake because of morally arbitrary considerations. We can distinguish between two chief types of fairness to which competitions must conform: (a) procedural fairness, which reflects a concern with the basic rules governing a competition, and (b) background fairness, which is concerned with competitors' starting positions. As far as the global economy is concerned, these two kinds of fairness dictate the removal of, respectively, formal and substantive market access restrictions that arbitrarily prevent some participants in the global economic order from carving a niche for themselves in the global marketplace.

Global procedural fairness

Procedural fairness is essentially a principle of formal equality of opportunity. It obtains when there are no legal or quasi-legal restrictions on agents of similar talent and motivation having an equal chance to attain the goods at stake in a competition (Daniels 1978: 217; Joseph 1980: 394; Rawls 1999a: 62; Kivimäki 2001; Jacobs 2004: 15–16). We can think of different kinds of procedural unfairness: some agents or classes of agents can be arbitrarily excluded from a decision-making process, which affects their interests; some parties to a decision-making process can be arbitrarily denied as much voice as other parties; the interests of some parties can be arbitrarily underrepresented, those of others arbitrarily overrepresented, and so forth (Beitz 2001: 107).

In the global economic context, the idea of procedural unfairness is often associated with the imposition of barriers to the free movement of 'certain' goods and 'certain' persons, and more specifically, with the fact that developing countries face obstacles in selling their agricultural and manufactured products in developed countries (while these are sectors of great importance to them and in which they have a comparative advantage) or that their unskilled workers face obstacles in migrating to developed countries (while this is a factor of production that they possess in abundance and in which in which they have a comparative advantage) (The World Bank 2005: 208; United Nations Development Programme 2005: 115).

The removal of these barriers can be supported on different grounds: libertarians may regard it as the direct implication of persons' absolute right to negative freedom,¹⁰ while utilitarians may think of it as an efficient way to use talents and skills.¹¹ From a fairness perspective, such removal is mainly motivated by a concern that global economic rules give due weight to the interests of all the parties concerned. More specifically, these rules are said to be unfair when they do not represent the interests of all parties concerned as equal parties and when this lack of representativeness is attributable to arbitrary factors like differences in bargaining power (Miller 1999: 204–08; Beitz 2001: 107–09).¹²

Part of the unfairness of existing global economic rules can be explained by the conditional nature of global agreements, and more specifically, by the norm of specific or strict reciprocity on which these agreements are built in practice (Kapstein 2006: 59–65). Put in a nutshell, the norm of specific reciprocity entails that the concessions made by each of the parties in an agreement should be of roughly equivalent value, in a tit-for-tat fashion. In the context of international trade negotiations, this often translates into a 'setting apart' of those countries that have little to offer in terms of market access that is of interest to other countries. Since the domestic costs of equivalent exchange will, on a relative basis, be higher to poor countries than to rich countries, the former may be expected to be given relatively little voice in the negotiation process and not to obtain liberalization in the sectors of most importance to them. That is, they may be expected not to influence and to benefit from the design of the international trade regime as much as do great powers with large economic markets.

The challenge will then be to see to it that the outcome of global economic negotiations does not reflect differences in bargaining power or that the capacity of the parties to represent their interests is not impaired by a lack of wealth and power. The most straightforward way of preventing global economic rules from being shaped by power relationships is either to strengthen the bargaining position of the weaker parties or to restrict the freedom of action of the stronger parties. With respect to the former option, a first thing to note is that, although all parties are formally given an equal voice in the WTO rule-setting process, some of them may still find it difficult to follow negotiations either because of a lack of expertise or because they simply cannot afford being present at the WTO headquarters in Geneva (Pogge 2005; The World Bank 2005: 66–7, 212; United Nations Development Programme 2005: 146–7). Therefore, if all societies that are parties to international trade negotiations must have the capacity to influence the rules-setting process, it is necessary that they all be given the capacity for understanding and analysing the negotiations taking place, and this in turn requires that they all be provided with qualified personnel, having knowledge of international trade law and of multilateral trading systems. But improving the parties' 'bargaining skills' is of course not enough to equalize their 'bargaining power'. And in this regard, the suggestion that the weaker parties' position is to be reinforced is hardly realistic as it would require transfers of resources to eliminate not only societies' absolute deprivation, but also their relative vulnerability (Miller 1999: 208).

In order to neutralize the unfairness that might result from differences in bargaining power, the second option looks more promising. Now, saying that the freedom of action of the great powers should be constrained is another way of saying that global economic negotiations should be sensitive to each party's particular situation — for example, its level of resources, its capacities, and that the weaker parties should be provided with benefits that they cannot fully reciprocate. At this point, the obvious question is why the great powers would agree to constrain their freedom in this way. A possible answer has been offered by Ethan Kapstein (2006: 31–8). According to him, given the uncertainty of world politics, integrating weaker societies into the international trade regime might be a way to ensure that potentially dissatisfied societies do not seek to shake the existing world order, and might thus redound to the benefit of the stronger parties in the long run. Put another way: great powers might exchange the principle of 'strict' reciprocity for that of 'diffuse' reciprocity and renounce some specific advantages now in order to obtain some more general advantages in the future, such as a robust and stable world economic order. Now, whether the great powers can really be expected to open their markets or to stop subsidizing their farmers in anticipation of a future return is a controversial issue. But the point is that something like a principle of diffuse reciprocity must be at work if procedural fairness is to be achieved in the global economic order.

Global background fairness

Dismantling barriers to global markets constitutes only part of the strategy toward a fairer world economic order, in the same way that removing racial and sexual discrimination in the allocation of jobs within domestic societies is only a partial scheme too. Even if global market competitions were procedurally fair, this would not guarantee that all equally talented and motivated competitors have an equal chance to attain the newly opened advantages. The reason for this is that success in a competition depends not only on the way the rules of the game are defined, but also on competitors' respective endowments. Exploiting the new possibilities generated by the global economy requires more than open markets: it also require a capacity to respond to market openings. Yet, as already pointed out (in the first section), inequalities in the capacity to use and to benefit from the global marketplace can persist because of initial differences in wealth, social and cultural background, health and educational infrastructure, geographic location, and so forth.

By way of illustration, consider the debate that has been going on for several years regarding the potential impact of fairer rules for the functioning of global markets on the condition of the poor. While there is considerable controversy as to whether the removal of barriers would lift up many people out of absolute poverty or not, most research tends to agree that its effects will vary with agents' endowments. More concretely, it is widely assumed that gains would accrue mostly to agents already significantly integrated in global markets; by contrast, agents that are disconnected from markets, technology, and information would benefit little or not at all (The World Bank 2005: 211). This hypothesis has mainly been advanced with regard to countries — the greatest beneficiaries being assumed to include Brazil, China and India, and the smallest ones sub-Saharan countries — but it can plausibly be extended to individuals.

Background fairness or substantive equality of opportunity demands that participants enter the competition on roughly equal terms. More specifically, a level playing field must be established, where competitors are compensated for all morally arbitrary factors that prevent them from acquiring the qualifications needed to attain the advantages at stake, in the case that concerns us here, to capture a share of the benefits generated by the global economy. Presumably, creating the conditions under which societies can build and maintain the endowments of their individual members entails a massive investment in human development, trade-related infrastructures and governance structures, social safety nets, trainings, etc. Here too, the most advantaged agents can be said to bear a special responsibility for promoting the less advantaged agents' inclusion in the global economy. For it is they who have the capacity for redistributing resources, whether by transferring resources to specialized agencies or by modifying the global economic rules and granting preferential treatment.

Agents

Whose opportunity to influence and to benefit from the global economic order must exactly be equalized? In the previous section, considerable attention was paid to domestic societies, which might suggest that the title of this article — 'global' equality of opportunity — has misled us or that the whole discussion has been biased from the start. Now, advocating a principle of equality of opportunity among societies rather than among persons seems to have some advantages. First, it can accommodate the fact that national governments are still the main and the most efficient actors of the international scene: it is still they that negotiate and sign international agreements, impose tariffs and quotas, and have the power to make profound and large-scale reforms in order to achieve justice at both the domestic and world level. Second, it can accommodate the fact that domestic societies vary greatly in their understanding of the person and of social goods, and consequently endorse very different conceptions of domestic justice. In this regard, it is worth remembering that one of the arguments used by John Rawls to invalidate principles of (egalitarian) distributive justice at the global level is that (a) these principles presuppose a liberal conception of the person as free and equal, which is not endorsed by all well-ordered peoples, and that (b) liberalism's own principle of toleration condemns the imposition of liberal values on well-ordered peoples that reject them. Hence, the importance of tolerating alternative reasonable ways of ordering society excludes the pursuit of global equality of opportunity.

Note that in order to discredit Rawls's claim, it is not sufficient to show that the principle of equal opportunity can be implemented among societies without compromising toleration. As Allen Buchanan (2000: 710) has rightly argued, nothing in Rawls's The Law of Peoples prevents the application of egalitarian principles among societies, since the principle of equality among peoples as peoples is one of its basic principles. Put another way, inter-national equality of opportunity cannot be characterized as ethnocentric for, as Rawls himself put it, societies 'cannot argue that being in a relation of equality with other peoples is a western idea! In what other relation can a people and its regime reasonably expect to stand?' (Rawls 1999b: 122) Yet, inter-national equality of opportunity is far from guaranteeing global equality of opportunity. It places no particular demands on societies to promote the capacity of their individual members to actively participate in the global economic order, except insofar as this is necessary to preserve social cohesion and international stability. It can in principle be achieved among societies in which birth prohibits some groups of persons from pursuing higher education or paid employment.¹³

Now, is it a problem if we strive to achieve equality of opportunity among domestic societies, and not among those who shape the world economy by their day-to-day interactions, namely, individual producers, workers, entrepreneurs, consumers, researchers, etc.? Suppose that significant steps are taken at the international level to equalize societies' capacity to benefit from the global economic order, but that a considerable number of individuals are still lacking or even denied a like capacity: can this activate some duties on the part of other societies and their members?¹⁴

It seems to me that if we truly believe that there is something unfair about a competition in which some competitors are barred from attaining the advantages at stake because of morally arbitrary considerations, then we cannot but acknowledge that we have at least 'some' responsibility for individuals' capacity to benefit from the global economic order, even if they belong to another society. At a minimum, we must acknowledge a duty to assist those societies which are willing to enhance the capacity of their individual members to benefit from the global economic order, but are lacking the material means to do so.

At this point it is worth examining the alleged advantages of an inter-national conception of equality of opportunity more closely: are 'global' principles of equality of opportunity doomed to be inefficient and intolerant? Concerning the question of efficiency, nothing prevents us from falling back on the existing states system and taking advantage of its institutional capacities. But states would then be accorded a special status not because they have inherent moral significance, but only insofar as they can be effective instruments for achieving equality of opportunity among persons worldwide. That is to say, their privileged status would have a cosmopolitan justification; hence, outsiders would not be indifferent to the way the resources made available by their transfers or preferential measures are distributed within societies.¹⁵

This immediately raises the question of toleration: how is a commitment to global equality of opportunity to be conciliated with liberalism's demand to respect other reasonable, yet inegalitarian, ways of organizing domestic society? On the one hand, it is important to realize that the world, while it can contain both egalitarian and non-egalitarian domestic societies, can itself be structured and organized in only one way (Pogge 1994: 217). At the world level, a choice must be made between egalitarian and non-egalitarian principles, and if we truly believe that persons of similar talent and motivation should have an equal chance to benefit from the global economy, then we should support the appropriate global economic order. That is, we should support global redistributive mechanisms aimed at promoting persons' active participation in the global economy. But on the other hand, it must be recalled that the concept of opportunity refers to a given agent's chance to achieve a given goal without the hindrance of a given obstacle if he so chooses. This means that global equality of opportunity aims only at granting persons a chance which they remain free not to seize (Caney 2001: 130). In particular, persons who do not share the egalitarian values underpinning the proposed principle of global equality of opportunity remain free to disregard any initiatives taken to enhance their inclusion in the global marketplace.

Obstacles

To round off this article, I would like to consider two kinds of objection that might be raised to the proposed version of global equality of opportunity. Both relate to how the phrase 'morally arbitrary factors' is best understood: the first holds that one's place of residence is not a morally arbitrary factor and that persons should therefore not be compensated for the global competitive disadvantages they may suffer as a result of their living in a particular society; the second holds that talents are undeserved and should therefore not be allowed to affect persons' access to advantages, including their success in the global economy. This section aims at showing that (1) as far as competitive equality of opportunity is concerned, whether an attribute of a person is to be considered morally arbitrary or not depends on the nature of the goal(s) at stake (not on the person's responsibility for possessing this attribute) and (2) insofar as the goal at stake is to benefit from the global economic order, it is reasonable to hold that persons' talents may, but their place of residence should not, be allowed to affect their chance of success.

At the core of the ideal of equality of opportunity is the compelling intuition that access to advantages must not be governed by morally arbitrary factors. Accordingly, it should come as no surprise that the most usual justification for extending this ideal to the global level consists in saying that persons' place of residence is a morally arbitrary feature and should therefore not be allowed to influence their attainment of advantages. To quote Caney (2001: 115) again: 'underpinning our commitment to equality of opportunity is the deep conviction that it is unfair if someone enjoys worse opportunities because of his or her cultural identity. [...] This deep conviction implies, however, that we should also object if some people have worse opportunities because of their nationality or civic identity'. The same idea is found in The World Bank's World Development Report 2006: 'The fact that country of birth is a key determinant of people's opportunities runs counter our view of equity–that is, that people should enjoy the same opportunities regardless of their background, including where they are born' (The World Bank 2005: 206).

The issue that needs to be investigated is thus whether and, if yes, why persons' place of residence should be seen as a morally arbitrary feature. Why, after all, would it be objectionable that persons have an unequal opportunity to succeed in the global economy because of their being members of different societies? Is it not precisely one of the main tasks of governments to decide which opportunities will be provided to their people and at what level, taking into account the cultural values they share? Should we not follow David Miller and admit that a national community constitutes a set of special relationships that are morally relevant and that may therefore legitimize the preferential treatment of one's co-nationals and thereby the rise of international inequalities? The underlying reasoning is that decisions made by domestic societies must inevitably affect the resources and opportunities that will be available to their present and future people, and that endeavouring to achieve global equality of opportunity — even in the form proposed here — would amount to undermining both the principle of political responsibility and the principle of political self-determination. Not only would domestic societies not have to bear the cost of their decisions (they would in fact be compensated for their 'bad' choices), but they would also be denied any control over their own fate, since the differential effects of their policy decisions on the opportunities of their members would have to be systematically neutralized.

For global egalitarians like Caney and Moellendorf, the primary reason why a person's place of residence is morally arbitrary and thus not allowed to affect her access to advantages is that it is undeserved. So Moellendorf argues that the parties in a cosmopolitan original position would opt for fair equality of opportunity, and specifies that 'The underlying idea is that one cannot claim to deserve things such as place of birth and education, race, or parents' privilege. Hence, these things should not be the basis of a distribution' (Moellendorf 2002: 79). In the same vein, Caney (2001: 115) claims that 'people should not be penalized because of the vagaries of happenstance, and their fortunes should not be set by factors like nationality or citizenship.' In keeping with the tradition of 'responsibility-sensitive egalitarianism', these authors assume that persons should be compensated for all those circumstances for which they cannot be held responsible and which negatively affect their access to advantages. What they add, or at least presuppose, is that persons are not responsible for those aspects of their place of residence which hamper their ability to attain advantages.

The question which immediately arises at this point is why proponents of the principle of global equality of opportunity, including Caney and Moellendorf themselves, then restrict the scope of this principle to equally talented and motivated persons (Pogge 1994: 196; Caney 2001: 121; Moellendorf 2002: 49, 79–80).¹⁶ Assuming that a feature of a person is morally arbitrary whenever it is undeserved, why should her possession of talents, but not her place of residence, be allowed to affect her access to advantages?

If supporters of global equality of opportunity want to hold on to both the thesis of 'responsibility-sensitive egalitarianism' and the thesis that one's talents (but not one's place of residence) may affect one's access to advantages, then they must provide an argument to the effect that one's talents (but not one's place of residence) are deserved. And one of the most plausible strategies in this regard is probably to say that the development and maintenance of talents most often presuppose a certain amount of voluntary choice and effort, and that the distribution of talents among persons is therefore not always or wholly the result of a natural lottery (Green 1989: 14). More particularly, it might be pointed out that in a world economy as complex and highly specialized as ours, it is very unlikely that competitors will owe their success only to their possession of natural endowments, without having taken some risks, made some sacrifices and spent a considerable amount of time and energy. And if the talents needed for success in the global economy are for a large part acquired, then the 'winners' of economic globalization cannot be said to be benefiting on morally arbitrary grounds.

Yet, however attractive and sound this argument may be, it faces two difficulties. The first is that exactly the same line of reasoning can be used to support the opposite conclusion. So it might be argued that because talents are always to some degree the product of fortunate circumstances, allowing them to affect the distribution of social and economic resources amounts to rewarding their possessors for features that are not due to any credit on their part. On this view, there is no reason why unequally talented persons should not have equal access to various advantages (why, for instance, the fruits of the exercise of talents should not be redistributed).

The second difficulty is that there may also be some sense in which persons are responsible for their place of residence, or more precisely, for the opportunities that are available to them as a result of their being members of a particular society. An argument of this type is found in David Miller (2004). According to him, the fact that members of different societies have different or unequal opportunities can often be traced back to political choices or social practices in which they are involved or which simply reflect the beliefs and values they share and regard as part of their identity. His point is that since persons may to some degree be held responsible for the political decisions in which they have a voice as well as for the beliefs and values articulated in the public culture of their society, they may also be held responsible for their effects on the bundle of opportunities with which they end up. Hence, they should not be compensated for the disadvantages they might suffer as a result, even if these disadvantages affect their chance to succeed in the global economy.

It appears then that the question of exactly where personal responsibility starts and where it ends remains largely unsettled. But this is not the only weakness of the 'deservingness'-criterion. An independent and more fundamental question is indeed whether persons' 'morally arbitrary features' can so simply be equated with their 'undeserved features'. One can think of many cases in which undeserved attributes like race, religion, class and gender may legitimately influence a person's chance to enjoy some kinds of advantages. In order, for instance, to qualify for a post of obstetrician in a hospital with a high proportion of Muslim patients, applicants may be required to be female. The same holds for persons' talents and place of residence: even if we assume that they are on balance undeserved, we may sometimes allow them to affect the allocation of social and economic advantages. Having the requisite physical skills for being a professional football player can hardly be discounted as a good reason for being appointed in a national football team. Similarly, few will disagree that living in France may be considered a necessary condition for qualifying for the presidency of France.

It is worth noting that the reverse is true too: it is doubtful that those aspects of a person's life which result from her voluntary choices should always be allowed to affect her access to advantages. In particular, we might consider that a person's talents and place of residence are on balance deserved and still deny that they should always be allowed to affect her access to advantages. So, a person may have developed an exceptional talent at doing something in which no one is interested. Or she may voluntarily have settled in France, and still not qualify for the presidency of France because she lacks a sufficient command of the French language.

What all this suggests is that whether some feature of a person should be deemed a morally relevant feature and thus allowed to influence her access to advantages depends on the nature of the advantages at stake. Or put another way: whether an agent's lack of talents or place of residence should be considered an obstacle that 'must' be removed depends on the nature of the goal of the stipulated opportunity. This takes us back to an idea developed earlier: namely, that a principle of equality of opportunity cannot make sense unless its agent(s), its goals(s) and the obstacle(s) it is intended to remove are specified. Our question must then be reformulated as follows: Does the nature of the goal of the opportunity under consideration in this article — namely, to benefit from the global economic order — justify that agents' chance of success be a function of their place of residence or of their talents?

I think that the following suggestions put us on the right track. To begin with, it is very unlikely that talents might be ranged among factors not allowed to count in the determination of persons' success in free market competitions, since economic globalization finds its raison d'être precisely in the gains that might result from the exploitation of their relative differences in productive resources (i.e. from the exploitation of their comparative advantages). That is to say, it belongs to the nature of economic globalization that those who can support market pressures by being more competitive than others or by differentiating themselves from others are also more likely to be appreciated by the global buying public and to carve for themselves a place within the global economy. This does not rule out that compensation might be provided to those who — in spite of the implementation of effective measures for ensuring the procedural and background fairness of the global economic order — remain unable to develop the talents that predict success in the global economy and are therefore still denied the fruits of global economic contacts. Perhaps the requirement that all persons who participate in the global economy should be treated as equal participants entails the joint adoption of a kind of global Difference Principle. But this question is beyond the scope of this article, whose main aim is to determine whether a principle of global equality of opportunity can make sense. It is also critical to emphasize that, even if the proposed principle of global opportunity does not consider talents as morally arbitrary factors, its implications in terms of human development are already considerable.

On the other side, the talents likely to predict success in the global economy are sufficiently residency-blind that agents' place of residence may reasonably be considered a morally irrelevant factor. Since the primary criteria of success include competitiveness, differentiation, innovation, and so forth and since these skills can in principle be acquired by agents regardless of their place of residence, it is unfair to deny them formal access to the advantages at stake in the global economy on the ground that they happen to live in a society lacking sufficient bargaining power. And it is also unfair to exclude agents formally permitted access to these advantages on the ground that they happen to live in a society lacking the capacity to provide them with the resources necessary to develop and exercise the relevant talents.

Conclusion

In this article, my principal objective has been to show that a plausible conception of global equality of opportunity can be constructed, which demands that equally talented and motivated persons participating in the global economic order should have a roughly equal chance to benefit from this order if they so choose, irrespective of the society to which they belong.

I started by arguing that the current process of economic globalization generates specific relationships, which render the recognition of principles of egalitarian distributive justice necessary at the global level. I have pointed out that members of different societies are now competing against one another for the same global economic advantages (e.g. labour, goods and services, and technology and capital) and can be expected to have an extremely unequal chance to attain them, depending on the amount and kind of resources made available in their respective societies. I have supported the view that global inequalities of resources that undermine the fairness of global economic negotiations and market competitions are objectionable.

I have then noted that the fairness of competitions is the object of a specific conception of equality of opportunity — namely, 'competitive' equality of opportunity — which is immune (at least to a certain extent) to the charge that there is no cross-cultural standard for comparing opportunities. The reason is that it is not concerned with persons' prospects for success in life as human beings, but only with their chance of success in a given competition as competitors, and that it does therefore not require the equalization of their complete opportunity sets or life chances.

In line with Peter Westen, I have claimed that in order for a conception of global equality of opportunity to make sense, it must specify (a) to what agent(s) the opportunity belongs; (b) toward what goal(s) the opportunity is directed; and (c) what obstacle(s) the opportunity is intended to remove. Elaborating on the idea of a 'fair global economic order', I have then proposed the following specification:

1. The goal of global equality of opportunity is to benefit from the global economic order. While there is no doubt that global markets create valuable economic opportunities for millions of persons, the crucial question is to determine how these markets can be used in a way that is fair. I have distinguished between two kinds of fairness to which the global economic order must conform: procedural fairness and background fairness. Procedural fairness demands not only that global economic rules duly represent the interests of all the parties concerned, but also that the capacity of the parties to represent their interests in global economic negotiations not be impaired by a lack of wealth and power. To this end, a first point of importance is to improve weaker parties' 'bargaining skills' — by providing them, for instance, with qualified personnel, having knowledge of international trade law and of multilateral trading systems. But since these measures are insufficient to equalize parties' 'bargaining power', it is also necessary that stronger parties constrain their freedom of action and provide weaker parties with benefits that they cannot fully reciprocate. Now, exploiting the new possibilities generated by the global economy requires more than open markets: it also requires a capacity to respond to market openings. And the trouble is that inequalities in the capacity to use and to benefit from the global marketplace are likely to persist because of initial differences in wealth, social and cultural background, health and educational infrastructure, geographic location, and so forth. It is here that background fairness comes into play: it demands that persons who participate in the global economic order be given the capacity to acquire the talents that predict success in the global economy. This, in turn, entails the adoption of redistributive measures targeted toward human development, trade-related infrastructures and governance structures, social safety nets, trainings, etc.
2. The agents of global equality of opportunity are persons who participate in the global economic order, that is, persons who contribute, benefit or are simply affected by this order. This, of course, raises a question of efficiency and of toleration. Now, as far as efficiency is concerned, nothing prevents us from falling back on the existing states system and from taking advantage of its institutional capacities. But states' privileged status would then have a cosmopolitan justification; hence, outsiders would not be indifferent to the way the resources made available by their transfers or preferential measures are distributed within societies. This leads us directly to the toleration issue. I have noted that, since the world can be organized in only one way, our belief that persons of similar talent and motivation must have an equal chance to benefit from the global economy should lead us to support global redistributive mechanisms aimed at promoting persons' inclusion in the global economy. But given that the concept of opportunity refers to an agent's chance to achieve a given goal if he so chooses, persons would remain free to decline it.
3. Finally, the question of obstacles requires that we determine what characteristic of the person is to be considered morally arbitrary, that is, a characteristic not allowed to affect her attainment of global economic advantages. Supporters of global equality of opportunity tend to identify persons' 'morally arbitrary' features with their 'undeserved' features; yet, one can think of many cases in which undeserved features may legitimately influence a person's chance to enjoy some kinds of advantages. It rather seems that whether a feature of a person should be deemed a morally relevant feature depends on the nature of the goal at stake. And I have suggested that as far as the goal of the opportunity under consideration is to benefit from the global economic order, it is reasonable to hold that agents' talents may, but their place of residence should not, be allowed to affect their chance of success. The reason is that economic globalization finds its raison d'être in the gains that might flow from the exploitation of some talents — for example, competitiveness, innovation, differentiation, etc. — and that these talents are sufficiently residence-blind that members of different societies can reasonably be held capable to acquire and exercise them.

Notes

¹ After some hesitation, I distanced myself from the leading interpretation of global equality of opportunity in terms of persons' 'nationality' and focused instead on their 'place of residence' understood as the society in which they are living. This shift is meant to accommodate the fact that persons of the same nationality can be living in different societies and as a result enjoy extremely unequal opportunities.

² To be sure, some 'responsibility-sensitive egalitarians' call the principle they support a principle of equality of opportunity too. But in this article, I reserve the denomination 'equality of opportunity' to refer to the traditional ideal of 'advantages open to talent and motivation', and borrow from Andrew Mason (2001: 762) the denomination 'responsibility-sensitive egalitarianism' to refer to what he calls the 'new' view of equality of opportunity. Put in a nutshell, the former is desert-based and holds that access to advantages must as far as possible depend on attributes for which people can be held responsible, while the latter is merit-based and holds that access to advantages must depend on some 'relevant' attributes, whether people can be held responsible for them or not. On the difference between desert and merit, see Lucas (1993) and Daniels (1978).

³ It has also been doubted whether a global principle of equality of opportunity, however morally appealing it may be, can be implemented in the absence of a world state, and if it cannot, whether the establishment of such a state is desirable (Boxill 1987: 155–68). This objection focuses on the idea that, without the institution of a government, principles of distributive justice pose a problem of mutual assurance — that is, there is no assurance that all parties will do their part because it may not be in their interest to do so or the requisite reciprocity may be lacking — and that, at the world level, such institution is likely to threat individual rights and to become tyrannical.

⁴ An exception to this is Ethan Kapstein's (2006) recent book Economic Justice in an Unfair World, which aims to present a theory of global justice that centres on equality of opportunity among states. However, although this book is explicitly about ethics, it remains primarily intended for scholars in international relations and contains no sustained discussion of the pros and cons for equality of opportunity as a normative ideal.

⁵ The same kind of conclusion follows from Michael Walzer's pluralistic account of distributive justice. On this account, 'All distributions are just or unjust relative to the social meaning of the goods at stake' (Walzer 1983: 9), and since there is no 'international community' united by shared understandings of the meaning of social goods, there can be nothing like 'international distributive justice' either. For a criticism of Walzer's view on global justice and Walzer's reply, see Barry (1995) and Walzer (1995: 292–3).

⁶ Gillian Brock too notes that Caney's interpretation of global equality of opportunity 'permits many cases which do not look at all like a robust account of blocking disadvantage and discrimination on morally arbitrary grounds' and, in particular, that on his account, there might be an enormous difference in persons' 'abilities to influence policy, political arrangements or global institutions, which shape their lives [...], even if they enjoy an equal standard of living, as cashed out by the relevant capabilities and their contribution to well-being' (Brock 2005: 349–50). However, instead of trying to remedy this weakness, she suggests that we focus on ensuring that all persons have a decent opportunity set.

⁷ By 'manipulable factors', I mean not only factors that can materially be manipulated (access to education can for instance be enhanced by the construction of new schools, the implementation of specific school policy, etc.), but also factors that may legitimately be manipulated (withdrawing children from their family in order to enhance their access to education may not be allowed if being raised in a family is considered an important good).

⁸ For different versions of 'competitive equality of opportunity', see for instance, Lloyd Thomas (1977), Green (1988, 1989), and Jacobs (2004).

⁹ It is less than a guarantee because an opportunity does not mean the absence of all possible obstacles between an agent and a goal; and it is more than a mere possibility because it does not mean the possible absence of all obstacles between the agent and the goal (Westen 1985: 839–40).

¹⁰ Not all libertarians will favour the removal of laws that restrict emigration and immigration. As Hillel Steiner (1992: 93–4) notes: 'libertarianism's attitude to transnational migration follows straightforwardly from its view that national boundaries possess no less — and no more — moral significance than the boundary between my neighbour's land and mine. Neither I nor any collectivity of which we are both members can claim a non-contractual power to prevent her or him from allowing, or compel her or him to allow, another person on to her or his property. And we similarly lack any non-contractual power to prevent her or him removing themselves and all their property from our jurisdiction.'

¹¹ Certainly, formal equality of opportunity maximizes the availability of talents, which would otherwise be wasted. Yet, the question of whether it is also the most efficient thing to do is an empirical question. It depends on how the costs incurred by the removal of barriers are balanced against the costs incurred by not making all as yet unavailable talents available. From the perspective of a single society (instead of the world at large), it might for instance be more efficient to protect some of its sectors and thereby avoid rising unemployment or discontent among powerful interests groups.

¹² Procedural fairness need not entail the removal of every barrier to trade and capital flows. The reason it demands that agricultural goods be allowed to move freely is that agriculture is a sector (a) which is of utter importance to poor societies and (b) in which poor societies have a comparative advantage. Yet, liberalization of other sectors might not be in the interest of these societies at all and might therefore not be required by procedural fairness.

¹³ That is to say, the requirement that equality of opportunity be achieved between persons who participate in the global economic order, whatever the society to which they belong, entails that their chance to attain global economic advantages should not be affected by the place they occupy within their respective society (on account of their gender or social class). This explains why the proposed principle of global equality of opportunity involves direct confrontation with inegalitarian domestic institutions.

¹⁴ As indicated at the outset, this article deals with the question of whether equality of opportunity does make sense between equally talented and motivated persons, whatever the society to which they belong. While the supported conception global equality of opportunity is based on persons' participation in the global economic order, it is left aside whether it may also apply to other participants in this order (e.g. firms, governmental agencies, etc.)

¹⁵ On the idea that a states system is not incompatible with moral cosmopolitanism, see Barry (1995), Beitz (1999: 520–1, 528–9), and Pogge (2002: 169).

¹⁶ The fact that these authors interpret the principle of equality of opportunity as 'advantages open to talent and motivation' does not rule out that they may be committed to the undeserved character of talents and abilities and, on this basis, support an 'additional' redistributive principle aimed at making inequalities resulting from differential talents and abilities to the benefit of the least advantaged. But unlike 'responsibility-sensitive egalitarians', they do not regard compensation for lack of talent as a requirement of background fairness and thus as part of the principle of equality of opportunity.

References

1. Arneson, Richard J. (1989) 'Equality and Equal Opportunity for Welfare', Philosophical Studies 56(1): 77–93. | Article |
2. Barry, Brian (1995) 'Spherical Justice and Global Injustice', in David Miller and Michael Walzer, eds, Pluralism, Justice, and Equality, 67–81, Oxford: Oxford University Press.
3. Beitz, Charles R. (1999) 'Social and Cosmopolitan Liberalism', International Affairs 75(3): 515–529. | Article |
4. Beitz, Charles R. (2001) 'Does Global Inequality Matter?', Metaphilosophy 32(1–2): 95–112. | Article |
5. Boxill, Bernard (1987) 'Global Equality of Opportunity and National Integrity', Social Philosophy and Policy 5(1): 143–168.
6. Brock, Gillian (2005) 'The Difference Principle, Equality of Opportunity, and Cosmopolitan Justice', The Journal of Moral Philosophy 2(3): 333–351. | Article |
7. Buchanan, Allen (2000) 'Rules for a Vanished Westphalian World', Ethics 110: 697–721. | Article |
8. Caney, Simon (2001) 'Cosmopolitan Justice and Equalizing Opportunities', Metaphilosophy 32(1–2): 113–134. | Article |
9. Charvet, John (1969) 'The Idea of Equality as a Substantive Principle of Society', Political Studies 17(1): 1–13. | Article |
10. Cohen, Gerald A. (1989) 'On the Currency of Egalitarian Justice', Ethics 99(4): 906–944. | Article |
11. Daniels, Norman (1978) 'Merit and Meritocracy', Philosophy & Public Affairs 7(3): 206–223.
12. Fishkin, James S. (1983) Justice, Equal Opportunity, and the Family, New Haven, CT: Yale University Press.
13. Green, Simon J.D. (1988) 'Is Equality of Opportunity a False Ideal for Society?', The British Journal of Sociology 39(1): 1–27. | Article |
14. Green, Simon J.D. (1989) 'Competitive Equality of Opportunity: A Defense', Ethics 100(1): 5–32. | Article |
15. Jacobs, Lesley A. (2004) Pursuing Equal Opportunities: The Theory and Practice of Egalitarian Justice, Cambridge: Cambridge University Press.
16. Joseph, Lawrence B. (1980) 'Some Ways of Thinking about Equality of Opportunity', The Western Political Quarterly 33(3): 393–400. | Article |
17. Kapstein, Ethan (2006) Economic Justice in an Unfair World, Princeton, NJ: Princeton University Press.
18. Kivimäki, Timo (2001) '"Humane and Market-oriented": An Explication of the Western, Humane, Market-oriented Concept of International Exploitation in the International Co-operation between the Rich and the Poor', Journal of International Relations and Development 4(3): 202–220.
19. Lloyd Thomas, David A. (1977) 'Competitive Equality of Opportunity', Mind 86(343): 388–404. | Article |
20. Lucas, John R. (1993) Responsibility, Oxford: Clarendon Press.
21. Mason, Andrew (2001) 'Equality of Opportunity, Old and New', Ethics 111: 760–781. | Article |
22. Miller, David (1999) 'Justice and Global Inequality, in Andrew Hurrell and Ngaire Woods, eds, Inequality, Globalization and World Politics, 187–210, Oxford: Oxford University Press.
23. Miller, David (2004) 'National Responsibility and International Justice', in Deen K. Chatterjee, ed., The Ethics of Assistance, 123–143, Cambridge: Cambridge University Press.
24. Miller, David (2005) 'Against Global Egalitarianism', The Journal of Ethics 9(1–2): 55–79. | Article |
25. Moellendorf, Darrel (2002) Cosmopolitan Justice, Boulder, CO: Westview.
26. Nozick, Robert (1974) Anarchy, State, and Utopia, New York: Basic Books.
27. Nussbaum, Martha (1992) 'Human Functioning and Social Justice: In Defense of Aristotelian Essentialism', Political Theory 20(2): 202–246. | Article |
28. Pogge, Thomas (1994) 'An Egalitarian Law of Peoples', Philosophy & Public Affairs 23(3): 195–224. | Article |
29. Pogge, Thomas (2002) World Poverty and Human Rights, Cambridge: Polity Press.
30. Pogge, Thomas (2005) 'Recognized and Violated by International Law: The Human Rights of the Global Poor', Leiden Journal of International Law 18(4): 717–745. | Article |
31. Rawls, John (1999a) A Theory of Justice, revised edition, Oxford: Oxford University Press.
32. Rawls, John (1999b) The Law of Peoples, Cambridge, MA: Harvard University Press.
33. Roemer, John E. (1998) Equality of Opportunity, Cambridge, MA: Harvard University Press.
34. Sen, Amartya (1993) 'Capability and Well-Being', in Martha Nussbaum and Amartya Sen, eds, The Quality of Life, 30–53, Oxford: Clarendon Press.
35. Sen, Amartya (1999) Development as Freedom, New York: Oxford University Press.
36. Sen, Amartya (2002) 'How to Judge Globalism?', The American Prospect 13(1): A2–A6.
37. Steiner, Hillel (1992) 'Libertarianism and the Transnational Migration of People', in Brian Barry and Robert E. Goodin, eds, Free Movement: Ethical Issues in the Transnational Migration of People and of Money, 87–94, University Park: Pennsylvania State University Press.
38. The World Bank (2005) World Development Report 2006: Equity and Development, New York: Oxford University Press.
39. United Nations Development Programme (2005) Human Development Report 2005, New York: United Nations.
40. Walzer, Michael (1983) Spheres of Justice, New York: Basic Books.
41. Walzer, Michael (1995) 'Response', in David Miller and Michael Walzer, eds, Pluralism, Justice, and Equality, 281–297, Oxford: Oxford University Press.
42. Westen, Peter (1985) 'The Concept of Equal Opportunity', Ethics 95: 837–850. | Article |

Acknowledgements

I thank the Fonds Wetenschappelijk Onderzoek — Vlaanderen (FWO), which provided me with financial support and allowed me to concentrate on the writing of this article. I also thank Antoon Vandevelde, the three anonymous reviewers, and the editors who provided me with helpful advice and criticism. Of course, despite my great indebtedness to them, they should not be held responsible for the views expressed here.

About the Author

Sylvie Loriaux is a postdoctoral researcher at the Centre of Law and Cosmopolitan Values at the University of Antwerp (Belgium). She received her Ph.D. in philosophy from the Katholieke Universiteit Leuven (Belgium). Her main research interests include theories of justice, global ethics, multiculturalism, human rights, and the work of John Rawls and Immanuel Kant.