‘Coercive’ Associations

Most liberal-egalitarian theorists of justice tend to assume, often implicitly, what we may call ‘a radical division of institutional labor’ between the state and civil society. Will Kymlicka summarizes the terms of this separation well: ‘the distinction between state and civil society maps onto the two parts of the liberal-egalitarian ideal: social justice is secured by the state … individual freedom is secured in civil society.’Footnote ³³ We find similar terms in Rawls: ‘As citizens they [people] cooperate to achieve their common shared end of justice; as members of associations they cooperate to realize ends falling under their different comprehensive conceptions of the good.’Footnote ³⁴ As long as this, highly moralized, radical division of labor is assumed, it is clear why associations have no relevant autonomy-infringing power within the liberal-egalitarian system – associations are left free to pursue their own ends in the knowledge that the conditions of freedom and autonomy are independently secured.

However, this radical division of labor is not, or not uncontroversially, a necessary requirement of justice and, in any case, it is empirically false. I say that it is not a necessary requirement of justice because we could, in principle, imagine equally just societies in which different divisions of institutional labor between the state and civil society are in place.Footnote ³⁵ It is empirically false in so far as no existing liberal democracy experiences such a radical division (and never has, and probably never will). By contrast, in many societies, states secure the conditions of autonomy through associations.

For instance, the welfare regime of the United States has long been characterized by a system of indirect provision.Footnote ³⁶ Government social-welfare spending is directed – whether through grants or contracts – to a large number of secondary associations that produce and deliver essential goods and services on behalf of government. In 2010 about 33,000 civil society associations were under a total of some 200,000 government contracts for social services delivery, including education, healthcare, childcare, and unemployment benefits.Footnote ³⁷ In some states, secondary associations can control up to 90 percent of overall social service delivery and up to 75 percent of state benefits delivery.Footnote ³⁸ Nonprofit organizations provide 71 percent of all hospitals nationwide.Footnote ³⁹

Now, it is not at all obvious that this system of public-private partnerships, where funding remains public and delivery is privately run, is in principle incompatible with the requirements of liberal-egalitarian justice. What matters from the perspective of justice is that social institutions, taken together, are arranged so as effectively to secure certain conditions of liberty, equality, and stability. If these conditions, as some argue, can be more effectively secured by a fully decentralized system that systematically relies on civil society actors for the production and delivery of justice-required goods and services, liberal-egalitarian conceptions of justice may have reasons to uphold this form of associationalism.Footnote ⁴⁰

Regardless, this and similar forms of institutional interdependence between state and civil society, as opposed to a radical division of labor, are an increasingly widespread and long-lasting empirical feature of many societies (precisely like the consolidation of new international institutions), and thus deserve to be taken seriously. Yet, once we take this interdependence seriously we realize that many associations come to share the same properties that, according to the Coercion View, make the state a special site of justice. This is because by acquiring the power to deliver justice-required goods and services, secondary associations often acquire that kind of autonomy-infringing and entitlements-determining rule-making power that, according to the very criteria established by the Coercion View, triggers special standards of justification. To illustrate, consider first the following examples.

Political Justification and ‘Street-Level’ Associational Power

A first necessary step to make the decision-making power of service-providing associations publicly justified is to ensure that the legal framework through which that power is delegated to them is itself just. Since in our societies decentralization generally happens through delegation contracts, the first way in which justice can govern service-delivering associations is by applying to contract law. This requires that the legal framework of delegation contracts, itself a part of the basic structure, should itself be assessed according to first principles of egalitarian justice (whatever their specific content). We may thus say that in the same way in which family law establishes the legal rights and duties of equal citizens in their roles as spouses, parents, and children, in this same way contract law, in the case of privatization, should establish the legal rights and duties of contracted parties (associations’ managers and workers) and beneficiaries, in accordance with first principles of egalitarian justice.

This is an important first way in which egalitarian principles directly apply to secondary associations (as directly as to the family). But it would be mistaken to think that this is all that justice requires, and this is so even if we assume that this is all that justice would require in the case of the family.

The main reasons why justice should not stop at the level of contract law in the case of service-delivering associations should be clear. First, privatization contracts are different from other types of contracts, such as marriage, for they delegate political decision-making power. In this respect decision makers within service-providing associations are more similar to legislators than to parents. Second, and most importantly, delegation contracts cannot a priori fix with precision the rights of beneficiaries and the obligations of service-delivery associations. For it is daily decisions concerning implementation and administration, happening within private associations, that ultimately fix the content of beneficiaries’ entitlements.

For both these reasons, justice cannot stop at the level of contract law. It must directly reach internal decisions. Indeed, justice demands that we treat ‘street-level’ decisions as themselves being a matter of basic justice, insofar as they ultimately determine beneficiaries’ entitlements. This is to say that a just system of delegation contracts should be complemented by a process of internal political justification at the associational level. This process requires that entitlements-affecting decisions made within private associations be justified to all those subject to them (such as service-recipients, beneficiaries, patients, etc.) in the light of public reasons. These reasons must make direct reference to constitutional values, thereby affirming the standing of all stakeholders as free and equal citizens.Footnote ⁵⁴

Therefore, the demands of public reason which, on Rawls’s view, do not apply to the discussions of citizens qua members of private associations, do extend within decentralized systems to the reasoning of individuals within associations.

But if the range of reasons that can justify inequalities in the setting of eligibility criteria or in decisions concerning what claims to prioritize, or what requirements to impose on beneficiaries, become limited to considerations that all stakeholders, including those with the most urgent claims, can reasonably find acceptable, the range of inequalities permitted by the resulting local distributive principles will be, in turn, severely limited. For example, some inequalities in the allocation of goods and services by particular associations may be justified on grounds that they go to the benefit of the worst-off (among the relevant stakeholders), but inequalities could not be justified by simple appeal to the comprehensive mission or religious values of the association delivering the goods. It is true, public reason may be inconclusive, insofar as multiple distributive principles at the local level could be compatible with affirming the standing of recipients as free and equal citizens.Footnote ⁵⁵ Yet, all admissible principles would most likely have an egalitarian outlook. For example, a process of public reasoning that takes the perspectives of all stakeholders into serious account would be unlikely to uphold a principle that allows denial of medical treatment to the worst-off (among the stakeholders) in order to provide higher medical benefits to those who have much less urgent claims.

We reach the conclusion, therefore, that, at least in the case of those (many) associations that make entitlements-defining decisions, principles of egalitarian distributive justice should directly guide the conduct of local associations, and this is so according to the very criteria established by the Coercion View. Within the range of distributive principles compatible with public reason, different associations may then be permitted to endorse different distributive policies, depending on the particular needs, as well as the sociological or demographic reality of the area within which they operate.

At this point, it might be objected that I have used the term ‘distributive principles’ improperly. Distributive justice, the objection goes, is not concerned with the allocation of particular goods. It is only concerned with the overall social process of production and the basic rules that govern it.Footnote ⁵⁶ The norms established by secondary associations count as ‘secondary’, rather than basic, for they are ‘supervenient upon the multiplicity of basic social institutions constituting the basic structure.’Footnote ⁵⁷

My answer to this objection is twofold. First, it is unclear why principles governing the distribution of goods and entitlements within an association should be understood as being different in kind from those governing the same process within the state. The kind of a principle depends on the principle’s basic function. But principles applying to associations may have the same basic function as principles applying to states: to establish who, in the presence of competing claims and in conditions of scarcity, is rightfully entitled to what.

Second, the distinction between ‘basic’ and ‘secondary’ rules upon which the objection rests, albeit analytically sound, is practically untenable. It completely disregards the complex and intertwined relationship between putatively ‘basic’ and ‘secondary’ or ‘street-level’ norms and decisions, within the social process of production. Associations, through their acting as consultants to government, as accreditation bodies, and as standards-setters, regularly participate in the very making of those myriad rules that provide basic structure and content to welfare policies. As implementers of those very policies, they ultimately specify their content and thus contribute to defining the terms of access, exchange, and consumption. These putatively ‘secondary’ norms and decisions thus end up constituting an important part of that ‘basic’ normative framework that regulates the production, distribution, and consumption of important goods.

In practice, the best available way of realizing the demands of political justification within local associations in situations of systematic decentralization might be to create local deliberative committees inclusive of associations’ professional staff, beneficiaries, members of local governments (representing the broader (instead of board’s) public point of view and thus connecting decision making within local associations to larger deliberative processes at the societal level), as well as members of local associations performing similar functions in different communities (so as to include an element of coordination and benchmarking across local associations). Note, as here justified, deliberative committees within nonprofit associations are understood neither as a demand of democracy per se nor mainly as a tool of problem solving. They are rather regarded as a demand of justice. They are institutional mechanisms for realizing (to the extent possible and under conditions of disagreement) a process of political justification, from which action-guiding principles of local distributive justice can be derived. Progressively, the action-guiding principles that emerge from decisions made by these committees would become embedded in a sort of associational constitution.Footnote ⁵⁸ National courts should then make sure that local constitutions retain compatibility with the national one. Furthermore, citizens should be able to enforce constitutional requirements directly against associations rather than through the mediation of government.

To sum up: at least within social systems (such as those we live in) that feature a strong institutional interdependence between political institutions and civil society associations, even those who support the Coercion View have strong internal reasons to extend the demands of both public reason and egalitarian justice to many local, nonstate associations.

The Limits of Exit and Consent

First, egalitarian demands would seem to arise within some associations from which exit without loss of autonomy is clearly possible. Municipalities provide a clear example. The same may be said of some nonpolitical associations as well. Consider the following example. At X University, a professor of sociology employs a team of three researchers, John, Jean, and Jim. The output of the project will be a book. The three researchers equally contribute to the project. John conducts interviews, Jean collects data, and Jim performs analysis. Finally, they write up a book together. The book is completed. At that point the professor decides that only Jean’s and Jim’s names will appear on the book; John’s name will not. When asked why, the professor answers: ‘John, unlike the others, you could easily exit the team and find another job.’ John, we think, has been treated unfairly. The rules of the game are set up in a way that denies him what is rightfully his own relative to other participants: an equal share of the product of their joint cooperation. Most importantly, whether John is able to exit the association without a loss of autonomy would seem to be irrelevant to whether his name should appear on the book together with the other names.

Note that John’s equal entitlement to have his name on the book is not based solely on a principle of interpersonal morality (for example, proper appreciation for other people’s work). John’s entitlement is also grounded on a principle of fair return. An obligation of fair return arises within this context because (1) the association (the research team) can be reasonably regarded as a cooperative venture since it is framed by a system of formal or informal norms that distribute rights and duties, and attach entitlements to positions of responsibilities; (2) John, by participating in the team project, intentionally benefits others – Jean and Jim’s talents would be insufficient to generate the book from which they all benefit, or it would be much more costly for them to reach those results, without the cooperation of John’s talents; (3) John incurs costs in doing so, his labor. Finally, (4) the good produced by John’s cooperation (the book) is a good that we can assume all participants in that cooperative venture have reasons to want – indeed, this is why they support that cooperative venture in the first place and they accept the benefits that scheme produces. For these reasons, taken together, it seems reasonable to argue that to deny John’s equal entitlement to the final product (once cooperation has already happened) is not simply disrespectful or unprofessional, it is also a breach of a principle of fair return.

It could however be objected that whether John can rightfully claim an equal share of the product depends entirely on whether John voluntarily consented to the terms of the enterprise, before joining. Before testing this objection, note that if this is the sense in which voluntariness blocks egalitarians demands, many so-called ‘voluntary’ associations would not meet this condition anyway. For example, we do not explicitly consent to all the terms of churches and we may be compelled to join trade unions.

However, let us imagine that the professor has spelled out to John the conditions of participation previous to his entrance. ‘John, you must contribute equally to the project, however, you will not have your name on the book. Jim and Jean will.’ John understands the terms of the contract and thinks that they are unfair. Yet since he admires his professor, he willingly consents to participate. Does the fact that John consents to this treatment automatically make the terms of the contract fair? I would say it does not. Indeed, I believe that the best way to describe what happened would be to say that John has voluntarily consented to be treated unfairly. This is to say that, by consenting to that agreement, John has waived his preexisting entitlement to an equal share of the fruits of the team labor, thereby releasing his professor from an otherwise-binding duty to treat him as an equal. Imagine the professor had forgotten to make John aware of the terms of the contract. It would seem that, in this case, the professor would be bound to act according to a default norm of equality (at least in the absence of overriding considerations).Footnote ⁶⁰ Assuming this view to be correct, it is not the case that the presence of consent makes John’s distributive entitlements inexistent.Footnote ⁶¹ At most, it disactivates them. But the ground of such entitlements is independent of consent.

The further question is whether John’s consent is sufficient to release his professor from the obligation to treat him as an equal? The answer to this question depends on a previous theory of contractual obligation. If we believe that the moral force of a contract derives exclusively from the conditions under which the contract is signed (absence of coercion, access to relevant information, etc.), then John’s consent does have that power. But there are reasons to believe that free consent is not always a sufficient condition for a contract to have moral force.Footnote ⁶² For instance, it is generally agreed that people who voluntarily and freely consent to become citizens of a country different from their original one, cannot (be asked to) waive their entitlements to be treated as equals. Those entitlements are attached to the type of relationship or association they are entering and cannot be waived.

Of course, one could object that while we cannot waive our rights to be treated as equals when we voluntarily enter a state, we can waive our rights to be treated as equals when we join a research team. This is because once we become citizens of a state we cannot easily exit it without loss of autonomy, while we can always exit a research team easily.

And yet, even in the case of the research team, consent might not always be sufficient to waive preexisting distributive entitlements. Imagine, for example, that the professor says to John, previous to the agreement, that despite John’s equal contribution to the book, he will not put his name on the book because he considers John to be morally inferior. We may think that even if John freely consents to the terms of the agreement, the agreement should have no moral (and no legal) force, because the practice of consenting to these types of agreements, once regarded as permissible, would undermine basic norms of moral and political equality. The professor then would retain an obligation to treat John as an equal, regardless of John’s consent.

It seems, therefore, that, at least in some cases, the demands of equality can arise within associations where exit is possible and also retain their moral force independently of a person’s consent at entrance. However, let us assume that the demands of equality, as demands of justice, can only arise within nonvoluntary associations. Would we have any reason left to regard secondary associations, at least some of them, as appropriate sites of justice?

Nonvoluntary Associations

In existing societies the state and the family can hardly count as the only nonvoluntary associations. This is for several reasons. First, secondary associations can and do sometimes hold monopolies or quasi-monopolies over access to basic material goods. In the United States, one striking example of this monopolist control is provided by Catholic nonprofit hospitals (as well as by MCOs and other organizations). These secondary associations represent 12 percent of all hospitals nationwide.Footnote ⁶³ However, more than a quarter of these hospitals are located in rural areas, where there are often no other viable health care options, especially for the poor, who may not be able to afford to travel to another institution. These organizations often refuse, on religious grounds, to deliver particular medical services (for example, reproductive health services to women). It would certainly be mistaken to regard these associations as ‘voluntary’ for individuals cannot choose to exit, or not enter, these associations without losing access to their basic conditions of autonomy.

It may be argued, however, that within states that are fair and just cases of nonvoluntary associations would not arise, for it is the responsibility of a state to secure alternative, nonreligious providers. This argument, albeit sound in its form, is too simplistic. It fails to take into account the possibility that, in some circumstances, all things considered it may be justifiable for a state not to secure alternative providers.

To illustrate: suppose, as it is often the case, that the region where the religious hospitals operate is mainly populated by religious communities, which live far removed from urban centers. Suppose that for this reason, it is difficult to motivate highly skilled citizens, external to those communities, to move to, live and work in these rural regions. As a result, there is a scarcity of labor. There are not enough people who are willing to work in the region as doctors or nurses. Lacking the necessary labor, an alternative hospital cannot be set up unless (1) people with the relevant skills are forced to relocate to that area and take up hospital jobs, or (2) sufficient incentives are provided to motivate those with the relevant skills to voluntarily move to that area.

However, (1) violates freedom of occupational choice. Option (2) may be extremely expensive. It is not unreasonable to think that, in conditions of scarcity, all things considered a state might be justified in spending its revenue on other causes. In this situation we would say that, all things considered it is justifiable for the state to let private organizations have monopolistic control over healthcare provision within that area. Yet, given this monopoly, those organizations should be regarded as nonvoluntary. It would then follow that the board of directors of these nonprofit organizations should make decisions about the eligibility for, and quality of, patient care according to reasons that all those subject to these decisions, whether religious or not, could reasonably accept.

Second, and more generally, in order to count as fully ‘voluntary’ all associations must operate against certain background conditions of justice. These conditions secure that agreements (including the agreements through which associations form) are fairly arrived at and subject to voluntary exit. These conditions of voluntariness reasonably include, beyond protection of property rights, also adequate access to basic goods such as healthcare and education. Yet, as long as states secure these very conditions by acting through secondary associations, rather than independently from them, these associations can no longer claim their own ‘voluntariness’ as a valid reason to be exempted from the requirements of justice themselves, for they come to share the responsibility for securing the conditions that make that voluntariness possible. In this respect, the many secondary associations, whether religious or not, that act as a constitutive part of the state’s capacity effectively to secure these background conditions should themselves be bound by the demands of political justice, whether they hold a monopoly or not.

Finally, sometimes we may have reasons to consider as nonvoluntary also associations that neither hold monopolies over material goods nor perform delivery functions. This is because, beyond material goods, also certain immaterial goods can be reasonably included in the list of goods, the deprivation of which imperils individual autonomy and thus voluntary exit. Consider the good of self-respect. Liberals generally regard self-respect as being crucial for acting on a plan of life, because ‘unless our endeavors are appreciated by our associates it is impossible for us to maintain the conviction that they are worthwhile.’Footnote ⁶⁴ Interestingly, Rawls himself, although being primarily concerned with the broader social bases of self-respect, points out that ‘what is necessary [for self-respect to be maintained] is that there should be for each person at least one community of shared interests to which he belongs and where he finds his endeavors confirmed by his associates.’Footnote ⁶⁵ Now, if self-respect is a basic good, any time exiting a particular association could be said to deprive an agent of access to this good, so we should count that association as nonvoluntary. But this potentially increases the number of associations that should count as non-voluntary for the purpose of justice, independently of state–civil society relationships.

Consider the example of churches. A dedicated Catholic nun who joined the Catholic Church at a young age and spent most of her life within a religious institution may not be able to exit the Catholic Church without being deprived of the possibility of having her conception of the good appreciated by others sharing that same conception of the good. This is because the Catholic Church (as all other churches) has a full monopoly over its conception of the good. Further, she may have a difficult time finding alternative means of subsistence, since her developed skills may not be of value in the labor market. Therefore, if self-respect is a basic good, and if it requires (at least for some people) participating in a community of common interests, we have reasons to regard the participation of at least some members in religious associations as nonvoluntary, according to the liberal definition of the term. This would provide a pro tanto reason to regard these members as having an equal claim to the fruits of their association (for example, veteran nuns may claim, on grounds of justice, an equal opportunity to access positions of power and responsibility within the Church).

To sum up, even if we assumed that egalitarian demands can only arise within non-voluntary associations, these demands need not be limited to the state-association. Indeed, the range of associations that can count as non-voluntary goes well beyond the state (and the family). Now I turn to the Reciprocity View.