Neighbourhood Demographic Quotas and Limits of the Right to Sort

I. SEGREGATION, INEQUALITY, AND DEMOGRAPHIC QUOTAS

Although there are many reasons to find sorting troubling, we will focus on an established account of how neighbourhood arrangements where there are class and ethnicity clusters correlate with inequality in access to various beneficial goods, such as income, jobs, education, and health care.

This happens partly due to the problem of spatial mismatch.¹⁰ This describes the geographical distance between people and work. One reason people in low-income areas struggle to find suitable job opportunities is their respective locations. While there may be decent prospects in other neighbourhoods, individuals living in disadvantaged places often do not hear about them or cannot easily reach them. A similar difficulty arises in relation to other goods.¹¹ Although reasonable health care and education is available in nearby locations, neighbourhood networks of people, communication, and transport act as a disconnection to access.

Another concern with residential clustering relates to social segregation. The issue here is that the existence of demographic concentrations in different neighbourhoods feeds into perceptions of in-groups and out-groups. It is a common phenomenon that people tend to hold prejudices about individuals they perceive as part of an out-group and there is increasing evidence that residential arrangements have bearing on this.¹² One explanation relates to social contact – because people interact less with those who live further away, they feel less social bond with them and tend towards feelings of dislike and distrust.¹³ Another, perhaps complementary, explanation relates to geographical space – when concentrations of socially salient groups live in distinct locations, it increases the perception of them as cohesive units and heightens the sense of in-group and out-group.¹⁴

The mechanisms of social segregation are important because they compound the relationship between demographic concentrations and inequality through what we might call the impacts of prejudice.¹⁵ This is the point that negative perceptions of out-groups manifest in behaviour that affects prospects. For example, those in advantaged enclaves are more likely to discriminate against those who live in disadvantaged neighbourhoods, unfairly denying them access to goods such as jobs and housing. They are also more likely to vote against progressive policies that would address aspects of disadvantage through political measures. In this way, demographic arrangements not only present a barrier to accessing valuable goods (as they do through spatial mismatch) but operate as a causal mechanism that perpetuates inequality.

We think this relationship between residential geography and inequality provides a presumptive basis for states to act against sorting. If the problems were not connected in this fashion, redistributing wealth might be a way to address differences in prospects across neighbourhoods. But the bearing of social segregation and prejudice suggests that a long-term response likely involves some attempts to advance integration.¹⁶

Demographic quotas provide a mechanism for addressing these concerns. We can observe this in several parallel contexts. Where there has been a high concentration of men acting as parliamentary representatives, quotas reserving a certain number of electoral seats for women has been a means to combat their underrepresentation. There has been a similar approach to addressing the dominance of men on company boards and other high-status positions, such as senior academic roles. Likewise, several countries have reserved a set of places on university courses for candidates with an ethnic minority background. While the exact reason to deconcentrate advantage in these spaces may vary, the virtue of setting threshold targets to achieve the desired effect is straightforward. Where the number of individuals from certain socially salient groups is disproportionately low, quotas act as a direct corrective.

Drawing on policy examples, we can differentiate three ways a government might implement a quota setting a minimum threshold for the proportion of individuals from disadvantaged groups that neighbourhoods where they are currently underrepresented must meet or surpass. First, it might pursue a force strategy, ordering individuals to move out of an area, ordering others to move into it, or both. Denmark’s Parallel Societies law worked this way by emptying and demolishing accommodation in certain neighbourhoods.¹⁷ We could imagine taking the same approach to enclaves of advantage, perhaps also relocating disadvantaged individuals into vacant or newbuild housing blocks there.

Second, a government could adopt a regulative strategy, such as restricting to and from whom people can buy, sell, and rent, as Singapore’s Ethnic Integration Policy works.¹⁸ There are several options for implementing a version of this in enclaves of advantage. One is pursuing regulative-replacement – changing a population over time by stipulating that any advantaged individuals exiting the area may sell their property to only those who are underrepresented there. Another is regulative-expansion, where a neighbourhood is required to create additional accommodation that is made available to only disadvantaged individuals. For example, a community of 300 could go from being 100% to 75% populated by people who are white if it built new homes to house an extra 100 and all of them go to individuals with a minority ethnic background.

Third, a government could use a penalty strategy that allows people to respond to a target in a variety of ways. For example, it could tax advantaged individuals who are overrepresented in a neighbourhood or fine local councils who do not meet the quota, thus presenting them with a choice between relocating or working to avoid the penalty through collective efforts to increase the numbers of those who are underrepresented. The latter could include combatting housing discrimination in the area to increase the likelihood of demographic change through neighbourhood turnover (penalty-replacement). It could also include relaxing exclusionary zoning laws that restrict housing to detached, single-occupancy options and building (or converting to) duplexes or apartment blocks that could accommodate more residents (penalty-expansion). In a broad sense, using penalties to pressure such responses was the approach of the Affirmatively Furthering Fair Housing rule and France’s Law on Solidarity and Urban Renewal.¹⁹ It is also how quotas have worked in several other contexts, the requirement that company boards have a certain proportion of female members being a known example.

This offers a range of possibilities for those convinced by the importance of integration to consider. In what follows, we set aside strategies of force (which face several accentuated or distinct objections) and focus on regulative and penalty quotas. In determining which, if either, of these measures is a good way to proceed, there are significant questions about how they compare to other policy options for advancing integration. We shall return to this in Section VIII. But they also face pressing normative challenges, which concern claims individuals may have to resist state interventions of these kinds. We turn now to these matters.

II. A RIGHT TO SORT?

To consider possible criticisms of neighbourhood demographic quotas, it is useful to sketch the general form an objection to them would take. What this policy does is prevent people from residentially sorting in a particular way – forming clusters with a higher-than-permitted concentration of certain socially salient groups. Normative opposition will arise if individuals have a decisive claim against states making such interventions. We might reasonably refer to this in terms of them having a right to sort.²⁰

We can elaborate this idea by detailing two incidents a right to sort would entail. First, it involves a liberty to inhabit a space where others in the nearby vicinity share socially salient characteristics. Second, it involves a mandate against others, including the state, interfering with one occupying such spaces. We can distinguish this conceptualisation from a more demanding right involving individuals having a sufficiently strong interest in clustered communities that others were under a duty to support their realisation. We are unsure what, if any, arguments could lend credence to such an entitlement. But, a more modest right to sort need not entail anything so strong. It could amount, rather, to individuals having an interest or claim sufficient to justify a prohibition on others preventing them from accessing residential concentrations of a certain kind.

It adds some plausibility to this notion that it seems justifiable in the case of disadvantaged communities. In this vein, let us consider defences of individuals in such groups self-segregating.²¹ The idea is that those who society disadvantages have a permission to separate from those who have oppressed them. There are various arguments that could support this conclusion. For example, it may be conducive for building solidarity and political mobilisation, forming a safe environment that can nurture self-respect and preserve culture, or securing an appropriate degree of collective autonomy. What is important for our purposes is what reasoning along such lines would support. If we understand the self-segregation literature correctly, the aim is to defend a liberty to reside in a location predominantly populated by those who share socially salient characteristics and an injunction against states interfering with this. In other words, it would be a right to sort of the form that we have sketched.

The pertinent question for this article is: do advantaged individuals have a similar claim? To assess this, it is useful to distinguish between symmetric and asymmetric justifications of a right to sort. Those from the self-segregation literature fall into the latter set, because they appeal specifically to features of disadvantage or oppression. There are other possible contenders, perhaps most promisingly those arising from interests all individuals might be thought to hold in residential choice, free association, located life plans, and anti-discrimination. It is conceivable that each of these concerns applies symmetrically across the population, and we consider them in the next sections respectively. However, it will be our argument that where articulations of them might plausibly support a right to sort, the reasoning also leads to asymmetric conclusions, ultimately denying that the advantaged have a defensible claim of this kind.

III. RESIDENTIAL CHOICE

A valuable place to begin is with the thought that a right to sort might derive from an interest in residential choice. It is notable that neighbourhood demographic clustering could plausibly result from a series of individuals opting to live in areas for innocuous reasons, such as wishing to be near family or others who share one’s hobbies, outlook, and lifestyle. One might worry that neighbourhood demographic quotas restrict acting upon what appear perfectly reasonable preferences. A principled explanation of this reaction is that we have an interest in being free to design our lives as we wish, and this applies as much to the type of neighbourhood in which we live as anything else. The thought, then, is that we should have freedom to decide on our housing and its surroundings, and this might underpin a right to sort.

To assess this argument, it is useful to separate different ways we might express the idea of a claim to residential choice. A first would be to hold that states must protect certain freedoms in this domain. There is credibility to this. However, we doubt that neighbourhood demographic quotas will undermine any plausible account of an entitlement to residential choice or that it could justify a right to sort.

A parallel with occupational choice is instructive here. It is common to think that this is among individuals’ core liberties.²² It plausibly follows that a government may not force people to work in specific professions. It may also entail ensuring they have a meaningful range of options.²³ But, it does not entitle people to any line of work they wish, and states do not violate these constraints if they shape the labour market in certain ways. To see this, let us consider a scenario in which a man seeks to secure one of 500 positions as a parliamentary representative for which he is eligible. We do not think that there is a transgression of a core liberty of occupational choice if other people win the elections and it does not make a difference to this if the 500 seats are a subset of 600 that include 100 reserved solely for women. Nor do we think there is a violation if those who secure a position must undertake the role on particular terms, such as within contracted working hours or subject to taxation on their pay or working in a setting that employs male, female, and non-binary staff. As Raz puts it, having meaningful choice ‘does not require the presence of any particular option’.²⁴

These points are relevant because the conflict between neighbourhood demographic quotas and the right to sort lies in the particulars of community composition. Neither regulative nor penalty strategies determine by force where people live. What they aim to entail is that the particular option of living somewhere that has precisely the social surroundings some might desire becomes unavailable. A regulative approach that operates through limits on sale and purchase may also mean that there are fewer houses available to certain buyers in specific places. These parameters would conflict with a right to sort. But, they parallel limits on various conditions of contract and the number of jobs available that are not what a claim to choice could contest.

It is possible to imagine scenarios where the range of options left available to some individuals is limited. For example, if many neighbourhoods have reached the upper limit of advantaged individuals living there, a person looking to relocate could face a choice between only a set of disadvantaged places, perhaps with other similarities, such as there being only apartments, and no detached houses. A critic might question whether this constitutes a meaningfully diverse set.

We have three responses to make here. The first is to note that, at most, this objection poses a challenge to replacement policies. It does not contest expansion policies (be they regulative or penalty), which do not limit the options available to advantaged individuals compared to the non-quota context. The second is to highlight that even in the given scenario, holding that a choice between disadvantaged neighbourhoods with certain similarities lacks variety overlooks the diversity that will exist within that set across a host of social, economic, and cultural aspects. The third is that claiming otherwise commits one to the conclusion that the disadvantaged in our world, who face exactly this choice set, lack meaningful diversity in their options. If this is right, and we must select between perpetuating this situation or introducing a quota that shifts the limits from those currently affected to the advantaged, our sense is that we should do the latter.

Proponents of the residential choice objection might respond that it overstates their point to address it through the conceptualisation of a core liberty. Even if our freedoms in this area are not what falls under this label, critics might argue that restrictions on them still need more justification than highlighting the bearing of certain residential patterns on inequality. A similar thought appears in the literature on private schooling, where Debra Satz highlights that lots of choices people make in relation to their children disrupt equality, but do not seem plausible candidates for interference.²⁵ Good examples include giving one’s children music lessons, foreign holidays, private tutors, and bedtime stories. Insofar as a main incident of a right to sort is a claim against the interference of others, perhaps the same point is relevant here.

The difficulty with this line of reasoning is that it does not establish a line on which choices are and are not reasonable candidates for intervention. There must be some threshold here. Even if we permit parents to benefit their children, we do not allow them to murder or bribe others to this end.²⁶ The simplest way to explain this is to contend that freedoms have their limits at the point of violating others’ entitlements (precisely the issue of concern in residential patterns that disrupt equality) and that the exceptions we allow are those that need further argument.

It is possible to offer such reasons. For example, in the schools literature, one way to defend actions like reading children bedtime stories, despite their equality-disrupting effects, appeals to the idea of familial value.²⁷ In the case at hand, then, a critic might proceed by identifying a further value that would support a right against interventions of the kind made by neighbourhood demographic quotas. The next sections look at several possibilities along these lines.

IV. FREEDOM OF ASSOCIATION

One way to develop the case is by appeal to freedom of association. This is a two-part idea – it licences individuals both, positively, to connect and, negatively, to disconnect with people and groups of their liking – and centres on the idea that, in Mill’s words, ‘we have a right to choose the society most acceptable to us’.²⁸ If this extends to licencing us choosing to live near (and far) from those with whom we wish (not) to associate, it might justify a right to sort.

An important strength of defending the idea in this way is that freedom of association is typically thought to fall among our most fundamental rights. Even theories that place considerable weight on egalitarian distributions of opportunities and wealth hold that these goals cannot come at the expense of such liberties.²⁹ Along these lines, Loren Lomasky argues that there is ‘a strong presumption in favour of the right of individuals to dissociate’ that cannot be overcome by ‘aspirations toward…programmatic models for implementing particular patterns of diversity’.³⁰ While he does not explore the exact policy position we are defending here, he holds that (at least some) integration interventions with egalitarian rationales of the kind we have employed fall to this presumption.³¹ Put concretely, the worry is that by preventing sorting and aiming for a certain demographic mix, neighbourhood quotas interfere with people connecting and disconnecting from associations of their choosing.

To address this challenge, it is useful to delve into some theoretical terrain concerning how neighbourhoods fit into ideas about types of association, their claims, and their limits. In the most basic sense, an association is a relationship with one or some others.³² This definition admits of a wide range along a plausibly fluid spectrum, but we can identify some place markers. Literature and jurisprudence focus on two forms particularly – on the narrower end, intimate relationships, such as family and friends, and, on the wider end, collective endeavours, such as bowling clubs, churches, and trade unions.³³ At its broadest, we might stretch the concept to our social surroundings – the general landscape of our interactions.

Connecting this with neighbourhood sorting, we can think about the positive aspect of free association in terms of a right to form particular relationships of one of these kinds. This would be to hold that residential communities constitute an intimate or collective association or our social surroundings and that our right to shape groups to our choosing entitles us to joining and maintaining a sorted version. The correlative of this is a claim against the state compelling us to admit members we wish to exclude, which is what quotas would deny. We can also think about the negative aspect of free association as permitting us to disconnect from neighbourhoods we dislike or granting us a claim against forced association with others. There is undoubtedly accuracy to these ideas, but they cannot make a case against quotas without falling back on the positive aspect of the right to establish sorted neighbourhoods. One can exercise a negative freedom to disconnect from a church by staying home, but if one disconnects to form or join a new church, it is also an exercise of the positive freedom. Likewise, a claim against being forced to live in an integrated neighbourhood may permit one an exit to solitude, but disconnecting to move to a sorted community involves a positive act of associating. Accordingly, challenging residential demographic quotas by appeal to free association requires assessing the plausibility of a claim to form and maintain groupings of this kind. We conduct this assessment in two parts.

V. LOCATED LIFE PLANS

Another way to develop a basis for sorting appeals to the value of located life plans. What underlies the latter words in this phrase is an interest in autonomy. Part of what it is to author a life is to shape plans around one’s interests. This is relevant to our case because our projects often have a spatial component. We form and orient them in and assuming a particular locational setting, with its physical and social features. Changes to this environment (at least of a certain kind) interfere with such plans, thereby potentially threatening our self-direction. If this is right, it might provide a powerful basis to resist interventions that would have this effect. It would ground a defence of sorting and a claim against demographic quotas if these efforts at integration would impermissibly disrupt the stability of plans people have made in sorted neighbourhood contexts.

One version of this argument in literature appeals to the notion of occupancy rights. Emphasising that life plans occur in a specific place and environment, Jakob Huber and Fabio Wolkenstein hold that people have a right ‘to reside permanently in a particular space and/or to make use of that area for social, cultural, and economic practices’ and a claim that others do not ‘interfere with one’s use of that space in ways that undermine the shared social practices in which one is engaged’.⁵⁵ They apply this idea specifically to formulate an account of what makes gentrification wrongful. They argue that an inflow of middle-class or white individuals to historically working-class or Black communities alters the social, cultural, and economic context in a way that is disruptive to life plans of long-term residents.⁵⁶ While it is not their focus, we take it that the bearing of Huber and Wolkenstein’s interpretation of how occupancy rights relate to housing on our subject is clear. In parallel to the changes gentrification might bring to disadvantaged areas, an increase in individuals of lower socio-economic status or minoritised ethnicities to enclaves of advantage might affect the social, cultural, and economic practices of the space. If people’s located life plans give them a claim against interference with this environment, it would amount to a right to sort, at least in places where this is already the state of affairs.

Another argument developing the same theme turns on the notion of legitimate expectations. Travis Quigley uses this idea to defend the Not In My Back Yard (NIMBY) movement’s objection to new-build accommodation in their neighbourhoods.⁵⁷ The complaint here often occurs in low-density areas that typically have single-family occupancy, minimum lot size housing stock, and centres on rejecting proposals to build higher-density options, such as duplexes or apartment blocks, that disadvantaged individuals could afford. Quigley argues that those who have bought housing in such areas have formed plans around the nature of that place and these ‘will routinely be frustrated if the nature of one’s community is significantly changed’.⁵⁸ Although it is not part of Quigley’s argument, we could extend this line of reasoning to plans people may have made with the expectation of stable or rising house prices. Part of what underlies exclusionary zoning practices is that this asset value may be tied to their location remaining an enclave of advantage and quotas that affect change to the neighbourhood and its housing stock may undermine this.⁵⁹

How might we respond to such concerns? We start by emphasising how time plays a role in moderating them. Objections to interference with life plans are strongest when targeted against sudden changes. The threat it would pose to their projects may ground a case against evicting people from their homes and communities without warning. But appeals to plan-making do not extend to claims for permanently unchanging circumstances, precisely because the former does not require the latter, and there is recognition of this in the literatures on both occupancy and legitimate expectations.⁶⁰ Accordingly, these threads present no objection to either regulative or penalty quotas if they are forecast and implemented with time to adapt.⁶¹

Our more challenging response is to deny that those in enclaves of advantage have a located life plan claim against the aspects of change that neighbourhood demographic quotas would entail. To elaborate this point, we can begin with a distinction from the literature on expectations between procedural and substantive accounts. The former give grounding to these claims with reference to actors, such as (legitimate) governments, having responsibility for their formation.⁶² It is not entirely clear that life plans located in advantaged enclaves can appeal to such an account. On the one hand, where they arise organically under a state permission, it is difficult to hold that governments are responsible for their formation.⁶³ On the other, where they arise more directly from zoning practices currently the progeny of local government, there is reason to doubt their normative force given the questionability of the democratic processes these standardly involve.⁶⁴ To put it sharply, claims to preserve life plans with a procedurally dubious foundation look suspect.

However, there is also a broader reason to reject procedural accounts of legitimate expectations in favour of substantive accounts, which ground these claims with reference to their moral defensibility, or their ‘justice-relation’. One way to articulate the rationale for this is given by Quigley, who does not extend protection to located life plans that are ‘clearly morally bad and should not be treated as morally salient’.⁶⁵ A similar clause also appears in reasoning about occupancy rights.⁶⁶ We can illustrate the force of it in concrete cases. For example, without this justice-relation, an account faces a difficulty with the interests of slaveholders in the Antebellum South. These slaveholders had life plans that were legally permitted (indeed, supported) by governments, but it is surely implausible to argue that they have a claim against the frustration of these plans or for compensation.⁶⁷

The relevance of this here is that we doubt that the claims of those in enclaves of advantage to maintain sorting have a defensible justice-relation. The central reason for this is that people living in these places have significantly more resources than their just entitlements, and we think a persuasive view will deny weight to the proportion of their life plans that rest on this excess. By analogy, let us consider the career designs made by men in contexts where they systematically have better prospects than women. When we introduce policies to even prospects between genders, it seems hard to argue that these men can complain about their plans having to accommodate losing the proportion of opportunities to which they were not entitled. This suggests that where individuals stand in relation to a just benchmark matters, which has the upshot that claims to the protection of located life plans is another instance where there will be some asymmetry to their force. Damage to them looks significant in the instance of a disadvantaged community, such as the gentrification case, partly because individuals there have fewer opportunities than their entitlement. Accordingly, protecting them is consistent with (albeit less than) what justice requires. But not so for enclaves of advantage. In this case, preserving current circumstances involves sustaining located life plans grounded on more than their holders are entitled to claim. Similar to the men who must adjust career designs in virtue of interventions to enhance gender equality, we do not think those who live in enclaves of advantage have a claim of any force against states acting in ways that alter their life plans to within what their just entitlements would allow.

Perhaps a critic might argue that this approach incorporates an overly strict justice-relation. There are threads of this kind in the literature on expectations, where some argue that grounding a view on a tight connection to fair institutions risks rendering no expectations legitimate in our current (unjust) world or that it provides too little leeway for epistemic uncertainty about our exact entitlements.⁶⁸ Even when people have more than they should, many will have formulated located life plans around this fact innocently, and we may still be concerned with a denial of their autonomy.

Our response to this is to deny that our view, or our defence of quotas, is hostage to these worries. To make this clear, it is useful to separate two readings of the framing we have employed. On the one hand, what we have not argued is that none of the located life plans of advantaged individuals warrant protection. It is compatible with our view that we should not interfere with those that could reasonably arise with a fair (or less than fair) share of resources. This is the same reasoning we gave for not interfering with self-segregating disadvantaged communities. Accordingly, the output of our view is not that people, the advantaged included, have no claims to located life plans in an unjust world. On the other hand, though, what we have maintained is that the proportion of their plans that could not arise in a just world do not warrant protection. The force of this, we think, is clear from the gender equality example we detailed above, and we can reinforce it by considering the case of progressive income taxation.⁶⁹ If we hold that the life plans of the advantaged above their fair share of resources entitle them to a claim against frustration or to compensation, we are committed, at least pro tanto, to the conclusion that we should refrain from imposing such taxation or compensate unjustly wealthy people for its introduction. We take it that this is a mark against views that involve a justice-relation much different to the version we articulated. If this is right, we think neighbourhood demographic quotas, and the restrictions to sorting they entail, can resist a challenge from autonomy interests in located life plans.

VI. ANTI-DISCRIMINATION

A final argument for a right to sort arises from concerns about discrimination. We can sketch this line of reasoning by thinking about the broader phenomenon in the context of housing. Simply defined, discrimination involves treating individuals or groups differently in the distribution of benefits and burdens.⁷⁰ It is cause for concern when those treated better or worse share socially salient characteristics, such as race and class. For example, when a landlord opts for tenants who are white rather than tenants with a minority ethnic background on account of the latter’s race, the landlord engages in housing discrimination

Defenders of the right to sort might turn these concerns against neighbourhood demographic quotas. At least some quota systems, such as regulative-replacement, will function to deny people access to housing based on their socially salient characteristics. Under this system, we can imagine a revised version of the example above where the rental accommodation is in a neighbourhood that has reached its upper limit of advantaged residents, and, because their socially salient characteristic is underrepresented there, goes to a tenant with a minority ethnic background. If this constitutes an act of discrimination that states should prohibit, we have a case for rejecting neighbourhood demographic quotas that work in this fashion. In this way, it would offer a defence of a right to sort, at least against these kinds of interventions.

We are familiar with arguments like this from affirmative action debates. A well-known example is the case of Allen Bakke, a white applicant who was denied a place at University of California, Davis medical school despite achieving higher test scores than students admitted to 16 places on the programme reserved for Black or Minority Ethnic candidates. In upholding his complaint, the US Supreme Court (and some subsequent literature has) maintained that a system that denied the availability of certain places to Black applicants would be discriminatory and that ‘equal protection cannot mean one thing when applied to one individual and something else when applied to a person of another colour [sic]’.⁷¹ The line of objection we are considering here is similar. It holds that we cannot reject housing discrimination when it treats advantaged applicants more favourably in virtue of their socially salient characteristics yet endorse it when those individuals are denied access for the same reason via a neighbourhood demographic quota.

Our response to this is to deny that we should treat these two cases symmetrically. To elaborate this reply, let us assume a case where a neighbourhood demographic quota does employ certain socially salient characteristics in the distribution of housing. This does not settle whether it is morally troublesome. When states consider sight in the criteria for the distribution of driving licences, they employ a socially salient characteristic in determining how benefits and burdens fall. But we do not think that this is problematic. Accordingly, in order to condemn discrimination we need to identify not only an act of it, but also that is has the feature that makes certain instances of it wrongful. This is important here because discussions of the leading theories conclude that none of them support the idea that affirmative action quotas have the same wrong-making property as commonly accepted instances of problematic discrimination.⁷² For our purposes, it is useful to consider how two prominent views account for the difference.

The first of these is an approach centred on harm.⁷³ On this view, the wrong of discrimination lies in making those subject to it worse off. However, to determine whether someone suffers harm, we do not ask solely whether they do not fare as well as they would have done without the discrimination. Rather, we judge it with respect to a more holistic baseline, considering whether they are worse off than if there had been no acts of discrimination. This makes an important difference in cases like affirmative action when those who miss out on jobs in virtue of quotas have lives that are otherwise advantaged.⁷⁴ The fact that these individuals are well-off means it is less plausible that the instance of discrimination leaves them in worse circumstances than the specified baseline. Accordingly, it does not involve harm of the kind that would make this act wrongful.

A second prominent account of discrimination focuses on whether it demeans its subject.⁷⁵ An act is demeaning if, within a social and cultural context, it is reasonable to interpret it as expressing that an individual or group has a lower moral status. This is important because not all discriminatory acts will have such traction. Their potential to demean varies depending on whether there is a past or ongoing usage of a socially salient characteristic to separate people in a hierarchical fashion. To put this concretely, it travels in a relationship with the extent to which those with a certain trait have been subject to a ‘history of mistreatment or current social disadvantage’.⁷⁶ This view makes clear sense of historical instances of discrimination widely accepted as problematic. For example, disenfranchising women demeaned them because it signalled a second-class position in a context of systematic denial of equal rights and status. However, the implications of this view for affirmative action quotas designed to rectify the mistreatment of disadvantaged groups is quite different. The rationale for these policies does not involve the premise that advantaged individuals are of lower moral status. More importantly for our purposes, their circumstances make it less plausible that any such signal has the relevant significance. Their position of advantage means that they do not experience the background social and cultural context that would give this message traction.

This discussion of discrimination and affirmative action is relevant for our purposes in two ways. The first is that it clears quotas of the charge that they are necessarily wrongful.⁷⁷ In the same way that their use to widen the spread of certain high-status jobs does not constitute a problematic form of discrimination, neither does their use to deconcentrate neighbourhood enclaves of advantage. Accordingly, this charge cannot defend against state interventions of this kind or, therefore, a right to sort. The second lies in the reasoning used to reach this conclusion. The way in which quotas designed to deconcentrate advantage are different from wrongful cases of discrimination, such as denial of access to housing on grounds of race, is due to facts about wider circumstances. They are asymmetrical because it matters morally whether the individual or group who experience discrimination fare better or worse more generally, and they cannot give enclaves of advantage a right to sort or a claim against neighbourhood demographic quotas for this reason.

VII. ASYMMETRIC RIGHTS

It is valuable at this point to draw together a theme emerging from the previous sections by addressing an objection that cuts across them. We have considered a series of normative values that plausibly could support a right to sort and have argued that they have some limit in the context of enclaves of advantage. Yet, we have expressed sympathy in several places for the idea that those who are disadvantaged may have a stronger claim to this right. Critics might baulk at this asymmetry. They might maintain that just as we should affirm equal rights to, say, voting or free speech, we should not give greater weight to the interests of some than others in choosing the neighbourhood context in which they live and form life plans.

We are resistant to this challenge and can begin to illustrate why by appealing to the following example:

Sorting support: across a society everyone has a modest level of opportunity to choose a neighbourhood that suits their various preferences about their residential surroundings, with whom to associate, and where they build a life plan. Then the government receives a windfall donation that must be spent supporting opportunities to sort. The amount allows them one of two policy options: (1) they could raise everyone’s level of opportunity a bit or (2) they could raise the opportunities of the disadvantaged a lot and do nothing to aid the opportunities of the better-off.

Our intuitions in this case are that the government should select Policy 2, and we think that, at least, two lines of theoretical reasoning support this. The first is thinking about social justice through an integrationist framework.⁷⁸ The central idea in this view is that individuals’ claims to goods in a domain are contingent on their broader level of advantage. To give examples from the literature developing this thesis, the thought is that someone’s entitlement to emissions or to a certain share of the gains of trade depends on how they fare in other respects, such as in their holdings of wealth or territory. We do not see any immediate reason to think that this approach would be ill-fitting in the case of supporting interests in neighbourhood choice and context. The second concerns reasoning about principles for transitioning from circumstances where there are injustices to improved states of affairs. In the literature on this, there is increasingly an emphasis on those in more advantaged positions bearing greater burdens than the disadvantaged.⁷⁹ Insofar as our discussion of neighbourhood demographic quotas began with rectifying a contemporary injustice, our sense is that this line of reasoning seems apt to the case. On both accounts, we think there is justification for the asymmetry in the right to sort that our arguments assert.

It is also worth noting that our analysis indicates that these frameworks extend further than advocates have previously argued. For example, in its main articulations, the focus of integrationism has been social and economic resources – emissions allowances and gains from trade and the sensitivity of these goods to background wealth, for instance. Our arguments extend this to the claim that there are also at least some rights and liberties that vary in force in relation to other aspects of advantage.

We must be careful not to overstate this. As we noted above, core liberties are often given priority over the distribution of opportunities and wealth. We have not sought to challenge this view in any fundamental respect. Even in our discussion of free association (which is often thought to fall within this group), our challenge centred only on its scope and the conditions that apply to (some instances of) its exercise.

Yet, neither should we overlook the significance of the conclusion that claims might vary in weight in relation to some aspects of certain rights. A main focus of this article has been the idea of a right to sort, and we have canvassed a series of normative values that could underpin it. In each of these cases, we observed some reason to think that there is a basis for asymmetry grounded in sensitivity to individuals’ overall level of advantage. In this respect, there is plausibly some terrain of rights and liberties that fall within an integrationist framework.

VIII. CONCLUSION

One of the main aims in this article has been to explore the normative defensibility of neighbourhood demographic quotas that set a minimum for the representation of disadvantaged groups in advantaged areas. While there is always more that it is possible to discuss, we hope to have made progress on this front in two respects. One is outlining the concerns that give some basis for advancing residential integration and, thereby, offer a presumptive rationale for neighbourhood demographic quotas. The other is to defuse some of the main objections critics might pose to this proposal and, more generally, to cast doubt on the permissibility of privileged individuals sorting into enclaves of advantage.

For all that we have said, questions remain about what policy specifics follow. Some of these relate to further details of what quota system to implement. Selecting between regulative and penalty systems would require resolving issues about comparative effectiveness and possible byproducts as well as potential normative differences in how they bear on property rights and whether there are reasons to prefer legal restrictions or taxation. Another issue is whether we might also bring other quota options under consideration. We set aside strategies of force, but perhaps what we have argued could support some such options, along similar lines to compulsory purchase policies states use when they build new transport networks. All of these issues, and many others, will affect what policy or combination of policies governments ought to adopt, and an article of this length cannot settle the full roster. However, it is important to note that any engagement with the subtle differences between quota options arises only if this broad policy platform is worth taking seriously. It is placing it on the agenda in this way that has oriented our approach here – seeking to show that there is enough rationale and defensibility against a series of broad objections to bring them forward as a strategy for tackling the inequalities of residential sorting.

This may prompt a further question as to why we have focused so extensively on neighbourhood demographic quotas. Those working in public policy may be quick to highlight that there are other strategies available for deconcentrating advantage, including some already employed in contemporary housing contexts. Perhaps the most common method here is inclusionary zoning, which involves requiring that new housing developments in advantaged areas market some stock at rates affordable to individuals on lower incomes.⁸⁰ One reason we have set this approach aside is that we have some concern that inclusionary zoning places too much of the burden of achieving desegregation unfairly.⁸¹ It asks very little of advantaged neighbourhoods other than that they make some housing lots available if they permit new developments, while, in cases where this happens, expecting the disadvantaged to bear the costs of moving, to face the dangers of relocating to environments often unaccommodating or hostile to their presence, and to assimilate to a new cultural context. There is, at least, one sense in which neighbourhood demographic quotas avoid such charges. For example, a penalty strategy places the onus of deconcentrating advantage appropriately on those most aptly place to bear the costs and presses them towards a proactive response to tackling underrepresentation. It goes some way to supporting this point that the Affirmatively Furthering Fair Housing rule that operated in a similar fashion had success in prompting robust action plans with measurable objectives associated with enhancing housing mobility.⁸²

However, our concern here is not to demonstrate that neighbourhood demographic quotas are superior to all other ways to advance residential integration, but, rather, to scope some of the possible arsenal and arguments for deconcentrating enclaves of advantage. This partly reflects our general scepticism that there are panaceas in this area and our belief that a multi-pronged strategy is likely needed. It was to this end that we focused on a policy that has not received much attention. In the same light, our exploration of the case has aimed to offer more of use to than in contest with advocates of other options. The broader point that we have defended is that arguments for resisting interventions to deconcentrate advantage in housing have been found wanting, and this finding is important for the cadre of policies we may use to advance integration. Thus, while we remain reserved on whether, when, and how exactly to deploy neighbourhood demographic quotas, we see significance in our defence of them as an option, if only because they show important limits on the advantaged having a right to sort.

ACKNOWLEDGEMENTS

We are grateful for discussion of this article with Matthew Clayton, Tom Geeson, Meg Huntingford, and Tom Parr and for comments from audiences at the ‘Just Housing Network Annual Meeting’ at the University of Hamburg, ‘Northern Bridge Political Theory Annual Meeting’ and the ‘Cities NUCoRE Research Seminar Series’ at Newcastle University, the ‘Economic Ethics Network Annual Meeting’ at the University of Oxford, the ‘Urban Political Theory Workshop’ at Queen Mary University of London, and the ‘Political Philosophy Colloquium’ at the University of Zurich. We are also grateful to two anonymous reviewers and the editorial guidance at Political Philosophy.

COMPETING INTERESTS

The authors declare that they have no competing interests.