Introduction

The international community has long recognized the principle that countries should share responsibility for hosting and supporting refugees.[2] The 2018 Global Compact on Refugees’ recognition of an “urgent need” for greater responsibility sharing across borders reflects widespread agreement that the existing distribution of responsibility among countries is unjust and unworkable.[3] A number of low- and middle-income countries in the Global South are the primary hosts of refugees,[4] while wealthier countries, such as Australia, many European countries, and the United States, deliberately prevent arrivals and strictly limit their acceptance of refugees.[5] The resulting distribution of responsibility tracks the proximity to the countries from which refugees flee, and parallels neither the capacities of states to support refugees, nor the extent of their moral duties to do so.[6]

The irony is that wealthy countries that have not taken their fair share of international responsibility have tried to ensure responsibility sharing within their borders. Responsibility for hosting and supporting refugees might be concentrated within a locality, province, or region. Or responsibility might be shared among communities within a country. Western European countries, for example, have adopted dispersal policies in part to achieve an equitable distribution of responsibility for providing resources and services to asylum seekers and refugees within their borders.[7] In the United States, Congress twice amended the Refugee Act of 1980 partly in response to complaints about “maldistribution” of refugees within the country.[8] National governments in these wealthy countries have treated demands for responsibility sharing within borders seriously, even as some schemes may harm refugees and undermine the aims of refugee law.

This Essay addresses questions of principle and policy that arise when we consider responsibility sharing for refugees within countries, not just among them. Scholars have explored the normative justifications for responsibility sharing across borders;[9] potential designs for multilateral systems of responsibility sharing;[10] the reasons why countries, particularly wealthy countries, have failed to share responsibility equitably;[11] and the consequences of these failures.[12] By contrast, domestic schemes of responsibility sharing for refugees have received far less attention than they should.[13]

Responsibility sharing within borders is important for a workable and just refugee system. The Global Compact itself suggests that the distribution of responsibility within borders is worth taking seriously, at least in cases involving “large” refugee movements.[14] Distributions of responsibility within borders may not track local capacities to support refugees. Turkey, for example, hosts more than 3.6 million Syrian refugees and lacks a system for responsibility sharing within its borders, with some municipalities struggling to provide services to their residents, refugees, and citizens alike.[15] At the same time, skepticism about demands for greater responsibility sharing within borders may be warranted, not only because these demands sometimes are xenophobic, but also because responsibility-sharing schemes may undermine refugee flourishing and the aims of the international refugee regime.

This Essay argues that there are important distinctions between responsibility sharing within borders and responsibility sharing across borders in principle and as a matter of policy design. One difference is that the case for solidarity is more straightforward within borders than across them. At the same time, current distributions of responsibility within borders are not as obviously unjust or unworkable as the global distribution. These differences are relevant to the criteria for distributing responsibility within borders. The capacity to host and support refugees is the most important criterion within borders, while concerns about consent and culpability are less important domestically than they are internationally.[16]

I. Responsibility Sharing Across Borders

Responsibility-sharing schemes aim at an equitable distribution of responsibility for hosting and supporting refugees. As of June 2020, 85 percent of the world’s refugees resided in developing countries.[17] This inequitable distribution of responsibility is not a new problem. Instead, it is baked into a system that imposes legal duties upon the first country to which a refugee flees, including the obligation of non-refoulement, while not mandating a system of responsibility sharing among countries.[18] International law has long recognized responsibility sharing in principle without requiring its implementation in practice.[19] Some wealthy countries make the problem worse by extending their migration controls beyond their formal borders, deterring asylum seekers, and dumping responsibility onto other countries.[20] Thus, the inequitable distribution of responsibility across borders persists, though that may start to change.[21]

Responsibility-sharing schemes to address this global inequity can involve people moving or sharing funds. Refugees may move across borders in a scheme to (re)distribute those seeking protection among different countries. One example is the U.S. system of refugee resettlement, under which the United Nations High Commissioner for Refugees identifies refugees for potential resettlement within the country.[22] Or countries may move money across borders to share in the costs of supporting and protecting refugees,[23] as with the European Union’s funding for Syrian refugees in Turkey.[24]

II. Distributions of Responsibility Within Borders

As with responsibility sharing across borders, responsibility sharing within borders can involve the movement of people (as in the case of policies to disperse asylum seekers throughout a country) or sharing money (as, for example, when a central government pools and distributes funds for services to resettled refugees). Using both mechanisms, national authorities, particularly in wealthy countries, have taken demands for responsibility sharing within borders seriously—in some cases, too seriously. Local demands for greater responsibility sharing may be xenophobic, and schemes designed to meet those demands may place refugees in communities where they are unlikely to flourish. Yet it is not hard to identify distributions of responsibility within borders that do not track local capacities to support refugees and that may fuel xenophobia.

Thus, on the one hand, responsibility sharing may be necessary for a workable and just system. On the other hand, schemes for responsibility sharing within borders may undermine refugee flourishing and the aims of refugee law and policy.

A. Concentrations of Responsibility within Borders

The Global Compact on Refugees’ concern for the capacities of local authorities during large-scale movements of asylum seekers points toward an argument for greater attention to responsibility sharing within borders. The Global Compact suggests that the distribution of responsibility within borders is important to ensure that localities have the capacity to host and support large numbers of refugees.[25] Local authorities, the Compact recognizes, “are often first responders to large-scale refugee situations.”[26] It therefore calls for “strengthen[ing] institutional capacities, infrastructure and accommodation at [the] local level.”[27] The Compact also identifies “local authorities” as “relevant stakeholders” that should be engaged in developing responsibility-sharing schemes.[28]

The Global Compact’s call for attention to local capacities to host and support refugees is borne out by recent examples. Consider, for instance, concerns about the concentration of Syrian refugees in North Lebanon and the Bekaa—regions “characterized by poverty and underdevelopment”[29]—or concerns about the initial concentration of Syrian refugees in southeastern Turkey.[30] Başak Kale and Murat Erdoğan have argued that refugee policy in Turkey shows “the importance of carrying responsibility sharing to [the] local level.”[31] Turkey hosts more than 3.6 million Syrians and lacks a centralized national system for refugee integration.[32] Syrian refugees have not settled evenly across provinces (or within them).[33] And that matters for municipal public services. Under Turkish law, the national government provides funding to municipalities from tax revenue in proportion to the number of “citizens” within municipal districts.[34] Syrian refugees, though classified as “fellow citizens” entitled to public services by most Turkish municipalities, do not count toward the budget allocation.[35] As a result, some municipalities struggle to provide adequate services for refugees and citizens.

Strains on schools in Turkey are one example. According to a 2018 International Crisis Group report, schools in Istanbul’s Sultangazi District and Ankara’s Altındağ District were stressed to the breaking point while trying to serve both Turkish and Syrian children.[36] In Sultangazi District, the arrival of six thousand Syrian students, all refugees, “dramatically increased the size of public school classes, reversing recent progress.”[37] And in Altındağ District, teachers were left to rely upon “Syrian children who have learnt a bit of Turkish to translate for them in the classroom.”[38] EU funding for education—that is, responsibility sharing across borders—helped address the issue. But it did not fully address the capacity mismatch arising from the uneven distribution of refugees within Turkey.[39]

In addition to this concern about local capacity to host and support refugees, another concern is xenophobic backlash to the refugee system. E. Tendayi Achiume has argued that the absence of an international system of responsibility sharing is a “xenophobic anxiety ratchet.”[40] Taking Syrian refugees in Turkey as an example, she has explained that the combination of regional concentration of refugees and inadequate responsibility sharing across borders creates conditions for increasing xenophobia and violence toward refugees.[41]

This concern does not stop at a country’s borders. To the contrary, some evidence suggests that the concentration of responsibility within borders may also ratchet up xenophobia. One recent study concluded that “mere exposure” to chaotic, poorly managed arrivals increases xenophobic opposition to asylum seekers, even when the asylum seekers are in transit to other countries.[42] This study compared the attitudes of Greek citizens on islands that briefly hosted massive numbers of Syrian and Afghan asylum seekers between 2015 and 2016 with those of Greeks on similar islands that did not.[43] It found that this direct exposure increased both anti-asylum and anti-Muslim attitudes, even though the refugees almost always left the islands within twenty-four hours on their way to other countries.[44] By concentrating the arrival of asylum seekers on a few islands, rather than distributing their arrival around the country, the Greek government created the appearance of a crisis that fueled xenophobia.[45] As in the case of global concentrations of responsibility for refugees, the distribution of responsibility within borders may fuel xenophobic opposition to the refugee system.

B. Schemes for Sharing Responsibility within Borders

It is not clear, however, that responsibility sharing within borders is a solution to the problem of xenophobia. Local demands for greater responsibility may be xenophobic and opposed to the international refugee system.[46] Moreover, responsibility-sharing schemes can harm refugees.

European countries have taken demands for responsibility sharing within borders seriously through both the movement of people and sharing money. Dispersal policies in Denmark,[47] Germany,[48] Italy,[49] the Netherlands,[50] Sweden,[51] and the United Kingdom (U.K.)[52] have all been justified as necessary to achieve an equitable distribution of the responsibility for providing resources and services to refugees.[53] The dispersal systems take different forms. Some use rigid quotas and residence requirements. In 1982, for example, Germany codified a quota system to allocate asylum seekers among the Länder and require them to reside in a particular place.[54] Other systems condition assistance on dispersal. The U.K., for instance, adopted in 1999 a “housing-led” system that conditioned an asylum seeker’s eligibility for housing assistance on accepting a dispersed placement.[55]

Critics of dispersal policies have argued that they have costs, including denying refugee autonomy, depriving refugees of kinship and social networks within which they might flourish, and triggering xenophobic backlash rather than preventing it.[56] Germany’s system of quota-based dispersal, as Christina Boswell has discussed, was “seen more or less explicitly as a means of reducing the attractiveness of seeking asylum in Germany.”[57] Vaughn Robinson, Roger Andersson, and Sako Musterd have argued that dispersal schemes in the Netherlands, Sweden, and the U.K. harm refugees by excluding them from communities with the capacities to support them, and thus “privilege[] the perspective of the majority population by proposing a solution that inconveniences them least.”[58]

Within the United States, commentators have begun to discuss “refugee federalism,” which raises issues of responsibility sharing within borders.[59] As the United States shows, demands for greater responsibility sharing may express xenophobic anxiety about refugees. For example, some state officials in the United States responded to the November 2015 terrorist attack in Paris by trying to bar the resettlement of Syrian refugees within their borders. Their rhetoric played upon tropes from the War on Terror and characterized refugees as a public safety and national security threat.[60] As Stella Burch Elias has argued, these state officials’ legal and policy arguments reflected “unprecedented levels of hostility” within the United States toward refugee resettlement.[61]

Under the Refugee Act of 1980, the resettlement process requires federal agencies and resettlement agencies to “consult regularly (not less often than quarterly)” with state and local governments concerning the placement of refugees.[62] In addition, the Secretary of State must consider four factors when developing policies for resettlement, including the “proportion of refugees and comparable entrants in the population in the area”; the availability of employment, housing, and services in the area; the likelihood that refugees will become self-sufficient if placed in an area; and the “secondary migration of refugees to and from the area that is likely to occur.”[63] Congress enacted the consultation requirement and four factors through amendments to the Refugee Act in response to complaints about, among other things, the “maldistribution” of refugees within the United States.[64] The federal government helps share the costs of refugee resettlement by providing funding to states, localities, and resettlement agencies.[65] Even so, some have argued that state and local governments should have rights to refuse the resettlement of refugees within their borders.[66] In 2019, then-President Donald Trump issued an executive order permitting states and localities to do just that.[67] The federal courts, however, enjoined the order’s implementation, citing concerns that requiring state and local consent to resettlement would have dire consequences for the United States’ involvement in the international refugee system.[68]

In contrast to the global distribution of responsibility for refugees, it is far from obvious that there are problems with the distribution of responsibility for refugee resettlement within the United States. The question of refugee resettlement is not one of local capacity to support large-scale movements of asylum seekers. While the United States has traditionally been a leader in refugee resettlement,[69] the number of resettled refugees remains strikingly small next to the total number of refugees globally. In FY2019, for example, the State of Texas, with a population of roughly twenty-nine million people,[70] was the top destination for refugee resettlement within the United States. 2,500 refugees resettled there, which was more than Washington, New York, and California, each of which accepted between 1,800 and 1,900 refugees.[71] That distribution is not clearly unjust, much less unworkable.

Thus, while distributions of responsibility within borders may be important for a workable and just refugee system, there are reasons for skepticism about some demands for greater responsibility sharing within borders. Moreover, the costs and harms of responsibility-sharing schemes suggest that we need to ask what the justification for responsibility sharing within borders is and what the criteria and limits should be. The next Part turns to those questions.

III. Normative Dimensions of Responsibility Sharing Within Borders

Responsibility sharing within borders may be more administrable than responsibility sharing across borders and more straightforwardly justified by a principle of solidarity. The principle of national solidarity and the relative administrability of national schemes suggest that an equitable distribution of responsibility sharing within borders is a goal worth pursuing. In evaluating responsibility sharing as a goal, however, we should emphasize refugee rights and the problem of xenophobia. Both provide reasons for rejecting some demands for responsibility sharing within borders and selecting some schemes for responsibility sharing over others. The norms that ground responsibility sharing within borders point toward capacity as the focal point for schemes to distribute responsibility equitably. Responsibility-sharing schemes on the international plane are founded upon principles of consent, culpability, and capacity. We might think that these principles matter domestically in the same way they matter internationally. But consent and culpability play lesser roles within borders than they do across them.

A. Responsibility Sharing as a Normative Goal

The greater administrability of responsibility-sharing schemes within borders distinguishes it from responsibility sharing across borders. Both sharing money and sharing responsibility through the movement of people are more readily achieved within borders than across them. Domestic taxing and spending to subsidize local hosting and supporting of refugees is easier to achieve than international redistribution of funds for refugees. So too is the movement of people within a country rather than across international borders. This distinction suggests that responsibility sharing is more readily achievable within borders than across them, with the comparatively lower administrative costs counting in favor of responsibility-sharing schemes.

To consider whether responsibility sharing within borders is not only administrable, but also an important normative goal, it is useful to start with the justification for responsibility sharing across borders. The most encompassing normative justification for responsibility sharing across borders is that all states share a general duty toward refugees and should act in solidarity in fulfilling that duty. The general duty of states to protect and support refugees is necessary for the legitimacy of the international system of recognizing states as territorial sovereigns.[72] Sarah Song has argued that this legitimacy condition may be derived from a global original position or from states’ mutual participation in recognizing each other’s sovereignty.[73] As she puts it, “[i]t is only in a world carved into states that people can become refugees.”[74] In such a world, it is fair to expect states “to step in” to protect refugees whose home states have failed to protect their human rights.[75] This shared responsibility does not depend upon signing onto the Refugee Convention. Rather, “even states that have not signed onto the [Convention] bear collective responsibility for refugees because they participate in and benefit from the state system.”[76] When we think about state legitimacy in this way, Rainer Bauböck has argued, it follows that solidarity in the protection and support of refugees is an important principle.[77]

If anything, the case for solidarity in the protection and support of refugees is more straightforward within borders than across them. Responsibility sharing within borders may be grounded in national solidarity: the idea that the members of a nation, either individually or collectively through their participation in subnational communities, owe a duty to their fellow members to share the responsibility for collective projects. As Oregon Governor Kate Brown put it in 2016 when announcing that her state would welcome Syrian refugees: “The words on the Statue of Liberty apply in Oregon just as they do in every other state.”[78] All states, in other words, should share in the American project of assisting refugees.

One objection is that a commitment to national solidarity does not determine which projects should be shared ones of the nation. One might argue, for instance, that national solidarity secures justice among citizens, not justice toward strangers.[79] On that view, one might question whether national solidarity entails responsibility sharing for refugees within borders.

In response to this objection, there are two accounts of responsibility toward refugees that explain why it is a shared national project. The first is a humanitarian account under which a collective duty runs from members of a nation toward refugees. The duty-bearers should distribute this collective duty fairly among themselves. In this account, the duty to assist refugees is akin to a duty to rescue, under which a person who has the capacity to rescue another has the duty to do so in an emergency.[80] Duties to rescue can be shared among multiple potential rescuers who each have the capacity to intervene. When there are multiple potential rescuers—when there are, say, multiple potential host communities with the capacity to support refugees—they have a collective responsibility that they may divide, so long as they do so fairly.[81] When those multiple potential rescuers are strangers to one another, we might question whether they share a collective duty to rescue. It is one thing to ask whether people share a duty when they happen to be present at the scene of an emergency but have no other connection with one another. It is quite a different matter, however, to ask whether members of a nation-state, who are connected by their common membership, share a collective duty to rescue. Multiple potential rescuers who have formed a collective may bear a collective humanitarian duty even if multiple potential rescuers who have not formed a collective do not share this duty.[82] A humanitarian duty to rescue, in other words, is the sort of project that should be shared among the members of a collective, such as a nation-state.

The second account is grounded in the legitimacy conditions for state sovereignty internationally. If the legitimacy of states on the international plane depends in part upon their general duty toward refugees, then the constituent political units of a state should share responsibility for fulfilling this duty.[83] The argument has its most obvious application within federal systems where authority is divided subnationally. Consider, for example, a case where a particular country—say, the United States—has a responsibility to accept the resettlement of some number of refugees. If each of its subnational units—for instance, all the states within the United States—refused resettlement, the country would be in violation of its moral duty. A subnational unit’s duty to take some responsibility for refugees thus may be derived from the legitimacy conditions for state sovereignty.

B. Criteria for Sharing Responsibility

1. Capacity. — The norms that ground the importance of responsibility sharing within borders point toward capacity as an essential criterion. The burdens and benefits of national projects rarely fall evenly across communities, and a commitment to solidarity does not require that responsibility for refugees be shared irrespective of a community’s capacity to support them. The humanitarian account begins with a duty toward refugees and makes the capacity question relevant at the stage of distributing responsibility among potential rescuers. Humanitarianism does not require that all potential rescuers contribute in the same amount or way toward the rescue. Rather, humanitarianism is sensitive to differences in capacities among potential rescuers, with capacity as a measure of fairness in the distribution of responsibility.[84] And to the extent that responsibility sharing within borders is derivative of the legitimacy conditions of state sovereignty, capacity is an important criterion, too. As Bauböck put it for the international context, “[i]f refugee protection is a legitimacy repair mechanism in the international state system, then all states ought to contribute in proportion to their resources.”[85]

Thus, capacity is an essential criterion in both the international and domestic contexts. States’ capacities to support and integrate refugees are “essential component[s]” of responsibility sharing across borders.[86] So too, distributing responsibility in light of capacity is essential to any fair scheme of responsibility sharing within borders.

“Capacity” is not self-defining, of course. Some material components, such as the capacity to provide public services, are clearly relevant. Consider again the case of schools in Istanbul’s Sultangazi District and Ankara’s Altındağ District.[87] Both school systems were struggling to provide adequate education to Turkish and Syrian children. This example raises concerns about justice for citizens within those host communities who receive fewer, or less effective, public services than citizens whose communities are not hosting refugees. The lack of capacity also raises concerns about justice for refugees themselves. The problem, to borrow a phrase from Hiroshi Motomura, is a policy of municipal funding that “fail[s] to funnel tax revenues fairly to the government entities that need them most.”[88]

The harder question is whether “capacity” should include a community’s willingness to accept refugees or resistance to them. Responsibility-sharing schemes within borders are sometimes justified by reference to the risk of xenophobic backlash within communities.[89] This justification is particularly important for schemes that disperse refugees across a country without honoring their preferences.

However, I agree with Song that capacity should not be defined by reference to a community’s level of xenophobia.[90] As Song put it, “[s]uch attitudes say more about the willingness, not the capacity, of states to take in refugees.”[91] The same is true with respect to local communities within states. Xenophobic opposition is an illegitimate normative criterion for responsibility sharing “because [it] violate[s] the democratic value of equal respect.”[92] It follows that national responsibility-sharing policies should not aim to distribute responsibility within borders by defining capacity in terms of willingness to accept refugees. In designing such policies, therefore, it is important to disentangle demands for greater responsibility sharing within borders from xenophobic anxiety about refugees.

2. Culpability. — Culpability plays an important role in international responsibility sharing because some states owe special duties to certain refugees. A refugee’s home state is responsible for making them a refugee, of course. But other states may be culpable as well. States may owe special duties to particular refugees when they are culpable for making them refugees, such as when one state makes war upon another, and thus contributes to instability and the threat of persecution.[93] For example, the United States owes a special duty to Afghan interpreters who assisted its military during the United States’ twenty-year war in Afghanistan.[94] When a state owes a “refugee debt,” it is unjust to rely upon other states to pay for it.[95]

There are limited ways in which we might take culpability into account when designing a system of responsibility sharing within borders. Culpability might matter when there are individuals or corporations residing within a state that are uniquely responsible for creating conditions abroad that forced refugees to flee their home states. Refugees may be entitled to an avenue of redress against these culpable persons. Consider human rights litigation in the United States under the Alien Tort Statute or state tort law, which may provide avenues for redress for human rights violations abroad.[96] We might view the availability of law for redress of those wrongs as a component of a scheme of responsibility sharing. For example, we might imagine a system that reduces the amount of public benefits to a refugee who has received compensation through tort litigation.

Culpability cannot, however, do the same work within borders as it does across them. Responsibility sharing within borders does not flow directly from the special duties that some states owe to specific refugees. Opening up the black box of the state to attribute responsibility to subnational governments or communities for their role in displacing particular refugees would make an already difficult question much more difficult. On administrative grounds, then, there is good reason not to make culpability a guiding criterion for responsibility sharing within borders. In principle, moreover, making culpability central would be hard to square with the principle of state sovereignty that—for better or for worse[97]—is at the center of the international refugee system.

3. Consent. — Consent of states is fundamental to the current international approach to responsibility sharing across borders. The Global Compact on Refugees “is not legally binding” and is “operationalized through voluntary contributions.”[98] States and relevant stakeholders determine those contributions, “taking into account their national realities, capacities and levels of development, and respecting national policies and priorities.”[99]

Should consent be fundamental to schemes of responsibility sharing within borders? In 2019, as Part II noted, then-President Trump tried to make consent constitutive of the refugee resettlement scheme within the United States. His executive order directed the “Federal Government” to “resettle refugees only in those jurisdictions in which both the State and local governments have consented to receive refugees,” with “limited exceptions.”[100]

One normative argument for making consent a constitutive feature would point to a right of collective self-determination. Within the United States, that argument might appeal to federalism. Perhaps Texas, as a sovereign state within a federal system, should have the right to refuse refugees on principle. That is not what U.S. law says, of course. But should it?

Concretely, what is at stake here is whether subnational communities can refuse to participate in schemes that involve sharing people. We might imagine an argument that the central government should obtain the consent of a subnational community before taxing its members to share money for the care and support of refugees. But the more likely scenario is one in which a community objects to the settlement of refugees within it.

What are we to make of the argument that communities within states should have the right to opt out of responsibility-sharing schemes, specifically those involving the settlement of refugees within particular places? The question turns not only upon which collectivities are “peoples” with rights of self-determination, but also upon what those rights to collective self-determination necessarily entail.[101] I cannot resolve those questions in this Essay. But it is possible to clarify how they bear upon responsibility sharing within borders.

In principle, there is a mismatch between the demand for a local right to exclude refugees and a professed concern about responsibility sharing within borders. As Part II explained, responsibility sharing is a scheme premised upon solidarity in the sharing of a collective burden. Such a scheme cannot “operate on the principle that each participant can determine and limit her contributions unilaterally.”[102] It is odd for a local community to appeal to solidarity—we are all in this together and must each bear our fair share—as a basis for the right to choose whether to contribute at all.

Domestic law may, of course, require consent of subnational communities to responsibility sharing. Some domestic constitutional systems may require, or at least support, a consent principle. Some have argued as much with respect to U.S. federalism. This Essay’s argument goes no further than to question the normative case for such a consent requirement within borders.

There is, however, a purely instrumental argument for consent to responsibility sharing within borders. Within the United States, James Xi has argued for amendments to the system of refugee resettlement that would afford localities the right to exclude refugees and require the federal government to purchase their consent.[103] Xi contended that this market-based approach would be the most efficient way to match refugees with communities that have the capacity to support them.[104] My arguments do not rule out that design option, nor do they bear upon the debate about the appropriateness of market mechanisms in refugee law.[105] Rather, they show that there is good reason to be skeptical that consent has independent normative value in the design of schemes of responsibility sharing within borders.

C. Refugee Autonomy and Flourishing

As Part II.B described, responsibility-sharing schemes may harm refugees by placing them in communities where they are unlikely to flourish. This harm might be discounted if refugees have no right to settle in any particular country, much less a right to settle in a particular community where they want to be and are likely to flourish. For this reason, refugee autonomy and flourishing are sometimes disregarded in discussions of responsibility-sharing schemes.

Refugee rights should not be discounted, however. The Global Compact commits the international community to “strengthened cooperation and solidarity” not only with “affected host countries,” but also “with refugees.”[106] This normative commitment to solidarity and meaningful engagement with refugees makes for more “effective” responses to refugee movements.[107] Therefore, states should include refugees in policymaking processes and the design of “appropriate, accessible and inclusive responses.”[108] The Global Compact thus treats refugees as agents whose voices and choices matter.

There is a normative argument for treating refugees as agents. As summarized by Matthew Gibney, this normative argument takes account of what refugees have lost and why their choices matter.[109] What a refugee has lost is “her social world,” that is, the many relationships and associations within which she has organized her life.[110] To repair this loss requires “physical security and basic rights” and also “a place where she can rebuild a meaningful social world.”[111] To determine where such a place might be, we should ask refugees themselves.[112] Doing so will not only increase the likelihood of refugee flourishing, but will also help to repair the sense of profound loss and disempowerment that refugees typically experience.[113]

Taking refugee autonomy and flourishing seriously has concrete implications for the design of responsibility-sharing schemes. It suggests that sharing money is a reasonable starting point for addressing capacity mismatches within borders. Responsibility-sharing schemes that disperse refugees may place them in communities where they are unlikely to flourish. This is especially true if their movements are restricted, such as in schemes that confine asylum seekers to certain places within a country.[114] Just as there may be particular states within which refugees are more likely to flourish, so too there may be particular communities within countries where they are more likely to flourish. If those communities to which refugees have special ties[115] lack capacity to provide services to their citizens and to additional refugees, the presumptive solution should be additional funding for services, not a dispersal scheme that places and even confines refugees in communities where they are more likely to struggle. Not only do dispersal policies have administrative costs, they also can deny refugees agency and undermine their flourishing.[116] Those costs are not presumptively unjustified if the central state has the capacity to compensate those local communities that take more refugees.

IV. Complications and Applications

Thus far, this Essay has argued that we should open the black box of the state to think through responsibility sharing within borders. It has distinguished responsibility for refugees in the international and domestic contexts. This Part introduces some complications to the analysis and identifies some issues for future research. It differentiates refugee resettlement from large movements of asylum seekers, raises the role of private actors in responsibility-sharing schemes, and discusses refugee movement within borders as it bears upon normative analysis of responsibility sharing.

A. Distinguishing Contexts: Refugee Resettlement and Large Movements of Asylum Seekers

The question of responsibility sharing arises in a variety of contexts. Some involve small numbers of refugees, as with demands for greater responsibility sharing for resettled refugees within the United States. Others involve the distribution of responsibility for protecting and supporting large numbers of asylum seekers that arrive at a country’s borders, such as with the many Syrians who have fled to Turkey during the Syrian civil war that began in 2011. Within the United States, states have objected to the resettlement of small numbers of refugees. In contrast, municipal districts in Turkey have been unable to provide adequate public services, such as public education, without more funding for the support of refugees.

This Essay’s normative analysis helps distinguish among these different contexts. For refugee resettlement within the United States, demands that the federal government obtain the consent of states are presented in the rhetoric of responsibility sharing. But the numbers involved are so small as to belie this rhetoric. Rather, these demands for greater responsibility sharing express xenophobic anxiety about refugees. To grant states a veto over refugee resettlement would enshrine xenophobia into refugee law. As Song argued, this approach cannot be reconciled with the basic democratic principle of equal respect.[117]

By contrast, large movements of asylum seekers into neighboring countries, such as Syrians fleeing into Turkey, present different issues that require different responses. One issue concerns the material capacity of municipal districts to provide services to refugees and citizens alike. Sharing money differently is a potential solution to this problem. The lack of such a solution may contribute to xenophobic backlash; concentrating large numbers of asylum seekers in particular places and then restricting their movement may exacerbate the perception of a crisis and thus fuel xenophobia.[118] Sharing money in this context may reduce the risk of xenophobia without violating a basic principle of equal respect.

B. Private Responsibility Sharing

There is another way in which contexts of responsibility sharing may differ across countries: the relative roles of public and private actors in distributing responsibility for refugees. Within the United States, for example, the resettlement of overseas refugees depends upon the work of nine voluntary agencies (VOLAGS) that receive public funding and resettle refugees within states. Under the terms of the federal program, VOLAGS must consult with state and local authorities concerning resettlement decisions. They match refugees with particular locales based upon the refugees’ needs and the services that the responsible settlement agency is able to provide in light of local capacities to support refugee flourishing.[119] This approach tracks the normative criteria discussed above.

The role of private organizations presents an important set of questions that become sharply defined when we consider responsibility sharing within borders. The Global Compact’s commitment to a multi-stakeholder approach highlights the importance of “private responsibility sharing,” or “public-private responsibility sharing.” This Essay’s analysis suggests that a normative analysis of this question should focus upon whether public and private responsibility contribute to solving capacity mismatches that undermine refugee flourishing and the aims of refugee system.

Whether privatization is a mechanism of responsibility dumping is an important question worth considering.[120] The question has arisen recently within the United States. In the past few months, the U.S. Government has left significant responsibility to private individuals for the resettlement of Afghans, such as U.S. veterans of the twenty-year war in Afghanistan who have assisted Afghans seeking refuge after the Taliban entered Kabul in August 2021.[121] The “unprecedented” use of private sponsorships may be necessary because of a lack of capacity among local resettlement offices within the United States.[122] But problems may arise from overreliance upon private organizations and individuals.[123] For example, Meryll Dean and Miki Nagashima have contrasted Japan’s approach to responsibility sharing across borders, which has involved the donation of “vast sums” for the support of refugees overseas, with its approach to responsibility sharing for the few refugees within its borders, which involves “narrow and ungenerous protection” from the government and a dumping of the responsibility onto NGOs.[124] Thus, what appears to be private responsibility sharing may prove to be responsibility dumping that shirks shared humanitarian responsibilities and undermines refugee flourishing.

C. Refugee Movement within Borders

This Essay has critiqued responsibility-sharing schemes that disperse refugees within a country without regard to their preferences and potential for flourishing. This critique is most applicable to schemes that not only assign but also confine refugees within particular places. In the case of large movements, some countries have confined asylum seekers in camps, including by placing them indefinitely in “prison-like conditions.”[125] For instance, one refugee from Somalia thus described the Dadaab refugee camp in Kenya: “As soon as we crossed the border we were registered, put in an isolated camp and basically quarantined from the rest of the Kenyan society.”[126] Large-scale refugee scenarios are not the only example of national policies that limit where refugees can move. Some Western European countries have also limited the movement of asylum seekers as part of responsibility-sharing schemes.[127]

Not all national policies limit refugee movements within borders. The evidence on refugee mobility within countries that permit it is mixed, with some studies suggesting that a significant number of asylum seekers and resettled refugees move within a few years.[128] Refugee movement within borders may mitigate the harms of responsibility-sharing schemes and thus lower the stakes of the debate. But there are enough examples of schemes that confine refugees, and enough reasons to doubt that refugee mobility eliminates the potential harms of dispersing persons, to make the problem one worth further research.[129]

Conclusion: Responsibility Sharing Across and Within Borders

Responsibility sharing within borders is an important goal because of both refugee flourishing and fairness within states. It is not, however, the same goal as responsibility sharing across borders. There are normative distinctions between the two goals that bear upon the design of responsibility-sharing schemes, as this Essay has argued.

This Essay’s analysis also suggests, however, that we need to think about responsibility sharing in an integrated way. To think through responsibility sharing within borders requires thinking through responsibility sharing across borders, and vice versa. For example, it requires thinking through whether demands for greater responsibility sharing within borders should be met with skepticism when a country is failing to take in its fair share of responsibility across borders. It requires thinking about the consequences of responsibility-sharing schemes within borders for refugee flourishing. Finally, it requires considering the aims of the international system and developing institutions in order to facilitate multi-stakeholder thinking within and across borders.

© 2022 Seth Davis, Professor of Law, University of California, Berkeley, School of Law. Email: sethdavis@berkeley.edu. I would like to thank Leti Volpp for her incisive commentary on an earlier draft, as well as E. Tendayi Achiume, Jennifer Chacón, Katerina Linos, Paul Messick, and Ayelet Shachar for their helpful feedback during and after the symposium. DOI: https://doi.org/10.15779/Z38HQ3S04S.

2. The principle of responsibility sharing stretches back to the 1951 Refugee Convention. See Convention Relating to the Status of Refugees pmbl., para. 4, July 28, 1951, 189 U.N.T.S. 137 (recognizing that “the grant of asylum may place unduly heavy burdens on certain countries, and that a satisfactory solution . . . cannot therefore be achieved without international co-operation”). ↑

3. Rep. of the U.N. High Comm’r for Refugees, Global Compact on Refugees, ¶ 1, U.N. Doc. A/73/12 (Pt. II) (Aug. 2, 2018) [hereinafter Refugee Compact] (recognizing “an urgent need for more equitable sharing of the burden and responsibility for hosting and supporting the world’s refugees”). On the widespread agreement concerning the current global distribution of responsibility, see T. Alexander Aleinikoff & Leah Zamore, The Arc of Protection: Reforming the International Refugee Regime 13 (2019) (“The absence of any [implementation] mechanism [for the principle of responsibility sharing] is a driving force behind the [refugee] regime’s present-day dysfunction.”); E. Tendayi Achiume, Governing Xenophobia, 51 Vand. J. Transnat’l L. 333, 376 (2018) (describing the “dynamic of grossly asymmetrical refugee distribution and support” under which there is a “regional concentration of refugees in the Third World”). ↑

4. See Press Release, U.N. High Comm’r on Refugees, 1 Per Cent of Humanity Displaced: UNHCR Global Trends Report (June 18, 2020), https://www.unhcr.org/en-us/news/press/2020/6/5ee9db2e4/1-cent-humanity-displaced-unhcr-global-trends-report.html [https://perma.cc/KG82-M2NJ] (reporting that as of June 2020, 85 percent of refugees “are in developing countries, generally a country neighbouring the one they fled”). ↑

5. See Ayelet Shachar, Instruments of Evasion: The Global Dispersion of Rights-Restricting Migration Policies, 110 Calif. L. Rev. 967, 969 (2022). ↑

6. See infra notes 16–17 and accompanying text. ↑

7. See infra notes 46–57 and accompanying text (summarizing these policies). ↑

8. See James Y. Xi, Note, Refugee Resettlement Federalism, 69 Stan. L. Rev. 1197, 1208–11 (2017) (quoting H.R. Rep. No. 97-541, at 11 (1982)) (discussing 1982 and 1986 amendments to Refugee Act). ↑

9. See, e.g., David Miller, Strangers in Our Midst: The Political Philosophy of Immigration 86–92 (2016); Sarah Song, Immigration and Democracy 117–27 (2018); Rainer Bauböck, Refugee Protection and Burden-Sharing in the European Union, 56 J. Common Mkt. Stud. 141, 143–44 (2018); Matthew J. Gibney, Refugees and Justice Between States, 14 Eur. J. Pol. Theory 448, 456–57 (2015); Tally Kritzman-Amir, Not in My Backyard: On the Morality of Responsibility Sharing in Refugee Law, 34 Brook. J. Int’l L. 355, 359–72 (2009). ↑

10. See Peter H. Schuck, Refugee Burden-Sharing: A Modest Proposal, 22 Yale J. Int’l L. 243, 247 (1997); James C. Hathaway & R. Alexander Neve, Making International Refugee Law Relevant Again: A Proposal for Collectivized and Solution-Oriented Protection, 10 Harv. Hum. Rts. J. 115, 118 (1997); see also Deborah Anker, Joan Fitzpatrick & Andrew Shacknove, Crisis and Cure: A Reply to Hathaway/Neve and Schuck, 11 Harv. Hum. Rts. J. 295, 300–01 (1998) (critiquing Hathaway, Neve, and Schuck’s proposed frameworks). ↑

11. See, e.g., Ashley Binetti Armstrong, You Shall Not Pass!: How the Dublin System Fueled Fortress Europe, 20 Chi. J. Int’l L. 332, 336 (2020); Tally Kritzman-Amir & Yonatan Berman, Responsibility Sharing and the Rights of Refugees: The Case of Israel, 41 Geo. Wash. Int’l L. Rev. 619, 622 (2010); Eiko R. Thielemann, Between Interests and Norms: Explaining Burden-Sharing in the European Union, 16 J. Refugee Stud. 253, 254 (2003). ↑

12. See, e.g., Aleinikoff & Zamore, supra note 2, at 13; Achiume, supra note 2, at 374–78. ↑

13. The existing literature offers case studies of systems for sharing responsibility within borders. See, e.g., Vaughn Robinson, Roger Andersson & Sako Musterd, Spreading the ‘Burden’? A Review of Policies to Disperse Asylum Seekers and Refugees 1–2 (2003). Some scholars have looked to such policies as potential models for responsibility sharing across borders. See, e.g., Walter Bartl, Institutionalization of a Formalized Intergovernmental Transfer Scheme for Asylum Seekers in Germany: The Königstein Key as an Indicator of Federal Justice, 34 J. Refugee Stud. 2613, 2616 (2019); Christina Boswell, Burden-Sharing in the European Union: Lessons from the German and UK Experience, 16 J. Refugee Stud. 316, 319 (2003). ↑

14. See Refugee Compact, supra note 2, ¶ 12 (explaining that Comprehensive Refugee Response Framework “relates specifically to large refugee situations”); id. ¶ 37 (recognizing that local authorities “are often first responders to large-scale refugee situations”). ↑

15. See Başak Kale & Murat Erdoğan, The Impact of GCR on Local Governments and Syrian Refugees in Turkey, 57 Int’l Migration 224, 224–25 (2019). ↑

16. *For discussion of criteria for responsibility sharing across borders, see Michael Doyle, Janine Prantl & Mark J. Wood, Principles for Responsibility Sharing: Proximity, Culpability and Capability*, 110 Calif. L. Rev. 935, 955–58 (2022). ↑

17. See Press Release, U.N. High Comm’r on Refugees, supra note 3. ↑

18. See Miller, supra note 8, at 83–84. ↑

19. In 1951, the Refugee Convention recognized a need for “international co-operation” in light of the “unduly heavy burdens” that a “grant of asylum may place . . . on certain countries.” See Convention Relating to the Status of Refugees pmbl., supra note 1, para. 4. In recent years, the U.N. General Assembly’s New York Declaration, as well as the ensuing Global Compact on Refugees and the Global Compact for Safe, Orderly, and Regular Migration, have restated the importance of cooperation and responsibility sharing among countries with respect to refugees and other migrants. See G.A. Res. 71/1, New York Declaration for Refugees and Migrants, ¶ 11 (Oct. 3, 2016) (recognizing that countries have shared responsibility with respect to refugees and migrants); Refugee Compact, supra note 2, ¶ 1; G.A. Res. 73/195, annex, Global Compact for Safe, Orderly and Regular Migration, ¶ 15 (Dec. 19, 2018) (identifying “international cooperation” as one “guiding principle[]”). ↑

20. See Shachar, supra note 4, at 969 (“[P]rosperous nations increasingly rely on sophisticated legal tools to expand the reach of [their migration and] border control[.]”); Katerina Linos & Elena Chachko, Refugee Responsibility Sharing or Responsibility Dumping?, 110 Calif. L. Rev. 897, 900–01 (2022) (offering framework to distinguish responsibility sharing from responsibility dumping). ↑

21. See Refugee Compact, supra note 2, ¶ 1 (recognizing the “urgent need” for greater responsibility sharing); Linos & Chachko supra note 19, at (arguing that “[o]ver the last several years, concrete responsibility-sharing arrangements have been developed and implemented, translating the concept from theory to practice”). But cf. BS Chimni, Global Compact on Refugees: One Step Forward, Two Steps Back, 30 Int’l J. Refugee L. 630, 630 (2018) (“[T]he Refugee Compact is a flawed text for several reasons[, including that it] . . . is short on real mechanisms for responsibility sharing . . . .”). ↑

22. See HIAS, Inc. v. Trump, 415 F. Supp. 3d 669, 673–74 (D. Md. 2020), aff’d, 985 F.3d 309 (4th Cir. 2021). ↑

23. This discussion tracks Noll’s distinction between “sharing people” and “sharing money.” Gregor Noll, Risky Games? A Theoretical Approach to Burden-Sharing in the Asylum Field, 16 J. Refugee Stud. 236, 243 (2003). I bracket “sharing norms,” the harmonization of refugee laws across borders. See id. ↑

24. See European Council Press Release 144/16, EU-Turkey Statement (Mar. 18, 2016) https://www.consilium.europa.eu/en/press/press-releases/2016/03/18/eu-turkey-statement/pdf [https://perma.cc/9A27-ZZ9N]. ↑

25. See Refugee Compact, supra note 2, ¶ 12. ↑

26. Id. ¶ 37. ↑

27. Id. ↑

28. Id. ¶ 3. In making this call for a “multi-stakeholder and partnership approach,” the Compact “recogniz[es] the primary responsibility and sovereignty of States” and envisions “a supportive and catalytic role” for the U.N. High Commissioner for Refugees. Id. ¶ 33. ↑

29. See Sam van Vliet & Guita Hourani, Regional Differences in the Conditions of Syrian Refugees in Lebanon, Civ. Soc’y Knowledge Ctr. (Apr. 1, 2014), https://civilsociety-centre.org/paper/regional-differences-conditions-syrian-refugees-lebanon [https://perma.cc/2HAV-8BB3]. ↑

30. See Alan Makovsky, Turkey’s Refugee Dilemma: Tiptoeing Toward Integration, Ctr. for Am. Progress (Mar. 13, 2019), https://www.americanprogress.org/article/turkeys-refugee-dilemma/ [https://perma.cc/C623-XEW6]. ↑

31. Kale & Erdoğan, supra note 14, at 231. ↑

32. Id. at 224–25. ↑

33. M. Murat Erdoğan, Konrad-Adenauer-Stiftung, Syrian Refugees in Turkey 15 (2019), https://www.kas.de/documents/283907/7339115/Syrian+Refugees+in+Turkey.pdf/5d3d4091-e56d-3c42-2a9c-4e7f5d98706f?version=1.0&t=1571303379232 [https://perma.cc/22G7-NEJN] (“[T]here are critical differences in the distribution of Syrians throughout the country, politically and proportionally.”). ↑

34. Kale & Erdoğan, supra note 14, at 229. ↑

35. Id. ↑

36. Int’l Crisis Grp., Turkey’s Syrian Refugees: Defusing Metropolitan Tensions 19 (2018). ↑

37. Id. ↑

38. Id. (quoting an NGO representative). ↑

39. See id. ↑

40. Achiume, supra note 2, at 379. ↑

41. Id. at 377–78. ↑

42. Dominik Hangartner, Elias Dinas, Moritz Marbach, Konstantinos Matakos & Dimitrios Xefteris, Does Exposure to the Refugee Crisis Make Natives More Hostile?, 113 Am. Pol. Sci. Rev. 442, 442 (2019). ↑

43. See id. at 443. ↑

44. Id. at 444. ↑

45. See id. (“[M]ere exposure to the chaos of the refugee crisis generates a feeling of threat that can activate latent predispositions against immigrants and mobilize support for exclusionary policies.”); see also id. at 454 (highlighting the “inability of the local and European authorities to effectively manage the refugee flows and provide medical support and sanitary services” at islands where asylum seekers were arriving). I thank Katerina Linos for raising this point. ↑

46. Cf. Noll, supra note 22, at 236 (“[B]urden-sharing continues to be a desideratum at best, a deceptive rhetorical veil at worst.”). ↑

47. See Birgitte Romme Larsen, Becoming Part of Welfare Scandinavia: Integration Through the Spatial Dispersal of Newly Arrived Refugees in Denmark, 37 J. Ethnic & Migration Stud. 333, 334–35 (2011) (describing the Danish mandatory spatial dispersal policy requiring refugees to live in a particular community for three years). ↑

48. See Bartl, supra note 12, at 2630–38 (describing the German scheme and detailing its history). ↑

49. See Francesca Campomori & Maurizio Ambrosini, Multilevel Governance in Trouble: The Implementation of Asylum Seekers’ Reception in Italy as a Battleground, 8 Compar. Migration Stud. 1, 4 (describing the Accordo Conferenza Unificata and the subsequent legislative decree responding to arrival of 170,000 asylum seekers in 2014 and “huge imbalances” in their geographic distribution within Italy); see also Giuseppe Campesi, Between Containment, Confinement and Dispersal: The Evolution of the Italian Reception System Before and After the ‘Refugee Crisis,’ 23 J. Mod. Italian Stud. 490, 498 (2018) (explaining how “logic of ‘dispersal’ drives the functioning of both the S.P.R.A.R. and of the network of Extraordinary Reception Centers” and how that logic includes idea of responsibility sharing within borders). ↑

50. See Linda Bakker, Sin Yi Cheung & Jenny Phillimore, The Asylum-Integration Paradox: Comparing Asylum Support Systems and Refugee Integration in the Netherlands and the UK, 54 Int’l Migration 118, 121 (2016). ↑

51. See Yitchak Haberfeld, Debora Pricila Birgier, Christer Lundh & Erik Elldér, Selectivity and Internal Migration: A Study of Refugees’ Dispersal Policy in Sweden, 4 Frontiers Socio. 1, 2 (2019) (explaining that dispersal policies in Sweden have been justified in part to share “financial and social burdens of immigration that fall upon refugees’ preferred destinations”). ↑

52. See Boswell, supra note 12, at 320–21 (describing the U.K. dispersal system). ↑

53. For a collection of detailed case studies, see Robinson et al., supra note 12, at 1–2 (presenting case studies of policies adopted in the Netherlands, Sweden, and the U.K. to disperse the settlement of asylum seekers, based in part upon fairness arguments concerning the local costs of providing services for them). ↑

54. See Liza Schuster, A Sledgehammer to Crack a Nut: Deportation, Detention and Dispersal in Europe, 39 Soc. Pol’y & Admin. 606, 615–17 (2005) (comparing and contrasting German and U.K. dispersal policies). ↑

55. Id. at 616–17. ↑

56. See id. at 617 (discussing “negative effects that dispersal can have on asylum-seekers”). ↑

57. Boswell, supra note 12, at 319. ↑

58. Robinson et al., supra note 12, at 164. Some of the potential harms to refugees may be mitigated if refugees are free to move within a country’s borders. For further discussion, see infra Part IV.C. ↑

59. The more general literature on immigration federalism has also addressed issues of responsibility and burden sharing. See, e.g., Peter H. Schuck, Taking Immigration Federalism Seriously, 2007 U. Chi. Legal F. 57, 90 (arguing that “some states do in fact shoulder a disproportionate share of the burdens imposed by undocumented immigrants”). ↑

60. Then-Indiana Governor (and Vice President-elect) Mike Pence’s proclamation exemplified this rhetoric: “[I]n the wake of the horrific attacks in Paris” in November 2015, for which the Islamic State of Iraq and the Levant claimed responsibility, “I am directing all state agencies to suspend the resettlement of additional Syrian refugees in the state of Indiana . . . .” See Stella Burch Elias, The Perils and Possibilities of Refugee Federalism, 66 Am. U. L. Rev. 353, 399 (2016) (quoting then-Governor Pence); Mariano Castillo, Paris Suicide Bomber Identified; ISIS Claims Responsibility for 129 Dead, CNN (Nov. 16, 2015), https://www.cnn.com/2015/11/14/world/paris-attacks/ [https://perma.cc/K7U9-7EEC]. For discussion of how “the figure of the migrant has come to be seen as a potential terrorist in the West,” see Thomas Nail, A Tale of Two Crises: Migration and Terrorism After the Paris Attacks, 16 Stud. Ethnicity & Nat’lism 158, 160 (2016). ↑

61. See Elias, supra note 59, at 353, 397–99. ↑

62. 8 U.S.C. § 1522(a)(2)(A). ↑

63. 8 U.S.C. § 1522(a)(2)(C)(iii). ↑

64. See Xi, supra note 7, at 1208–11 (quoting H.R. Rep. No. 97-541, at 11 (1982)) (discussing 1982 and 1986 amendments to Refugee Act). ↑

65. See Elias, supra note 59, at 374. ↑

66. See Xi, supra note 7, at 1221–22. Others have argued the existing system of consultation provides states and localities with meaningful opportunities to contribute to policy development. See Elias, supra note 59, at 413. ↑

67. See Exec. Order No. 13,888, § 1, 84 Fed. Reg. 52,355 (Sept. 26, 2019). ↑

68. See HIAS, Inc. v. Trump, 415 F. Supp. 3d 669, 686 (D. Md. 2020), aff’d, 985 F.3d 309, (4th Cir. 2021). ↑

69. Resettlement in the United States, U.N. High Comm’r for Refugees (Mar. 2020), https://www.unhcr.org/en-us/resettlement-in-the-united-states.html [https://perma.cc/9D2D-62HA] (“The United States has a long history of welcoming refugees and remains one of the largest refugee resettlement countries in the world.”). ↑

70. QuickFacts: Texas, U.S. Census Bureau, https://www.census.gov/quickfacts/TX [https://perma.cc/A8HY-HCUH]. ↑

71. Jens Manuel Krogstad, Key Facts About Refugees to the U.S., Pew Rsch. Ctr. (Oct. 7, 2019), https://www.pewresearch.org/fact-tank/2019/10/07/key-facts-about-refugees-to-the-u-s/ [https://perma.cc/D5EY-RNK8]. ↑

72. See David Owen, In Loco Civitatis: On the Normative Basis of the Institution of Refugeehood and Responsibilities for Refugees, in Migration in Political Theory: The Ethics of Movement and Membership 269, 274–75 (Sarah Fine & Lea Ypi eds., 2016) (calling refugee protection a “‘legitimacy repair’ mechanism”). ↑

73. Song, supra note 8, at 116. ↑

74. Id. at 116–17. ↑

75. Id. at 116 (arguing for “a system in which each state bears the primary responsibility for protecting the human rights of its members, but if it fails to do so, the responsibility falls on other states to step in”). ↑

76. Id. ↑

77. Bauböck, supra note 8, at 143 (“As this general responsibility is shared by all states, it seems natural to assume that states should also share the burdens of refugee protection and integration.”). ↑

78. See Elias, supra note 59, at 355 (quoting Governor Brown). ↑

79. See Will Kymlicka, Solidarity in Diverse Societies: Beyond Neoliberal Multiculturalism and Welfare Chauvinism, 3 Compar. Migration Stud. 1, 4 (sketching, but not adopting, a view that “[n]ationhood has helped to secure . . . an ethic of membership” that results in “bounded solidarity” and “egalitarian” commitment to “justice amongst members”); id. at 5 (“[A]s we all know, this link between nationhood and liberal-democracy creates endemic risks for all those who are not seen as belonging to the nation, including indigenous peoples, substate national groups and immigrants.”). ↑

80. Song, supra note 8, at 115–16. ↑

81. See Miller, supra note 8, at 83. ↑

82. See Stephanie Collins, Collectives’ Duties and Collectivization Duties, 91 Australasian J. Phil. 231, 234–35 (2013) (arguing that persons who have formed a collective may share a duty to rescue even if strangers do not). ↑

83. The argument is not that binding international law requires responsibility sharing within borders. Of course, local authorities’ actions still may violate a state’s duties under international law. A local authority might, for example, expel a refugee to a place of danger in violation of the state’s non-refoulement duty. But a local authority’s duty to comply with international law does not entail a responsibility to share in the costs of other local authorities’ compliance. Rather, responsibility sharing on the international plane remains a non-binding goal, and responsibility sharing within borders does not follow from a state’s binding obligations under international refugee law. ↑

84. See David Miller, The Nature and Limits of the Duty of Rescue, 17 J. Moral Phil. 320, 334 (2020) (“[T]he distribution of individual duties must be fair: the costs of carrying out the rescue need not be the same for all, since some may have capacities that others lack . . . .”). ↑

85. Bauböck, supra note 8, at 147. ↑

86. Song, supra note 8, at 123. ↑

87. See supra notes 35–38 and accompanying text. ↑

88. Hiroshi Motomura, The New Migration Law: Migrants, Refugees, and Citizens in an Anxious Age, 105 Cornell L. Rev. 457, 533 (2020). ↑

89. See, e.g., Boswell, supra note 12, at 321 (discussing U.K. policy following the 1999 Immigration and Asylum Act, which was based, in part, on an “assumption that dispersal could relieve social tensions caused by the concentration of asylum seekers in particular areas”). ↑

90. Song, supra note 8, at 123 (arguing that “nativist attitudes are not a legitimate reason for refusing to take in refugees,” at least “with respect to democratic states”). ↑

91. Id. ↑

92. Id. ↑

93. See id. at 115; see also Bauböck, supra note 8, at 144 (“[P]articular states can also have responsibilities for refugees because of their involvement in the refugee generating situation, for example if they supported a repressive regime or a warring party in an armed conflict.”). ↑

94. See Song, supra note 8, at 115 (arguing that the United States owes a special duty to Afghan and Iraqi interpreters who assisted its military during its wars in Afghanistan and Iraq). Aziz Huq has argued this principle also applies to the United States with respect to recent Syrian refugees. Aziz Z. Huq, America’s Refugee Debt, Bos. Rev. (Nov. 25, 2015), http://bostonreview.net/world/aziz-huq-syria-refugee-debt [https://perma.cc/X2YB-WVSS]. ↑

95. See Huq, supra note 93. ↑

96. See Seth Davis & Christopher A. Whytock, State Remedies for Human Rights, 98 B.U. L. Rev. 397, 398 (2018) (arguing that U.S. state tort law may provide redress for some wrongs that violate human rights). ↑

97. Compare E. Tendayi Achiume, The Fatal Flaw in International Law for Migration, 56 Colum. J. Transnat’l L. 257, 258 (2018) (arguing that “the conception of state sovereignty operational” within the “extant global governance framework” for migration is a possibly “terminal” flaw), and Chantal Thomas, What Does the Emerging International Law of Migration Mean for Sovereignty?, 14 Melb. J. Int’l L. 392, 442–49 (2013) (sketching a “new organicism” that would “shift away from atomism” and “turn towards interconnectedness” as the starting premise for international ethics of migration), with Schuck, supra note 9, at 247 (“[F]or the foreseeable future, genuine human rights protections—particularly the protection of refugees—can only be enforced and implemented by sovereign states or by other entities such as supranational agencies and nongovernmental organizations (NGOs) working with their assistance or sufferance.”). ↑

98. Refugee Compact, supra note 2, ¶ 4. ↑

99. Id. ↑

100. Exec. Order No. 13,888, supra note 66, § 1. A federal court enjoined this order. See HIAS, Inc. v. Trump, 985 F.3d 309, 316 (4th Cir. 2021). Then, President Biden revoked it. See Exec. Order No. 14,013, § 2, 86 Fed. Reg. 8,839, 8,840 (Feb. 4, 2021). ↑

101. See Song, supra note 8, at 58–66 (addressing these questions). ↑

102. Bauböck, supra note 8, at 146. ↑

103. Xi, supra note 7, at 1201 (proposing a system “to persuade localities to accept a group of refugees,” under which “the federal government must encourage them to do so through federal grant money”). ↑

104. Id. (“[G]iving local governments a more meaningful role in the refugee placement process—a right to refuse refugees—will better match refugee groups with communities capable and willing to welcome them.”). ↑

105. See supra note 9 (listing sources in that debate). ↑

106. Refugee Compact, supra note 2, ¶ 4. ↑

107. Id. ¶ 34. ↑

108. Id. ↑

109. Gibney, supra note 8, at 460. ↑

110. Id. ↑

111. Id. ↑

112. Id. (“Certainly, the refugee is probably best positioned to judge the country where they could best rebuild their social world or integrate successfully, particularly if provided with appropriate information on destination countries.”). ↑

113. Id. (“[T]he very act of choice may be a way of empowering people who, in Arendt’s words, have been deprived a place in the world where their opinions are significant and their actions effective.” (citing Hannah Arendt, The Origins of Totalitarianism 296 (1986))). ↑

114. See infra Part IV.C (discussing refugee movements within borders). ↑

115. See Gibney, supra note 8, at 460. ↑

116. Robinson et al., supra note 12, at 166. ↑

117. Song, supra note 8, at 123 (“[T]he better argument with respect to democratic states is that such nativist attitudes are not a legitimate reason for refusing to take in refugees because they violate the democratic value of equal respect.”). ↑

118. See supra notes 41–44 and accompanying text. ↑

119. See Laura P. Lunn, Displaced and Disillusioned: “Free-Case” Refugees and the Government’s Obligation to Facilitate Effective Resettlement, 14 J. Gender Race & Just. 833, 846 (2011) (“Refugees are matched with resettlement agencies depending on their specific needs in relation to the services that each agency is able to provide.”). ↑

120. Privatization of immigration enforcement in other contexts has raised related concerns. Jennifer M. Chacón, Privatized Immigration Enforcement, 52 Harv. C.R.-C.L. L. Rev. 1, 3 (2017) (“In the detention context, and in many other immigration enforcement contexts, privatization can facilitate the development of structural mechanisms that undermines governmental accountability for individual rights deprivations.”). ↑

121. See Steve Walsh, Coalitions of Veterans Forming to Help Their Afghan Allies Get to US, KPBS (Aug. 27, 2021), https://www.kpbs.org/news/2021/08/27/coalitions-veterans-are-forming-help-their-afghan [https://perma.cc/H26X-7Z5P]. ↑

122. See Priscilla Alvarez, White House Makes Massive Change to Resettlement Program to Help Afghan Refugees, CNN (Oct. 23, 2021), https://www.cnn.com/2021/10/23/politics/afghan-refugees-private-resettlement-veterans/index.html [https://perma.cc/9QLB-3LCH] (“[T]o increase options to evacuees, the Biden administration is launching a program that would allow veterans with ties to Afghans, as well as others, the opportunity to bring them to their cities and serve as a support network as they get their lives started in the US . . . .”). ↑

123. Meryll Dean & Miki Nagashima, Sharing the Burden: The Role of Government and NGOs in Protecting and Providing for Asylum Seekers and Refugees in Japan, 20 J. Refugee Stud. 481, 484 (2007). ↑

124. Id. ↑

125. Moulid Hujale, If You Felt Cooped Up in Lockdown, Think of Refugees Confined Indefinitely in Camps, Guardian (Aug. 24, 2020), https://www.theguardian.com/global-development/2020/aug/24/if-you-felt-cooped-up-in-lockdown-think-of-refugees-confined-indefinitely-in-camps [https://perma.cc/4SGE-Z7H3]. ↑

126. Id. ↑

127. See, e.g., Kathrine Vitus, Zones of Indistinction: Family Life in Danish Asylum Centres, 12 Distinktion: Scandinavian J. Soc. Theory 95, 101 (2011) (discussing Danish policy circa 2011, under which “asylum-seekers [had to] live in the asylum centres, prohibited not only from free movement out of the country but also in some cases from free movement within Danish borders”); Boswell, supra note 12, at 319 (describing then-extant policy under which asylum seekers in Germany had “no say in choosing their place of residence” and were “confined to the Kreis [district] in which they [we]re staying,” meaning that they “require[d] a special permit if they want[ed] to travel outside of this area”); see also supra notes 53–54 and accompanying text (discussing German and U.K. schemes for responsibility sharing within borders). ↑

128. A 2020 study from the Netherlands found that about half of refugees remained in the municipality that they were assigned to ten years after the assignment, if migrants who departed the Netherlands altogether during the period of the study are excluded. See Marloes de Hoon, Maarten Vink & Hans Schmeets, On the Move Again? Residential Trajectories of Refugees After Obtaining Asylum in the Netherlands, 27 Population Space Place 1, 11–12 (2021). ↑

129. Even with secondary migration, we should still be worried about the impact of dispersal schemes upon refugee flourishing. Some evidence for this concern comes from the impacts of Sweden’s “Whole-Sweden” policy, which randomly assigned asylum seekers and allowed them to move. One study found that highly educated and skilled refugees “were more likely to migrate to labor markets with a wide structure of opportunities relative to less skilled refugees.” Haberfeld et al., supra note 50, at 1. A study by Olof Åslund and Dan-Olof Rooth focused upon the initial dispersal of refugees across Sweden, and concluded that initial placement in less prosperous communities was correlated with lower job prospects and income levels for refugees. Olof Åslund & Dan-Olof Rooth, Do When and Where Matter? Initial Labour Market Conditions and Immigrant Earnings, 117 Econ. J. 422, 424 (2007) (“Local conditions in the year of immigration affect initial individual outcomes, which in turn influence future earnings and employment. Those who start in poor regions tend to also experience high local unemployment in the future.”). Another study by Per-Anders Edin, Peter Fredriksson, and Åslund found that “the increased long-run dispersion of refugee immigrants that was achieved with the policy came at the expense of individual outcomes in the labor market,” but that refugee movement mitigated the negative impact. Per-Anders Edin, Peter Fredriksson & Olof Åslund, Settlement Policies and the Economic Success of Immigrants, 17 J. Population Econ. 133, 149–50 (2004). ↑