Introduction
Over 43 million refugees – individuals forcibly displaced by conflict and persecution who cross international borders – exist in the international system.Footnote 1 Geographic proximity is the main criterion shaping which countries take responsibility for refugees, where the closer a country is located to a refugee’s country of origin, the more responsibility it will have for the refugees.Footnote 2 This responsibility regularly extends beyond emergency care, where these states host refugees for the long term without adequate support from less proximate countries, whether in the form of financial assistance or offers to relocate refugees to their territories. Due to the disproportionate role that the ‘accidents of geography’ play in determining who has responsibility, there is a wide consensus among scholars and practitioners that the existing distribution is not only highly unequal but unfair. Footnote 3 While scholars have proposed possible criteria for creating a fairer distribution, they tend to ignore the overarching question of how a set of ethical principles ought to relate to one another. This raises what I refer to as the weighting question: what should each principle’s importance be within a composite framework? How should the principles be combined to assign responsibility to states?
Without clear guidance on this question, we cannot adjudicate between claims of responsibility made by political actors such as state leaders, who strategically invoke different sources of responsibility to minimize their own level of responsibility while elevating those of others. For example, U.S. President Donald Trump argued in 2017 that countries which share closer ties to the refugees ought to help, claiming that the ‘safe, responsible, and humanitarian approach’ would be to ‘host refugees as close to their home countries as possible.’Footnote 4 Russia’s representatives argued in 2018 that responsibility for refugees should be given to the countries that have ‘irresponsibl[y] interfere[red] into the internal affairs of sovereign States of Middle East and North Africa,’ implicating the interventionist role of the United States and its allies.Footnote 5 And at the height of the Syrian refugee crisis, Jordanian Minister of State for Media Affairs and Communications Mohammad Momani stated that ‘Jordanians are proud that their country has surpassed capable and great countries’ by hosting in excess of 1.4 million, more than 20% of Jordan’s population.Footnote 6 Underlying each of these claims are implicit theories of what gives rise to responsibility, where the responsibility either lies with the state that shares closer ties to the refugees, caused the exodus of people, or has more material capacity. Assuming each of these arguments is normatively plausible, it remains unclear how the responsibility ought to be distributed among each state.
I identify and overcome two necessary hurdles for resolving the weighting problem, as it relates to the specific context of assigning long-term hosting obligations to states. The first step is determining the set of principles to include in the composite framework. Scholars have proposed many possible criteria for allocating responsibility but some condensing of the list is likely necessary as a practical matter if the primary goal is combining them to generate fair shares. I review the possible allocation principles used in the literature – with a focus on sources of special responsibility – and argue for the use of liability, community, and capacity. Each of the principles is thoroughly detailed, where liability assigns responsibility to the external states that caused the refugee flows, community to those that share ties with the refugee groups, and capacity to those most able to help.
Once the principles have been selected, the primary challenge is theorizing solutions to the weighting problem: how should the principles relate to one another and which should carry more or less weight than the others? The difficulty is that there are likely various reasonable ways to rank order the principles rather than only one ‘master’ solution; what is a proper ordering will depend on the framework’s underlying goals. By specifying what these goals are, we can theorize the range of possible solutions. The first dimension by which the ordering might vary is whether the framework is intended to operate in an ideal world – one that assumes favorable conditions like the identifiability of each principle and compliance by states – or a non-ideal one that more closely resembles the real world where these conditions are absent. I argue that capacity should be prioritized over the other principles for any non-ideal formulation because it is the most identifiable of the three principles and thus the most amenable under real-world constraints. Applying the framework to an ideal world instead opens up additional possibilities for how to order the principles, where there is a second dimension by which the ordering could vary: whether the primary goal is to make the distribution more fair for states or for refugees. Liability should be given primacy if mitigating interstate inequity is the main concern but this can harm refugee well-being if refugees are compelled to live in unfit, liable states. As such, community and capacity should be granted precedence if refugee well-being is the main concern, and I discuss the possibility of using an alternative principle – refugee choice – to further this objective as well.
Throughout, the paper focuses on hosting as the main obligation states have toward refugees rather than alternative means of assistance like financing, where states pay others to host. It is important, as a first attempt at answering the weighting problem, to fix the content of responsibility to one behavior, as the relevance and fittingness of each principle will likely depend on the mode of contribution one is talking about. Limiting the scope of the study to hosting ensures a clear discussion of the principles and that they are being evaluated with respect to the same target. While this paper answers the fundamental question of how to think about creating the (hosting) quotas, financing could still be relevant. Many responsibility-sharing proposals often envision the role of financing after the initial quotas have been assigned, where states can ‘trade’ their quotas with one another by paying others to discharge portions of their quota.Footnote 7 It remains an open question, and a promising direction for future research, how solutions to the weighting problem would speak to these post-quota considerations.
Overall, clearing the two hurdles moves us toward a determinate weighting scheme, which is necessary for creating a fairer distribution. By discussing the crucial choices that need to be made, from selecting the principles to guidance on how to weigh them, I lay the groundwork for such schemes to be possible. In particular, the second hurdle of theorizing different versions of a weighting proposal will hopefully push scholars toward thinking more about the question of how the principles relate to one another. My contribution threads a middle ground between arguments that insist on one correct scheme and those that pessimistically view the weighting question as inherently indeterminate. Rather, there could be multiple reasonable schemes, each based on the optimization of distinct goals. It may be impossible for scholars and policymakers to agree on a single framework of responsibility allocation. But if we can at least agree that there are a multitude of valid goals and that these differences in priorities map onto distinct weighting approaches, then we can make progress toward resolving this difficult problem.
The paper proceeds as follows: First, I establish the significance of addressing the weighting problem and argue that existing approaches have not sufficiently explored the relationship between a set of principles. Next, I justify the inclusion of liability, community, and capacity in the framework and describe each principle, noting how it would identify responsible states. I theorize different possibilities for rank ordering these principles, which depend on the framework’s underlying goals like whether it intends to apply to a non-ideal or ideal world and whether it is designed to serve the collective interests of states or refugees. Finally, I conclude with some implications.
The weighting problem
Among multiple possible principles, the weighting problem refers to the challenge of how to specify the importance of each principle to determine each state’s fair share. It occurs after the refugee status determination stage and asks how to allocate a specified number of recognized refugees fleeing from particular countries among a specified population of states. In essence, the question is how to fairly allocate these refugees among the population of states globally or with respect to a region. I restrict the scope of the weighting problem to focus on the hosting obligations of the international community, particularly long-term hosting under the assumption that short-term repatriation is not feasible.Footnote 8 Thus, the content of the obligation can be applied to the long-term integration of recognized refugees either by the country of first asylum or from other states’ territories (resettlement). While other modes of contribution like financing are certainly important, the benefit of focusing on one behavior is that it helpfully narrows the discussion when comparing and evaluating each principle. Each principle is evaluated in relation to hosting only, which greatly streamlines the exploration of the weighting problem.
The weighting question starts with the premise that to create a fairer distribution of refugees, multiple distributive principles appear plausible. Imagine in this stylized scenario that there are three principles – 1, 2, and 3 – where Principle 1 points to Actor A as responsible, Principle 2 to B, and Principle 3 to C.Footnote 9 Assume that they generally point to distinct actors as responsibleFootnote 10 and that each of these principles fully apply to the case under investigation, so that none can be easily dismissed due to the lack of empirical relevance. Note that this stylized example is not so different from the example presented in the introduction. Answering the weighting question means we must determine the weights for Principles 1, 2, and 3, which will yield precise ‘fair shares’ of refugees A, B, and C ought to host. The weights could be determined simultaneously – assign a weight to each principle all at once, where the weights would sum up to one – or sequentially – create a priority ordering of the principles where the principle deemed most important is applied first, the second applied next, and the least important applied last. Rules would need to be set as to when we move from one principle to the next, where each subsequent principle assigns a residual amount of responsibility from the previous stage.Footnote 11
While this may seem a highly difficult and intractable task, the argument is that it is necessary to answer if we want to avoid unsatisfactory alternatives. The first of these options is to simply select the most favored principle and assign all of the responsibility using it, effectively giving the other principles zero weight. Perhaps due to its relative ease of measurement, for example, capacity has been widely used as the sole allocation rule to create fair share quotas.Footnote 12 Note that this position is often made due to theoretical simplicity; scholars are not arguing that other principles do not matter, just that they have strong reasons to focus on one in particular.Footnote 13 The obvious problem with a single-principle approach, however, is that it overlooks other relevant principles, each of which would result in a highly distinct distribution. A more compelling option is to consider each of the principles and theorize about their relationship. While one could opt for a practically convenient approach here too – asserting simultaneous equal weights – this would preclude a richer and perhaps more intuitive solution, that is, even if each of the principles has plausible normative value, some should be considered weightier than the others.
To work toward a satisfactory answer to the weighting problem, I illuminate and overcome two hurdles. The first is specifying the set of principles to be weighed against one another, where scholars have offered a wide array of possible allocation criteria. How should the principles be selected, and how can they be used to identify responsible agents? Once the principles have been selected and explicated, the second and primary challenge is theorizing possible solutions to the weighting problem. I argue that there is not just one correct ranking of the principles but rather multiple reasonable possibilities based on the optimization of different goals. These different goals, and the distinct ordering schemes that they produce, are explored in detail.
The plausible principles
The first hurdle is figuring out the set of allocation principles that can be used to create a fairer distribution. A natural place to look for these principles is to examine sources of special responsibility, which identify specific states for action and their magnitude of responsibility.Footnote 14 I provide an overview of the broad menu of principles to choose from, which originate from the fields of political theory and international climate governance. I justify why liability, community, and capacity are chosen as the three core principles for the composite theory and conceptualize each for the refugee domain. Due to their prominence in debates on how to assign responsibility in general and for refugees specifically, I argue for the inclusion of these three principles in any weighting scheme. That said, the focus on the three principles is meant to be a starting point rather than the final say on which principles should be included, where the advantage of focusing on three and not more is that it makes the weighting problem more tractable to solve.
Political theory provides a set of principles to allocate remedial responsibility to states, which can be applied to questions of how to assign responsibility for refugees specifically. David Miller identifies several possible sources of responsibility: causal, outcome, moral, community, capacity, and benefit.Footnote 15 We can broadly categorize the first three criteria as components of a ‘backward-looking’ liability approach which looks at how the conduct of perpetrators led to the harm; causal responsibility refers simply to A’s causal role in bringing about the refugees, outcome responsibility adds onto this relationship of bare causation by assuming that A had agency, and moral responsibility considers the extent to which A’s conduct deserves blame.Footnote 16 Next, community examines the special bonds that a prospective host state may share with the refugee group, necessitating that it help more than non-affiliated states. Capacity assigns responsibility to states in accordance with their different levels of material ability to host refugees while benefit envisions that states which have benefited from the global and systemic processes that produce refugees have a particular obligation to host them.
The other main source of the principles is drawn from international climate governance, particularly the concept of common but differentiated responsibilities (CBDR). Recognizing that international law overseeing refugees does not supply clear responsibility-sharing criteria, scholars have sought to import the concept of CBDR to the refugee domain.Footnote 17 CBDR can be operationalized by several ‘equity’ principles, where scholars identify five that are relevant and applicable to the refugee domain: polluter pays, capacity, egalitarian, cost sharing, and merit.Footnote 18 The first two principles overlap with the political theory criteria; capacity, as noted, would assign responsibility to states using economic and material indicators while polluter pays resembles liability in making the perpetrators that caused either high emissions or refugees the main duty-bearers. The egalitarian and cost sharing principles each strive to equalize the distribution of responsibility across states; the first envisions that states accept an equal number of refugees per capita while the second equalizes the costs of admitting refugees where states that can accept refugees at a lesser cost should accept more than those that accept refugees at a higher cost. Finally, the merit principle examines the prior contributions of states, where those that have contributed more to refugee protection in the past should have their assigned responsibility reduced.
While the literature has proposed a broad menu of principles, they have focused less on how the principles ought to relate to one another within a unified framework. I propose focusing on the following three principles for the purposes of weighting – liability, community, and capacity – where liability holds agentic states responsible for their refugee-producing actions, community to the states whose populations share bonds with the refugee groups fleeing, and capacity to the states that are the most materially able to help. This list represents both an aggregation of several principles mentioned earlier – liability is a catch-all category for causal, outcome, and moral responsibility – as well as the discarding of other candidates. As I will discuss, each of these principles is prominent in the broader political theory literature and also commonly invoked as sources of responsibility for refugees specifically. Moreover, they are each able to identify and differentiate actors’ levels of responsibility, which is a necessary criterion of this entire endeavor: use some combination of the principles to identify each state’s fair share.
While scholars may buy the inclusion of these three principles, an obvious question is why restrict the focus to only three and not more? The decision to focus on these three sources of responsibility and not others is driven primarily by concerns about tractability rather than a claim that they are more worthy than the other candidates. Rather, some condensing of the master list is likely necessary, given that the primary goal is aggregation into a workable scheme rather than providing an exhaustive list of principles. It simply becomes more difficult to discuss the relationship between the principles and map the different ways of combining them as the number of principles under investigation increases. To be clear, the focus on these three principles does not preclude the addition of other principles in a weighting scheme. Future work can certainly add more principles to this foundation.
A detailed description of each of the three principles is provided below, where I discuss its broad ethical rationale and how to conceptualize it within the refugee domain. Each principle is important in its own right – the challenging question this raises is how they ought to be weighed against one another to assign responsibility to states.
Capacity
Capacity assigns responsibility to actors based on their ability to address a given problem. This principle enjoins actors with more resources to do more, because they can help at less cost than an actor with fewer resources.Footnote 19 It coincides nicely with a utilitarian approach, which strives to minimize suffering and maximize welfare. Given this objective, a utilitarian view would look to actors who can help ‘without sacrificing anything of comparable moral importance … [or] anything morally significant.’Footnote 20 By looking to the actors that can help without compromising their well-being, their assistance would minimize the suffering of others while not taking away from their own welfare, thus raising the average level of well-being within the system. Capacity also aligns well with cosmopolitan thinking, which views actors as having obligations to others regardless of the recipient’s identity or whether one shares a pre-existing relationship with it. A cosmopolitan perspective would argue that actors like states ought to help those who are most vulnerable to global and transnational threats, regardless of where they live; capacity comes into the picture because capable actors are best positioned to reduce the vulnerability of others.Footnote 21
In the refugee domain, scholars commonly argue that states with greater capacity should be tasked with more responsibility than those with less.Footnote 22 To adequately care for refugees, states must protect their human rights as outlined by the UN Refugee Convention and Protocol and meet their basic needs like food, water, education, and access to medical care. Two indicators appear particularly relevant for capturing a state’s ability to fulfill these rights and needs: its wealth and size.Footnote 23 A country’s wealth, or level of development, can be thought of as its economic capacity to provide for refugees without adversely affecting its ability to provide for citizens’ basic needs. Wealthier countries – those with greater GDP per capita or median income – can provide basic needs and government services to refugees at less burden to themselves than less wealthy countries.Footnote 24 Meanwhile, countries with large economies, populations, or territories are better able to absorb a large influx of refugees than smaller ones.Footnote 25 The underlying commonality across these measures of size is that they reduce the degree of contact and potential competition between refugees and the local population.Footnote 26 Different combinations of these indicators have been used to assign fair share quotas to states, making capacity the dominant principle used to allocate responsibility for refugees.Footnote 27
Liability
Liability assigns responsibility to actors for their role in harming others. Miller’s three criteria of causal, outcome, and moral responsibility can be thought of as sub-components of this principle, where the first two are necessary conditions for liability. First, a core component of liability is establishing a causal connection between two events, X and Y, where Y is some adverse outcome by which we seek to assign responsibility to one or more actors. In the case of a direct and necessary cause, to say X caused Y is to pick out X from many other antecedent conditions for its abnormality; a cause is what ‘make[s] the difference’ between an event happening or not.Footnote 28 In addition, X is typically distinguished by the fact that it was a voluntary human action.Footnote 29 When a chain of preceding events produces Y, we typically look to where the voluntary action took place to locate liability. This requirement that actors must act of their own volition distinguishes outcome responsibility from bare causal responsibility. When an actor causes an outcome but has no meaningful agency, we cannot hold them liable. Finally, moral responsibility – the extent to which an agent’s conduct warrants blameworthiness – is a sufficient but not necessary condition for liability.Footnote 30 This means that actors can be held responsible for the consequences of their actions independent of fault.Footnote 31 The concern with insisting that actors must act culpably to be liable is that it may set too high a bar to identify the responsible agents, which endangers the objective of adequately compensating the victims.
Refugee movements can be caused by external states, creating an obligation for those states to help.Footnote 32 External states acting with agency can produce refugees in various ways: relatively directly, exacerbating the effect of others’ actions, and more diffusely. I provide a brief overview of these distinct causal mechanisms, highlighting that it becomes more difficult to isolate states’ contributions to refugees as the causal relationships become more diffuse. First, external states can cause refugees in a direct sense. Direct liability refers to the presence of a relatively clear and unmediated link between an external state’s policies and the production of refugees, thereby warranting primary liability for the perpetrators. Military intervention is a particularly salient example where it can destroy a country’s infrastructure and economy, prompting people to flee land that is no longer inhabitable.Footnote 33 In addition to purely direct liability – State A’s policy leads straightforwardly to the production of refugees – this category also includes instances when the initiating state’s policy clearly enabled other actors to produce refugees.Footnote 34 For example, a state’s use of military force may destabilize or even overthrow governments without adequately preparing for the aftermath, creating a power vacuum conducive to more violence and more refugees. While domestic actors may be the groups directly responsible for causing refugees, the external state’s intervention created the opportunity for these groups to commit harm, and this outcome was clearly foreseeable. In other cases, the external state may not directly involve itself in the conflict but instead may provide significant military and political support to oppressive regimes and other groups that go on to create refugees. These policies, while mediated through other actors, still warrant significant liability, particularly the more the groups are dependent on this aid.
External states can also be liable when their policies exacerbate an existing refugee flow caused by other actors.Footnote 35 Liability can still be traced to the external state, but its level of liability is now shared with the primary actor that caused the refugees. In contrast to the first category, the state’s policies did not directly cause or initiate a chain of events that produced refugees. Rather, they make the refugee flows primarily caused by another actor worse – a state’s policy (e.g., military response) to massive human rights violations initiated by a domestic government can magnify the number of refugees compared to a world in which it did not respond in that manner.Footnote 36 The implication is that the primary liability would go to the domestic government, though this depends on how much the external state’s policy intensified the problem.
Finally, states can be liable in more diffuse ways, where their policies create the conditions for refugees to eventually emerge. This category can be difficult to identify and requires careful linking of an action’s contribution to the production of particular refugees at a specific point in time. One relatively agreed-upon and identifiable example is the production of ‘climate refugees,’ where states’ collective contributions to global warming create the conditions for these refugees to emerge.Footnote 37 The states that have emitted the most should do more to help these refugees than those who have emitted less. Even if there are important issues that need to be resolved – like what year to start counting the emissions and how much excusable ignorance should be a mitigating factor – we can roughly determine the degree of causal contribution for each state by examining its proportion of global emissions. Other diffuse causes that may be more difficult to identify include historical injustices like colonial rule and states’ contributions to unjust global structures.Footnote 38 Due to the significant passage of time and the actions of mediating agents like domestic governments, it can be difficult to trace colonial rule to the emergence of a specific group of refugees in the postcolonial era, even as the colonizer caused general harm to the state. Another common example of a diffuse cause is the claim that states, particularly wealthier states, collectively contribute to an unjust global economic order that creates permissive conditions for refugees.Footnote 39 The problem is that the Global North as a whole is implicated but it is difficult to differentiate responsibility among individual states. In sum, it can be considerably more challenging to isolate the liability of external states here than in other mechanisms. Liability will generally be more promising as an allocation rule the more direct the causal relationship.
Community
Community assigns responsibility to actors based on the special relationship they share with others. The idea that communities form morally relevant ties where members owe more to each other than non-members is well-established by both commonsense morality and communitarian ethics.Footnote 40 Individuals may be part of many communities, ranging from personal relationships to broader groups in which they may never meet most of the other members. Whether they are small enough to feature direct interaction or much larger, communities are a group of people joined by shared history, identity, interests, and goals.Footnote 41 These ties generate obligations among members because they derive considerable value from these relationships and recognize that the needs and wants of the other participants can be a source of reasons for action.Footnote 42 Members of communities may also depend on one another more than they depend on non-members, which gives rise to obligations between them.Footnote 43
In the study of politics, we are often interested in communities at the group level, where the actor in need is a group of individuals requiring assistance. The group in need could share a tie with other potential duty-bearers, necessitating that those actors ought to help more than those who are non-associated. Community promises to be a highly relevant allocation principle within the refugee sphere due to the importance of inter-group dynamics in creating them. Refugees are often fleeing due to persecution based on group membership. Outside states which have groups in common with these refugees would be particularly well-suited to resolve the refugee group’s vulnerability by hosting them. More generally, actors that share a tie with the refugee group may be more equipped to host, as it may be easier to integrate the refugees if they share a similar background. In these arguments, there is some overlap with the capacity principle, though it is also distinct as it points to commonalities and shared ties rather than mere material ability as the relevant factors that make a state responsible.
Community envisions that states have an obligation to assist a refugee group if they have a shared bond, an obligation they do not share with others.Footnote 44 It can be conceptualized as different types of ties, particularly those based on common history and identity, that obligate a host state to accept refugees from a particular state. The relevant relationship is whether the host state and its broader society share affinities with various refugee groups fleeing a country of origin, where the more affinities they share, the greater the obligation to host refugees from the country of origin. It is important to look at the subnational rather than bilateral, intergovernmental level because it is presumed that the country of origin’s government does not represent the refugees; in fact, it often has fundamentally opposed interests to the refugees and may be responsible for their displacement. Relevant linkages include colonial, ethnic, linguistic, and religious ties. Some ties, like the former colonial relationship, affect the entire societies of both the host state and the country of origin and can thus be thought of as a tie that both countries broadly share.Footnote 45 Meanwhile, the other connections apply firmly at the subnational level, where we can ascertain whether the host state shares groups in common with the various country of origin-refugee groups on the dimensions of language, religion, or ethnicity.Footnote 46 Looking at the subnational level and disaggregating the populations of each state means that influential minority groups within host states would count as possible ties, not just majority groups. Thus, this principle will at times point to less proximate, capable states with established minority populations and not just to more proximate states which may generally share more of these ties.
Weighting the principles
Now that we have the list of principles, how should they be arranged and put together to assign refugees to states? A complete answer to the weighting question will determine precise weights for each principle that indicate each principle’s importance. Here, we can begin to answer this question by imposing some ordering (ordinal ranking) of a set of principles, ranking them from most to least important. This will serve as a pathway for a determinate weighting scheme that can uncover each state’s fair share.
The challenge is that there may be multiple solutions to the weighting problem, that is, various reasonable ways of ranking the principles, due to the optimization of different goals. Does one value a more realistic ordering that could potentially be implemented in the real world or does one want to create a moral ordering designed to apply to an ideal world? Does one care more for refugee or state interests when ranking the principles? Each of these goals (dimensions) could result in different weighting schemes. The first is whether the principles are intended to operate under non-ideal or ideal theory assumptions, where the first resembles more of the real world and the second adds favorable conditions like the neutral identification of each principle and the enforcement of the allocations. I argue that capacity is central in any non-ideal formulation due to its greater objectivity than the other principles. The second dimension by which the relationship between the principles could differ is whether state or refugee interests are the main concern. An approach that seeks to promote justice between states will likely prioritize liability while one that seeks to promote justice for refugees will prioritize community and capacity instead as well as consider the principle of refugee choice: using refugees’ preferences for host destinations to assign them to states. These different orderings show the promise of engaging more with the weighting question.Footnote 47
Capacity’s centrality for a non-ideal world
Ideal theory assumes a world with favorable background conditions where actors will generally comply whereas non-ideal theory invalidates one or both of these assumptions.Footnote 48 More specifically, the difference between ideal and non-ideal theory in the context of debating the ordering of the principles is that ideal theory assumes a hypothetical set of institutions that can (1) make accurate and neutral attributions of responsibility for each principle and (2) enforce the allocations. Meanwhile, non-ideal theory negates these assumptions, so there is no institution that can perform these functions, mirroring the real world. As such, the overriding question given non-ideal assumptions is which ordering of the principles would be most likely to be realized (feasible) in the absence of these strong institutions. The absence of enforcement affects all three principles equally, as none are enforceable, but some principles, like capacity, are already considered more identifiable than others. Therefore, I argue that capacity is the most realistic principle and thus should be given the greatest weight in this world.
The measurability of capacity makes it possible to make accurate and neutral attributions of responsibility using it. It is readily identifiable using material indicators like GDP, GDP per capita, population, and land area. While scholars and practitioners may quibble over the exact indicators to use and the amount of importance to place on each, they generally appear to recognize the importance of capacity as a distributive principle and for evaluating state contributions. The UNHCR, for example, has historically used measures like population and GDP to annually evaluate state contributions.Footnote 49 Perhaps the greatest evidence of capacity’s workability is the EU’s use of capacity-driven quotas to allocate 120,000 refugees among its member states in 2015.Footnote 50 Population and GDP accounted for 80% of the formula to calculate the quotas, and the other 20% looked at mitigating circumstances that capped a state’s capacity to take in refugees – the average number of asylum applications over the previous four years and the unemployment rate.Footnote 51 These real-world uses of capacity show that it is an eminently plausible method to assign responsibility under non-ideal theory conditions.
In contrast, liability and community lack the simplicity of capacity and are thus generally unfit as allocation principles in a non-ideal world. For either principle, it is difficult to imagine that any objective understanding of it could be reached by states without the promise of a hypothetical institution that can perform this function. With liability, this goes beyond the fact that it can be difficult to identify, where there could be reasonable empirical disagreements about the relative weight of various causes that produced the refugees in question, particularly as the causes become more diffuse. The deeper concern is that these attributions will be driven by self-interest and politicized in the absence of a neutral agency that can make factual attributions of liability. Allies of the perpetrating state will downplay claims of its liability while adversaries will amplify them. There is a similar concern with community, where objective measures appear elusive. Even if actors generally act as if they have greater obligations to their community, the meaning of community can be quite subjective, and actors may have different understandings of what criteria constitute community for them.
If instead it is assumed that there is an institution capable of making neutral attributions of responsibility for each of the principles and enforcing those determinations – which move us toward an ideal world – then liability and community will become more important. In conceptualizing each principle in detail, I see my role as the ‘institution’ that can objectively identify responsibility for each principle in an ideal world. These assumptions raise a different set of questions and open up more possibilities than when we are in a non-ideal world. Scholars can create a moral theory where the ethical strength of each principle is weighed against one another, and they are each assumed to be binding. I argue that the proper allocation of responsibility in this world will further depend on another dimension: whether the collective interests of states or refugees are prioritized.
Refugee versus state interests in an ideal world
An ideal world abstracts away issues of identification and enforcement of the principles, thereby creating more possibilities for how to order the principles. A crucial dimension by which the ordering could differ within this world is whether the framework is designed to fulfill the collective interests of states or refugees. The collective interests of states can be identified as creating a fairer distribution of refugees among them than the one that currently exists. Meanwhile, the collective interests of refugees can be defined as the distribution that maximizes their aggregate well-being, captured by the extent to which the distribution reflects (1) preferences for where they want to be hosted and (2) their likelihood of thriving in the host state. Notably, a scheme designed to promote state interests may come at the expense of refugee interests (and vice versa). An ordering that seeks fairness for states could leave refugees in undesirable and unfit states; conversely, one that is maximally responsive to refugee interests, like accommodating their preferences for host states, could compromise interstate fairness.Footnote 52 I illuminate each of these possibilities below.Footnote 53 If the main goal is making the distribution more fair for states, I propose an ordering where liability is given the most weight, community second, and capacity least. If the goal instead is to create a distribution that maximizes refugee well-being, I propose elevating the importance of community and capacity relative to liability and consider an alternative principle, refugee choice, to assign refugees to states.
Liability’s centrality for interstate fairness
Liability should be prioritized over the other ethical principles if addressing interstate inequity is the main goal, because it would most directly mitigate the moral hazard problem endemic in the current distribution. The presence of a moral hazard is perhaps the main reason why the distribution is considered unfair among states. It occurs when actors are shielded from the costs of their actions and are thus not sufficiently deterred from engaging in those costly actions. The current refugee distribution is a classic manifestation of this phenomenon, where states can cause refugees, knowing that they will not be held responsible. Due to the legal principle of non-refoulement, proximate states bear the brunt of the responsibility whether they caused the refugees or not, while states located farther away are better able to prevent refugees from arriving even if they were the ones that caused their flight.Footnote 54 Privileging liability would break the link between proximity and responsibility and place the bulk of the responsibility on the states that caused the refugees. In addition to a more just distribution of costs, privileging liability may also prevent states from engaging in refugee-producing actions in the first place, potentially reducing the supply of refugees in the system. Affording any of the other principles priority may have the unintended effect of exacerbating the moral hazard problem, as states may be emboldened to cause refugees knowing that high-community and high-capacity states would bear more of the costs.
Community and capacity serve important yet subsidiary roles after liability, as they would each further resolve the problem of interstate inequity. Community would place refugees in states that share more ties than non-affiliated, random states, while capacity would place them in states with greater ability to secure their needs. As a practical matter, community may be prioritized over capacity due to capacity’s more general orientation. Community, like liability, is directed at a specific other, which means it only activates for certain states while it does not for states it does not share a tie with nor has it harmed. Meanwhile, capacity generally applies across all refugee cases and is not recipient-dependent; those with more capacity should help more refugees in general rather than a particular group of refugees. As such, capacity would identify the same set of states as the most responsible across all refugee cases, whereas who is liable or shares a tie of community varies by case. Thus, prioritizing capacity over any of the other principles can potentially be over-demanding for the duty-bearers, who are tasked with the lion’s share of the responsibility across all cases, regardless of states’ values on the other two principles. Rather, capacity should become more relevant the more the first two principles are exhausted as options to assign responsibility. By placing it last in the sequence, it serves as a ‘backstop’ that can assign responsibility even if the previous two principles fail to apply.
A refugee-centered scheme
While the earlier scheme seeks to reform the system for states – namely, by holding states accountable via liability – a scheme that seeks to improve refugee well-being instead will likely emphasize different principles. Here, the primary goal is how to maximize the aggregate well-being of a set of refugees: create a distribution that places the refugees in states that they want to be in and are most likely to thrive. This will necessarily lead to ranking community and capacity over liability, as they each promise to fulfill these objectives more than liability. The concern with liability is that it may often place refugees in undesirable states; in other words, the quest to reduce moral hazards may come at the expense of refugee well-being. Finally, I discuss the possibility of using a principle that falls outside the scope of special responsibilities – refugee choice – and how it can work independently as well as in conjunction with the three core principles.
A refugee-centered scheme will prioritize community and capacity over liability, because they are each essential for ensuring that refugees are placed in states where they can flourish. The use of capacity places refugees in states that are materially well-off and able to provide the basic foundation for them to thrive, like securing their physical protection and basic needs. More than securing basic needs, capacity is central for improving their life chances and prospects more generally. In addition, refugees need a place to rebuild their social world that has been stripped from them. States that share a tie of community – those that share ethnic kin with the refugees or allow refugees to practice their religious and cultural traditions without discrimination – offer the best prospects for integration.Footnote 55 Due to the importance of each principle for refugee welfare, I am agnostic about their relative weight and am simply arguing that they should each be granted greater weight than liability.Footnote 56
Liability is given the least weight, because assigning refugees to liable states can often be detrimental to their well-being. While liability may be ideal for preventing moral hazards and making the costs more equitable among states, it can lead to undesirable consequences for refugees. Refugees may face hostility and ill-treatment if they are placed in states that have already caused great harm to their group.Footnote 57 To ask them to relocate to these states would be potentially cruel. One possible exception to this rule is if liable states are allowed to contribute by other means besides hosting, like financing: paying for the protection of refugees elsewhere. If this is granted, the role of liability may become more prominent, because unfit states that caused harm would finance instead, and refugees would not be forced to relocate to those places.Footnote 58 If focused only on hosting, however, community and capacity each promise a better environment for refugees than liability, thereby they are given precedence.
Incorporating refugee choice?
The above discussion argued that among the three sources of special responsibility, community and capacity are the most related to refugee well-being and should thus be emphasized. The argument is that refugees may benefit much more in countries with high capacity and community. While these quotas are designed with the interests of refugees in mind, perhaps a more straightforward way of ensuring their welfare would be to solicit their preferences for host destinations and use that information to allocate them to states.Footnote 59 The concern with an exclusive, top-down focus on special responsibilities is that it risks relocating at least some refugees to undesirable destinations, however well-intentioned it aims to be. Refugees may have preferences for host destinations that are not entirely captured by the ethical principles. Therefore, the only way to ensure refugees are placed in states that they want, which is a central component of their well-being, is to ask them directly. I discuss how refugee choice could be used as a standalone principle to assign refugees to states or used in tandem with the three core principles.
The first option is to simply bypass the quotas created using the sources of special responsibility and have the distribution directly and solely driven by refugee preference. Refugees can be granted varying degrees of choice in these proposals. Their choice of destinations could range from completely free – soliciting their list of preferred host countries – or could be restricted in several ways: for example, asking them to only list undesirable states, to list attributes of host countries but not particular host states that are desirable, or to choose from a limited set of countries.Footnote 60 These preferences would be collected and aggregated by some centralized agency to create an optimal distribution. The downside of this approach, however, is that it eschews any concern for interstate fairness. Having refugee choice be the sole determinant of the distribution can create new, unjustified inequalities between states if refugees consistently have preferences for host states that do not correlate with any of the ethical principles.Footnote 61 It could seriously undermine the objective of interstate fairness.
A more nuanced and balanced approach envisions that refugee choice be used to adjust the quotas, allowing it to be used alongside the three core principles. First, the quotas would be created, placing a premium on community and capacity. To ensure that all refugees are matched to appropriate states, the principle of refugee choice would be used to ‘fill’ the quotas as a second step. In a system where refugee choice is not considered, quotas would be filled with refugees randomly; it would not matter which specific refugees end up where.Footnote 62 Incorporating refugee choice means that it could shape how the quotas are filled; which refugees go where is now relevant. If refugees’ stated preferences generally point to the high-quota states, then the quotas would not change that much. If refugees systematically prefer low-quota over high-quota states, then the low-quota states would take in far more refugees than they should, and they should be compensated by the undesirable high-quota states. Keeping the quotas ensures some semblance of fairness between states because the desirable low-quota states would receive some compensation for their extra efforts. This approach of using refugee choice alongside the quotas promises to better reconcile state and refugee interests.
Conclusion
This paper has worked toward resolving the weighting problem: the puzzle of how to specify the importance of each principle to assign responsibility to states. Scholars have proposed various sources of special responsibility but it is less clear which of these principles should be selected for a composite framework and how they ought to relate to one another. I explored this problem in the context of assigning refugee-hosting obligations to states, arguing first that liability, community, and capacity should serve as core principles in the framework. Each represents a morally relevant reason as to why a specific state ought to host refugees and can differentiate levels of responsibility between actors. Next, I specified different solutions for weighting the principles. Given the diversity of normative aims that a framework could have, it is unlikely that there is one ‘correct’ or proper ranking of the principles. I argued that there are at least two dimensions by which the ordering of the principles could differ. First, if the framework is intended to apply to a non-ideal world that features real-world constraints, like the lack of robust institutions that can identify responsible states, then capacity should be given primacy over the other principles due to its greater identifiability. Second, if the primary goal is to make the distribution of refugees more fair among states, then liability should be granted the most weight; if refugee well-being is the main concern instead, then community and capacity should be given greater emphasis, and alternative principles like refugee choice should be considered as well.
This paper has cleared the necessary hurdles to develop a determinate weighting scheme, which is needed to specify each state’s fair share. By focusing on the question of how the principles relate to one another, it can advance the literature on conceptualizing state responsibility for global problems more generally. The weighting problem is not unique to refugees and arises in a variety of domains. While IR scholars have proposed multiple sources of special responsibility, there remains ambiguity over how to distribute responsibility for other shared, global problems like climate change, global health, and widespread underdevelopment.Footnote 63 Similar to the challenge with assigning responsibility for refugees, there are multiple plausible principles of responsibility, and state leaders seek to deflect their responsibility by emphasizing principles that would implicate other states. My approach of selecting the relevant principles and theorizing different possibilities for weighting them can be amended and applied to these problems, which will allow scholars to uncover each state’s fair share.
Acknowledgments
I would like to thank Audie Klotz, Minseon Ku, Ben McKean, Glyn Morgan, Greg Smith, Alex Thompson, Alex Wendt, participants at the 2024 Political Theory In/And/As Political Science (PTPS) Junior Scholars Workshop, as well as the two anonymous reviewers and the editors at International Theory, for their helpful comments.
Competing interests
The author declares none.
Open access
