- Original Article
- Open access
- Published:
Between meeting quotas and following the duty-bound heart: navigating the formidable dilemma of refugee protection in the EU
Comparative Migration Studies volume 12, Article number: 26 (2024)
Abstract
The provision of refuge to those fleeing persecution and danger is a complex matter that cannot be reduced to simply meeting admission quotas. While numerical targets for refugee admissions may seem like a straightforward way to ensure states are doing their fair share, an overemphasis on numbers risks obscuring the deeper moral issues at play. A truly robust approach to refugee protection requires that states act from the right kinds of reasons - a genuine commitment to the duty to provide safe haven to all refugees, regardless of their country of origin or the political expediency of admitting them. The worry is that if a state’s willingness to accept refugees fluctuates based on domestic political preferences and perceptions of national interest, rather than an unwavering recognition of the moral claims of the forcibly displaced, then the protection provided to refugees becomes contingent and precarious. A disposition to selectively protect only certain refugees undermines the very concept of refuge. However, in our non-ideal world, a commitment to the principle of robust refugee protection can run up against the hard reality of state non-compliance. If the only way to get some states to fulfil their duties is to allow them leeway in choosing which refugees to admit, we face a serious moral dilemma. Relaxing the demands of robustness may enable larger raw numbers of people to access refuge in the near term. However, it also risks eroding the norm of impartial consideration and creating a two-tier system of refugee protection. Ultimately, I argue, the dilemma between refugee numbers and robustness has no fully satisfying solution - it is an unavoidable product of our current circumstances of injustice and non-compliance. The best we can do is seek arrangements that ease this tension over time. This requires taking a nuanced, context-sensitive approach attuned to real-world constraints and trade-offs while still keeping sight of the moral lodestar of genuinely unconditional refuge for all who need it.
Big numbers and broken duties
The European Union’s treatment of refugees is a litmus test of its commitment to common projects serving international law and justice beyond the boundaries of states. As Frans Timmermans, Vice-President of the European Commission, has put it: “The European Union stands for principles of human dignity, solidarity, and respect for human rights. It is in our hands to turn these principles into realities for people in need of protection.” However, despite these bold proclamations, the question of how to effectively protect and integrate refugees within the EU remains a contentious issue.
This article aims to contribute to thinking about what such effective provision of refuge amounts to, especially in the context of the EU, but not exclusively in that context. It posits that a more profound understanding of this commitment can be attained by conceiving refuge provision as a rich good, necessitating actions grounded in pertinent reasons related to refugee status and the protection of refugees. I contend that adopting such a perspective fortifies the protection of all types of refugees. Moreover, it illuminates why higher refugee admission numbers and tradeable quotas aimed at increasing admissions may not inherently entail more robust refugee protection or even the genuine provision of refuge. The article further illustrates the usefulness of this analysis by delving into the intricate dilemmas arising from conceiving refuge provision as a robust good and explores time- and case-sensitive approaches to resolving them while also examining potential objections.
The EU, with its shared institutions and capacity to coordinate member states, is uniquely positioned to shape a just and effective framework for refugee protection that embodies the principles of robust refuge provision. The European Union’s distinct role in discussions on refugee protection stems from its common institutions, which aim to coordinate and monitor the actions of member states across various policy domains. By virtue of this institutional structure, the EU possesses the potential to facilitate the just and effective discharge of refugee protection duties by its member states (Bauböck, 2018). Within the realm of political theory, scholarly attention predominantly revolves around the obligations that member states bear towards refugees (Carens, 2013; Miller, 2016). Correspondingly, some consideration is given to appropriate refugee policies within the EU (Bauböck, 2017, 2018; Owen, 2019). Protection for refugees is offered via two policies. The temporary protection directive is meant to serve as a rapid response tool for mass displacement situations and refugee quotas that serve as a mechanism for sharing the burden of refugee resettlement during less immediate circumstances. In the European Union’s approach to distinguishing among various types of refugees through the implementation of two distinct policies, it is not inherently the case that these policies stand in opposition to one another. By conceptualising refugee protection as a rich good, we are afforded a clearer understanding of the rationale for their integration.Footnote 1
It is within this context that the demand for a just and effective EU framework for refugee protection arises, often taking the form of advocating for refugee quotas to be fulfilled by each member state (Hathaway, 1997; Anker 1998). Consequently, ongoing debates in political theory strive to delineate the normative requirements that this quota system should meet, while public policy debates focus on the practical implications of its implementation (Bauböck, 2018; EU Commission 2016). It is against this backdrop that four key desiderata have emerged within the relevant literature, deemed crucial for a satisfactory system (Gibney, 2015; Thielemann, 2017; Milazzo, 2023).
First, the system should allocate quotas based on a state’s capacity to provide the requisite resources for refugee protection, primarily considering GDP and population size. Second, it should account for refugees’ preferences regarding their destination, taking into consideration locations where their integration is more likely to succeed. Third, the quotas should be tradeable and enforceable, ensuring effective refugee protection. The underlying idea is that states disinclined to protect a significant number of specific refugees should have the opportunity to trade quotas with other member states and compensate those willing to accept payments in return for exceeding their allocated quotas. At the same time, states failing to meet their quotas should be held accountable for refusing to compensate other states. Finally, for these schemes to be fair, they must not reflect unjustifiable state preferences concerning particular groups of refugees (Bauböck, 2018).Footnote 2
The allure of meeting quotas by aiming at large numbers can be morally misleading, as it may not necessarily reflect the complex moral implications of this choice in our rather non-ideal circumstances. While the EU’s refugee protection framework places significant emphasis on quotas and their associated normative requirements, this approach may not effectively address the interests of all refugees, leading to a complex dilemma arising from conflicting desiderata. This dilemma becomes apparent when a member state fulfils its quota by selectively accepting refugees from a specific country driven by entrenched domestic political preferences. Consequently, a difficult choice arises either allowing such selective criteria, thereby compromising the commitment to providing robust protection for all refugees in need, or prioritising robustness while sacrificing the opportunity to offer protection to a larger number of refugees. This paper aims to address the moral concerns associated with this dilemma by providing a robust account of refugee protection while recognising the inherent tensions that arise from any normative approach attempting to satisfy multiple, often conflicting, requirements. These tensions, as we shall see, are particularly evident in our current situation.
A real-life example of non-robustness: disparities in refugee protection
A comparison of the cases of Ukrainian and Syrian refugees admitted to the EU helps illustrate these tensions. The first thing to note is that the current number of Ukrainian refugees globally is comparable to the total number of Syrian refugees, about 7 million (UNHCR as of May 2022). However, the total number of Ukrainian refugees in the EU is five times higher than that of Syrians. The second thing to note is that there are very significant cross-country differences that remain unexplained even if we consider the fact that some member-states are smaller than others in terms of GDP size and population. In Germany, for example, there are about 700,000 refugees from Ukraine, a number already higher than that of Syrian refugees. On the other hand, Poland has granted protection to 1,142,964 Ukrainian refugees as of May 2022 but to only 239 refugees from Syria.
The number of Ukrainian refugees in Poland corresponds to roughly 3% of its population. Although the Polish state had initially struggled to cope, and despite receiving relatively limited help from the EU, it neither has in any way become unstable nor unable to provide the basic public services it usually does to all those now residing within its territory. In other words, the normal functioning of institutions serving domestic achievements of political and social justice has not been significantly jeopardised by Poland admitting and granting protection to the equivalent of 3% of its population. We might think, therefore, that countries like Poland can use this number as a benchmark for a feasible quota of refugee admission and protection. We may argue that these numbers of admitted refugees approximate best practices of strict compliance with the duty to protect and integrate as many refugees as possible, at least in the here and now and up to now, for countries like Poland. When that capacity is not reached by a member-state sufficiently similar to Poland, we could claim with some degree of certainty that the member-state in question is unwilling, as opposed to incapable, of complying with duties to refugees. One positive aspect of this situation is that it establishes a new real-world precedent, making it more challenging for other nations with comparable capabilities to argue that they lack the capacity to provide refuge to a proportional number of displaced persons as a percentage of their population in the future. The issue is less about an inability to marshal the necessary resources and more about an unwillingness to do so.
There is also a darker side to this picture, however; a side this paper aims to highlight and caution against. Before admitting 5,3 million refugees from Ukraine, the EU had admitted only about half of that number of refugees from the rest of the world, about 2,7 million. The EU’s relative capacity to admit refugees, based on its GDP size, which is about 1/6 of the global GDP, is roughly 15 million, which amounts to about 3% of its population. The EU, as a whole, is far from this number, but this percentage is roughly equivalent to the number of Ukrainian refugees admitted to Poland.
These high numbers of Ukrainian refugees in countries like Poland raise a key question that motivates this paper. Are countries like Poland doing more than their fair share, even if we were to consider their past non-compliance? Imagine Poland were to say to the rest of the EU member-states: “Yes, in the past, we have not done our fair share in fulfilling the duty to protect and integrate refugees in the EU, and hence, as much as we could and should have done, but we have now provided protection to so many refugees that we have clearly exceed our fair share several times. Therefore, we need not host more refugees in the foreseeable future (and we may even be entitled to compensation for doing more than our fair share). We are willing to host large numbers of Ukrainian refugees, and they are willing to be hosted in Poland. Therefore, we do more than enough to fulfil our duty to protect and integrate refugees.” Further, acting upon the principle of non-refoulement for refugees from countries with which we share a border should take priority over accepting quotas from other countries. By just looking at the current numbers of protected refugees, it is difficult to challenge this claim, regardless of whether we look at them as a proportion of the population or weigh them for GDP size.Footnote 3
Are then countries like Poland morally exonerated and even praiseworthy for accepting so many refugees from Ukraine? Not if we adopt the robust conception of refugee protection outlined in this paper. As I shall argue, if we embrace the robust conception of refugee protection that this paper advocates, the moral commendation of countries such as Poland for their acceptance of numerous refugees from Ukraine becomes problematic. I will contend that it is morally disconcerting for countries like Poland to invoke the principle of non-refoulement not on the basis of a consistent and genuine commitment to refugee protection, but rather as a matter of convenience, aligned with their national interests and preferences.
There is thus a relevant wrong beyond the number of refugees a country accepts now or rejected earlier, namely its lack of a robust and, hence, reliable disposition to protect refugees as much as it can reasonably. As we shall see below, however, in practice, a commitment to such a robust conception of refugee protection does not avoid hard dilemmas in the here and now, even if it helps to better make sense of our current predicament. Footnote 4
The duty to provide refuge as a rich good
At this juncture, it becomes imperative to elucidate a conceptual framework of the duty to provide refuge. Conceiving the provision of refuge as a rich good is key here. Robust goods have three key characteristics. First, they are goods that we enjoy in the hands of others and that are central to living a good life. Second, the enjoyment of a robust good requires the steady, reliable provision of a corresponding ‘thin’ good. Third, for something to qualify as a robust good, it must be provided for the right reasons and with the right disposition. For instance, the rich good of love is central to living a good life; its reliable provision requires the thin good of care, and it is only provided when we act out of love. A carer who provides the thin good of care out of contractual obligation in exchange for a salary does not love us, even though she cares for us. Moreover, she can choose not to enter such a contract when it is not beneficial or convenient for her, unlike those who love us and should care for us even when it is not beneficial or convenient for them.
For refugee protection proper to obtain, the host state must also provide those in need of protection with that protection robustly. That is, even if the state or the person fleeing, or the circumstances somewhat changed. To use an analogy that might help to illustrate the point, to enjoy love robustly, others should act with care towards us out of a disposition of love even when our circumstances change (e.g. care for us even when it is not beneficial or convenient for them because of the costs involved in caring for us, for example when we are ill or old). The same goes for robust refugee protection. If I do not care for refugee protection, out of an appropriate disposition towards their condition and the interests jeopardised by the conditions they found themselves in, then I am more likely to abandon refugees to their fate as soon as it is not beneficial or convenient to me. So, robustness also requires an appropriate disposition.
Refugee protection is not offered robustly if a state provides protection to a person fleeing only when the circumstances are such that it is convenient or beneficial to the state in question to act in this way, for example, if it acts in this way only when it suits its national interests, or only because the state in question fears retaliation from other states if it fails to act, for example when it admits all types of refuges only because of threats of sanctions. In cases of non-robust provision of protection, a person fleeing has reason to doubt whether the state in question is committed to protecting persons like her and if the state in question constitutes a reliable source of protection. One country’s lack of commitment to robust protection is made all the worse by the tendency of such slack to produce compounding and downstream effects, allowing also other countries to claim a right to pick and choose their refugees, thereby reinforcing biases and stereotypes towards particular refugees and rendering the protection of refugees in general more conditional, contingent and precarious. When not coupled with a robust disposition to protect, big numbers can be deceptive and ephemeral.
This account of robustness also brings to the fore a dilemma currently at the core of refugee protection. Consider the following scenario: a given member-state is willing, due to prevailing domestic political preferences, to meet its quota of refugees, but insofar as it accepts refugees from a particular state only and not from other states. This dilemma highlights a difficult choice. If a state’s domestic political preferences prevent it from accepting refugees from certain countries, then we face a dilemma. On the one hand, allowing the state to select refugees according to these preferences in order to meet its quota would violate the moral duty to provide refuge without discriminating among those in need on grounds other than reasons relevant to refugee status. However, the alternative of disallowing such selection criteria in these circumstances could result in fewer total refugees being offered protection.
Therefore, increasing the number of refugees admitted via the introduction of a quota system could go hand in hand with undermining refugee protection in other ways. Therefore, and if the account proposed here is sound, we need to adopt what I dub a robust approach to refugee protection to see why this is the case. By adopting this robust approach to refugee protection, we can better understand the difficult dilemma at hand and the corresponding issues with quota systems. Furthermore, we can ease the tension by aiming to protect refugees as comprehensively as is reasonably possible rather than simply dismissing such considerations in the pursuit of maximising numbers in the here and now.
In what follows, this article gradually transitions from an ideal theoretical framework of robust refugee protection to non-ideal cases of partial and non-compliance. Section 2 of this article provides a detailed account of robust protection using stylised examples mirroring recent refugee influxes into the EU. Then, in Sect. 3, I examine the dilemma that arises in cases of non-compliance on the part of member-states with duties of refugee protection. The article examines two responses to resolving this tension, one based exclusively on condemning preferences for certain refugees and a second focusing solely on the number of refugees admitted. It finds both wanting in non-ideal conditions characterised by widespread non-compliance with the duty to protect refugees robustly. Instead, I argue that, in the here and now, one may need to respond to cases of entrenched non-compliance with a time-sensitive approach that aims not to resolve the dilemma but rather to ease it. This approach takes into account considerations often overlooked in the literature, such as the timing of admissions and sanctions, and highlights the enduring relevance of competing policy priorities, such as stability as a modus vivendi versus justice considerations. Section 4 addresses various objections to the argument, and finally, Sect. 5 concludes the article.
Robust protection vs. refugee quotas
The core desideratum of refugee protection in political theory and public policy is the effective protection of the largest number of refugees (Bauböck, 2018). Allocating quotas to states, depending on their GDP and population size, is seen as increasing capacity by distributing refugees in a manner that renders their protection in large numbers not only possible and more effective but also just, as it proportions refugees according to the relative capacity of each state to protect and integrate them (Carens, 2013; Gibney, 2015, Tielemann et al. 2010).Footnote 5
While the use of quotas could be, in some cases, an indispensable tool for ensuring the coordination of refugee protection, its exclusive focus on the number of refugees protected may be misleading. Consider the following scenario. Imagine a state admits its allocated quota of 1 million refugees because these refugees are of a particular type. More precisely, these types of refugees are citizens of a country with particular ties to the state in question. In this example, let’s assume that the state is Poland, and the relevant type of refugees would be refugees from Ukraine. The particular ties in question are motivated by the fact that Russia is conducting an unjust war of aggression against Ukraine, which also happens to constitute a threat to the national security of Poland. In other words, Poland has a national interest in hosting refugees from Ukraine. Contrast now this case with another one where another state admits its quota of 1 m refugees, but that million includes all types of refugees irrespective of whether their fleeing is the result of actions of a state that constitutes a threat to its national security. That state could be, for example, Germany.
How do we make sense of the differences between Poland and Germany, given that both states provide protection to the same number of refugees and meet their quotas? Relying solely on allocated quotas offers little help in understanding such cases. From a quota-oriented perspective, the only relevant aspect is whether the number of protected refugees meets the stipulated quota sufficient for safeguarding the largest number of refugees. Since the quota, by hypothesis, amounts to Poland and Germany protecting 1 m refugees, then both states simply meet their quota, and there seems to be little more to say here.
I propose that in order to make sense of the differences between Poland and Germany, we must move beyond the number of protected refugees and examine the robustness of their commitment to protecting refugees. Specifically, what matters here is that Poland is committed to protecting refugees in accordance with its quota due to their specific type, that is, in virtue of the fact that Ukrainian refugees are the product of an unjust war on the part of a state that poses a threat to its national security. Hence, protection in accordance with the stipulated quota is conditional in the case of Poland. In contrast, there are no such conditions attached to protection in the case of Germany, for instance. Hence, we could say that Germany’s commitment to refugee protection is more robust than that of Poland even though both meet their quotas because of the lack of such conditionalities in the case of Germany.Footnote 6
From this analysis, we can discern the essential components of robust refugee protection. Refugees are afforded robust protection when their admission and the associated rights tied to their refugee status are not contingent on the willingness of any particular member state to honour those rights in a given case. Furthermore, robust protection requires an auxiliary mechanism of compliance that empowers other states to prevent any individual state from acting in ways that undermine the fulfilment of refugee rights across all relevant cases, at least in the presence of an institutional structure of soft power, such as that found in the EU, which makes this preventative action possible.
This perspective on robust refugee protection sheds light on a range of concerns expressed in the existing literature on refugee protection. Most notably, it brings to the fore questions concerning the role of preferences of both states and refugees and the weight these preferences should carry in determining which state should provide protection and facilitate integration. Scholars differ in their views on the relevance and significance of such preferences (Miller, 2016; Carens, 2013; Gibney, 2015). While some contest whether states should be able to limit the number of refugees they accept to settle in their territory on the grounds of a concern for preserving their cultural identity, most political theorists agree that arbitrarily selecting certain groups of refugees over others constitutes a violation of equal concern and amounts to discriminatory treatment (Bauböck, 2017).Footnote 7 With regards to refugee preferences, there is consensus that refugees have no right to choose a particular destination country but that their preferences may amount to a pro tanto consideration that should be given weight. Still, there is disagreement on how relevant refugee preferences should be in determining which country they are protected and integrated into.
Returning to our earlier example, we can argue that the robust conception of refugee protection aligns with these concerns by highlighting how discrimination undermines robust refugee protection.Footnote 8 But it goes beyond that concern. Given that the set of potential refugees is not a homogenous socioeconomic, cultural or demographic group and that the same applies to the preferences of states and refugees, we can contend that the most robust protection is the one that treats refugees as a category of persons that is due protection irrespective of these characteristics, as long as they are in need of refuge. States are obliged to treat these characteristics as opaque in their admission policies and bring about conditions that render them irrelevant to protection.Footnote 9 They must, for example, create the conditions necessary for persons coming from different cultural backgrounds to be able to integrate, even if that entails significant changes in the role that the host state assigns to a particular religion, customs or even language. Otherwise, their refusal to admit on the grounds of difficulties related to the lack of such structures supportive of integration is a problem of their own making.Footnote 10
It is important to note that discrimination is not the sole factor undermining robust protection. For example, in the case of Poland and Ukraine, there might be an alignment between the preferences of the host state and that of the refugees. Not only Poland might prefer Ukrainian refugees to, say, Syrians, but also Ukrainian refugees might prefer to receive protection in Poland than in some other safe country, such as, say, Turkey. However, what matters from the perspective of robust protection is not the alignment of preferences but the removal of impediments that render protection less robust for all refugees. Fixed preferences for particular refugees on the part of states and for particular states on the part of refugees may constitute such impediments when they are inelastic and facilitate the entrenchment of inelastic preferences both within and across states. This results in making the protection of certain types of refugees more costly for reasons unrelated to a state’s resource capacity for protection, as they make resource provision for refugees depend not on resource availability as such but on the unwillingness to allocate and use available resources for the protection of refugees. Hence, such preferences could result in the withholding of resources from the common pool of resources available to certain types of refugees, thereby reducing the overall supply of such resources and increasing the cost of their protection.
It is important, however, that such considerations are examined on a case-by-case basis. In the case of countries like Poland, preferences are inelastic because they are fixed preferences for particular types of refugees. They are not, for example, the upshot of a prior policy of equal consideration for all types that has incidentally resulted in the formation of refugee communities of a specific type, facilitating easier protection of certain refugees by the earlier integration of refugees of the same type. While countries such as Sweden may be justified in expressing such preferences, countries like Poland cannot because they have never opted to offer such robust protection in the past. The point is that not all positive preferences for certain types of refugees are unjustified, even if they often are. If such preferences are not an expression of animus or do not result in increased costs of protection for certain types of refugees, then they are permissible as they do not undermine robust protection. The burden of proof, however, should always fall on countries that invoke such preferences to ensure that reasons of cultural affinity or even proximity do not merely serve to indirectly discriminate against particular types of refugees and hence undermine robust protection.Footnote 11 Additionally, it might also be important to factor in other considerations, such as historical responsibility, although at a later stage, once quotas based on GDP and provisions for equal consideration are in place. Beginning with such considerations is likely to be contentious and impede agreement among member-states in any EU scheme that is predicated on the cooperation and monitoring of inter-state agreements.
A hard dilemma for advocates of policies of robust protection
In this section, I discuss the dilemma that arises when member-states’ preferences for certain refugees do not align with the type of refugees in need of protection. This dilemma can be classified as a hard dilemma, which refers to a situation where conflicting values or principles cannot be easily reconciled due to their equally fundamental and competing nature. Resolving hard dilemmas requires engaging in a process of moral deliberation to weigh the relative importance of different values in the specific context and circumstances at hand. In some cases, hard dilemmas may require us to choose the “lesser of two evils” or to make trade-offs between competing values, even if this involves some degree of moral compromise (Bauböck et al., 2022).
As we shall see below, there is a dilemma at the heart of the robust conception when one must confront cases where states are unwilling to protect refugees robustly. The dilemma becomes a hard one in such conditions of non-compliance caused by states’ fixed preferences for specific refugees and their dis-preference for others for reasons irrelevant to their status as refugees. In such cases, due to these fixed preferences, one cannot easily avoid a decision between increasing the numbers of those admitted at the expense of offering admission to all those in need of refuge and insisting on robustness at the expense of numbers. In non-ideal circumstances of non-compliance, we are compelled to assign different weights to these two considerations to decide which of the readily available policy choices is overall more normatively appealing (or less normatively appalling). The overall aim is to minimise concessions made to each of the two desiderata not just from a view of the present but also of the foreseeable future and, in that sense, to mitigate the trade-off both in the present and in the future.
To begin addressing this issue, it is helpful to identify cases that involve dilemmas but may not necessarily amount to hard dilemmas in all instances. For consider a hypothetical case similar to the one discussed in the previous section, where state A accepts to protect and integrate either one million of X type of refugees or only half a million consisting, for instance, of 250 K of X types and 250 K Y types, while its quota is one million. Assume that this commitment is strong because state A is firmly committed to this preference structure. Assume further that other member-states cannot provide protection and integration to an extra half a million refugees (that is, they are not merely unwilling to do so at a moderate cost to themselves). In this case, we have a feasible set consisting of two options. The first option is to increase the number of protected refugees in order to meet the quota by compromising robust protection for both X and Y types. The second option is to prioritise robust protection for both X and Y types at the expense of a larger number of refugees receiving protection and integration from State A and meeting the stipulated quota.
This scenario undeniably involves conflicting desiderata, but it should be noted that the conflict arises solely due to State A’s non-compliance. If State A faces no hard constraints or adverse circumstances significantly limiting its capacity to protect and integrate one million of both X and Y refugees, then its refusal is only due to not being willing to bring itself to offer protection and integration for Ys. It matters, therefore, what renders the relevant preference so fixed. However, if State A were to face hard constraints and unfavourable circumstances (for which it could not be held accountable)Footnote 12 that make admitting one million X and Y refugees detrimental to effectively protecting and integrating one million Xs and Ys while still maintaining protection for one million Xs, then its choice to prioritise Xs would be normatively permissible. This resurfaces the dilemma as a genuinely hard one. In other words, a dilemma is not a genuinely hard one if it is not hard all the way down, that is if it arises solely due to an unwillingness in favourable circumstances to act in a way that satisfies simultaneously the otherwise competing considerations.
Resolving genuinely hard dilemmas is undoubtedly challenging. Suppose a State A, burdened by a lack of resources and poverty, were to decide that it prefers not to exclude Ys arbitrarily and to provide them with protection and integration at the expense of offering protection and integration to a higher number of Xs. In the case of a state burdened by the hard constraint of poverty, one could argue that both options are permissible. But if the state in question faces no such hard constraints in terms of its material circumstances and its corresponding institutional capacity to protect and integrate, then any lack of relevant policies and customs is merely due to the fixed nature of these preferences for particular types of refugees in favourable circumstances. It is also worth noting here, however, that fixed preferences as such do not necessarily entail the hard dilemma highlighted above in favourable material circumstances. For the dilemma to resurface as a politically intractable one, and in that sense at least as a contingently hard one, it is required these preferences to be not only fixed but also so entrenched that there is no legitimate and effective way to increase compliance without having to make concessions either to robustness or greater numbers.
Even though these cases are very different in nature, they might both call for the adoption of a similar policy in response to the dilemma. In other words, the optimal policy in terms of easing the dilemma by approximating robust protection or minimising concessions to it might not actually be different in cases where states are unable to comply with the robust protection vs. cases where states are unwilling to comply with robust protection. For instance, taking a dynamic time-sensitive perspective might suggest that in both cases, the best way to increase compliance is to do nothing in the here and now because, in our circumstances, this is the best of the available policy options, all things considered.
To see this, consider a case involving assisting a burdened non-EU state to protect refugees robustly is very likely to backfire, resulting in that state admitting half of its quota with a case involving sanctioning a non-compliant EU member-state is likely to also backfire in the same way. It is worth noting that these policy paths may vary significantly over time depending on the nature of the state in question. Different states with different capacities will require different amounts of time to implement different reforms. For a partially compliant burdened state lacking robust protection capacity, the path to follow will likely involve reforms aimed at enhancing basic material and institutional capacities. On the other hand, for a non-compliant liberal democratic and relatively prosperous EU member-state, the required reforms will differ when specific circumstances are taken into account. These reforms may also vary with respect to the circumstances at hand and even be counter intuitive. For example, in certain cases, the optimal path for such non-compliant EU member-states might be one that dynamically enhances the stability of the liberal-democratic institutions required to provide protection or integration and hence help these states develop the more specific institutions necessary for protecting different types of refugees by creating a more multicultural social structure over time. Further, taking immediate action against a state, for instance, in the form of sanctions, in cases where such action is going to be ineffective and could backfire, makes little sense. Inaction might be preferable in some cases and instances. There are sometimes better days to fight one’s battles, and that is when different, more favourable conditions are likely to obtain in the foreseeable future. It is possible that when confronted with such cases, inaction is a superior policy option over time compared to any action.Footnote 13
Even though we might be justified to select the same policy towards burdened and non-compliant states, that is, in cases where that policy is the most effective means to best-serving robustness considerations over time, it does not follow that we are justified to treat them always as the same cases policy-wise. The normative distinction between burdened and non-compliant states corresponds to a different range of our repertoire of legitimate policy actions. To increase robust protection, we are justified to name and shame and even sanction non-compliant EU’s member-states, even if we opt not to do so in this or that instance, but we are never justified to treat in this way burdened non-EU states even if in the current predicament and occasionally we consider inaction towards them and non-compliant states optimal all things considered.
The ideas presented above should not be misconstrued as concessions to non-compliance. On the contrary, they aim to address another relevant case. In this scenario, State A has the option to pay another member-state, such as State B, to protect and integrate its share of Ys. Assuming State B is willing and capable of providing such protection and integration (largely due to the financial compensation), State A can meet its quota of one million by protecting and integrating half a million Xs and paying State B, or other states, to protect the remaining half a million Ys. One might think that in such a world, both the robustness of protection of Xs and Ys is secured, and the overall quota, aiming to protect and integrate the largest numbers of refugees, is met. But this might not necessarily be the case in general and even less so in our rather non-ideal European Union.
Consider the following: if State B is a poorer member-state compared to State A, it would likely accept a price higher than its domestic cost of protection and integration but lower than the cost in the richer state. This allocation of resources may be efficient as fewer resources are required to protect and integrate the same number of refugees. However, it is also exploitative in that the richer state takes advantage of its higher GDP to not only offload its duty to another state, but also to treat refuge provision as an opportunity to self-enrich by reducing the per capita cost of discharging its duty to protect refugees.Footnote 14 Consequently, State A has an incentive to offload its responsibility even when it is capable and willing to protect and integrate refugees. To address this issue, it may be preferable to establish overall quotas based on GDP and then fix prices paid for protection to costs based on the GDP per capita of better-off states who wish to offload such costs to worse-off states, rather than subjecting them to negotiation between asymmetrically positioned member-states. Additionally, member-states that better protect and integrate refugees should be rewarded according to objective and subjective metrics measuring, for example, the social mobility of refugees and their satisfaction with refugee protection and integration, respectively. Conversely, states that fail to utilise allocated resources as intended could face sanctions.
It is important to emphasise that these contribution schemes towards collective EU refugee protection should not be seen as mere substitutes for states’ duties to refugees, achieved through financial contributions to other states. States that are unwilling to fulfil their duties as reasonably as possible but are willing to compensate others for fulfilling those duties on their behalf should transfer the full equivalent cost to other member-states, determined by higher domestic costs. Additionally, they should pay a penalty that can be used to reward states that excel in providing protection and integration. As previously suggested, the timing of these penalties is crucial and should be carefully considered. Nonetheless, the overarching goal over time should be clear: ensuring robust protection and integration for the largest possible number of refugees by ensuring that all member-states are nearly fully compliant with their duties, and not creating via tradeable quotas for refugee protection and integration regions within the EU where such protection and integration comes at the cheap and disproportionately.Footnote 15
Adopting a robust conception of refugee protection allows us to recognise that the aforementioned problems are not merely flaws within the quota system stemming from persistent inter-state inequalities within the EU; they might also be inherent to the system itself. As discussed in Sect. 2, expressing preferences for specific types of refugees reduces the cost of protecting and integrating them within a tradeable quota system. Conversely, the undesirability of other refugee types increases the price that states are willing to accept to protect and integrate them. In an ideal world, these pricing issues could be resolved by appropriately fixing prices and redirecting monetary rewards. However, these measures do not address the larger issue of limited protection and integration capacity when state preferences are rigid, as the robust conception aimed to highlight. In such a world, all potential refugees, especially the “unwanted” ones, recognise that their protection and integration are only provisionally secured when a sufficient number of states are willing to accept relocation payments, and global demand for refugee protection remains moderate. As the number of refugees in need of protection significantly increases, relocation budgets become strained, and willing states reach their maximum capacity, it becomes evident that tradeable quota schemes primarily serve to mask the underlying problem of insufficient protection supply due to a lack of commitment to robust protection by certain well-off states rather than offering genuine solutions.
Objections
In this section, I address several objections to the robust account of refugee protection and integration presented in this paper. The first objection challenges the assumption that all member-states collectively fail to exhibit adequate concern for the protection of specific types of refugees. It argues that if each member-state adopts a non-robust conditional refugee policy, there might be an alignment between the preferences for certain refugees and the corresponding groups. For instance, if State A wants refugees of type X and State B wants refugees of type Y, both types can be protected and integrated by different member-states. If Poland protects Ukrainians and Germany protects Syrians, both types are protected and integrated, albeit by different member-states. This non-robust protection and integration might even increase the number of refugees protected, as Poland, for instance, might be willing to take more refugees of the type it prefers, as an absolute number, than to meet a lower composite quota consisting of different types of refugees. In fact, one could press this objection further, and argue that the more selective refugee protection and integration becomes, not only the more willing states could become to admit greater numbers, but also the number of willing states might be enhanced.
In reply to this first objection, let me give a twofold answer. First, it is important to note that reality is less ideal than the objection suggests. Member-states of the EU, as well as most affluent states, tend to favour Christian, white, skilled, and young refugees from countries that are the victims of their geopolitical adversaries. The current refugee crisis follows this pattern, with the admission numbers of Ukrainian refugees versus other types of refugees revealing a preference for specific groups. Thus, the diversity of preferences for different types of refugees among member-states simply isn’t (and won’t be) as significant as the objection assumes. This bias reflects a strong preference for certain refugees that undermines robustness. This is the reason why the current higher numbers of refugees in the EU are both misleading and ephemeral. Barriers to unwanted refugees, due to these preferences, remain in place and will most likely keep the numbers of such refugees low, whereas the numbers of other refugees are likely to rise further, especially if the war in Ukraine does not come to an end soon.
Second, even if the preferences of states were sufficiently diverse, there is clearly something objectionable about a world in which every state is allowed to arbitrarily favour some refugees over others for reasons unrelated to their refugee status. Here’s an illustration that helps to tease this out.Footnote 16 Consider a scenario involving two different soup kitchens that provide meals to persons facing life-threatening poverty. Suppose soup kitchen 1 was willing to provide meals only to members of a particular ethnic group and soup kitchen 2 only to members of another ethnic group. Compare this state of affairs with one where the same number of meals is made available to both ethnic groups by both soup kitchens. If all that matters is the number of meals served, then we should be indifferent between the two scenarios. But the fact we would prefer the second scenario shows that we also value the equal concern at the heart of robust protection and hence that we might even be willing to serve a smaller number of meals to both ethnic groups out of a concern for that value rather than a larger number of meals to only one ethnic group.Footnote 17 The robust conception of refugee protection and integration helps us make better sense of the values involved in these different scenarios than an approach that focuses exclusively on maximising quota numbers. It could also point at policies, as we shall see again below, that approximate both conditions by rendering quotas composite and dynamic and hence adaptable to circumstances on the ground.
This brings us to a second objection. This is the claim that the nature of the value of robustness is unclear. At times, it appears as if robustness is reducible to a concern about numbers after all—that is, to a concern about numbers over different possible scenarios and in the long run. At other times, the objection goes, robustness seems to be anchored to a very demanding understanding of moral agency. One might object here that by treating states as capable of acting from moral reasons in the same ways as individuals, the robust conception is too demanding. It might be better to harness states’ rational self-interests than to design schemes that could work only if all states act for moral reasons.
In reply, there are two clarifications that need to be made. First, it is true that robustness has two dimensions in this paper, but the two dimensions are linked. Let’s get back to the illustration with the soup kitchens to clarify this connection. A world where there are soup kitchens only for Xs or only Ys is less robustly feeding all possible hungry people than a world with soup kitchens for each. Protection and integration of refugees in a world where an agent A caters only to X type and an agent B only to Y type is conditional on the number of Xs or Ys not being greater than the number, and the corresponding capacity, of states willing to protect Xs or Ys. Second, if agent A only protects X or Y because it is convenient for them, that is, for reasons of reward or punishment, then protection of X or Y becomes conditional also on the presence of such rewards and punishments rather than out of a concern for the circumstances of those in need of protection and integration. There is nothing left to fall back to for such an agent when such rewards and punishments are not available, and hence, their commitment to protection and integration faulters as soon as the geopolitical and socioeconomic calculus alters. The upshot of this picture and its emphasis on robustness is that it allows us to see the current higher numbers of refugees in Europe not as an aftermath of a concern for refugees in general but as an instrumental concern for certain refugees that is conditional on their characteristics and on broader geopolitical interests. A focus on the current bigger numbers simply means that one is unable to grasp what a robust account can do.
Further, as it is explained below, reasons of protection are distinct from reasons for protection. In the absence of the right reasons of protection, protection becomes merely instrumental and a concern for refugees a charade for other concerns. Robustness matters for numbers but also independently because, without the requisite disposition that renders numbers robust, one does not truly provide refuge but only offers conditional protection and integration as long as it suits one’s interests and geopolitical ambitions. This is, for example, why countries like Turkey might not be seen as trusted partners of the EU in the provision of refuge: agreements with such countries depend primarily, if not exclusively, on offers and threats.Footnote 18 Hence, if I lack the appropriate will or disposition to protect, then my protection is, at least in one respect, not robust, and this affects the (maximum) numbers. And if I lack the appropriate disposition but I do deliver in terms of the rights and resources to the protection and integration of refugees as long as it is convenient or imposed on me by others, then I do not really provide refuge in a deeper sense because I do not act out of reasons of protection. I act out of reasons of greed or fear, but not protection. This helps to analytically distinguish between agents that provide genuine protection and those that don’t within the broader refugee protection system. Reasons matter both for how robustly we protect and also for whether we provide refuge as a rich good.
This point brings us to a third objection. This objection is that it is difficult to see how the robust conception of refugee protection could provide guidance in the current non-ideal circumstances where member-states have very fixed and rigid preferences with respect to favouring and disfavouring certain types of refugees precisely due to reasons that lie far away from a general concern for reasons relevant to refugee status as such. It is difficult to see how the robust approach could help us guide policy towards countries like Poland that have strongly resisted accepting refugees from any part of the world except Ukraine. Furthermore, it questions whether the time-sensitive approach proposed in the paper can change attitudes towards unwanted refugees, such as those from the Middle East, as it primarily focuses on increasing long-term capacity rather than willingness to accept refugees from diverse regions.
In response, three replies. First, the concern for reasons of protection at the heart of robustness does not ignore the force of other incentives and motives in general. In fact, it aims to highlight their prevalence and ask us to recognise them for what they are. A concern for reasons of protection is also not incompatible with using incentives to render protection more robust where the will to protect is weak. A monomania, however, on offers and threats as policy incentives run the risk of neglecting other forms of diplomatic pressure and soft power, such as naming and shaming. It also runs the risk of crowding out reasons of the right kind and rendering policy recommendations wrong-headed.Footnote 19 At the end of the day, a state motivated also by reasons of the right kind is more likely to protect robustly than a state that does not.
Second, it could be argued that, all things considered, it may be preferable to refrain from taking action against Poland’s currently non-robust protections at present, allowing the country instead to develop the institutional capacities necessary for implementing more robust protections and integration in the future. This approach also avoids undermining broader efforts to support Ukraine against an unjust war that threatens the international legal order. Imposing sanctions on a country facing a war near its borders might backfire and hinder these broader efforts. The animating idea behind this proposal was that such inaction at present is preferable if it makes protection and integration easier in the future because of the institutions that it helps to build up and the effect the very practice of protection and integration could have on the mindsets of those who find it difficult to extend such protection to all refugees. The proposal does allow, however, for sanctions against countries like Poland at a later point if they persist in refusing to protect refugees from other parts of the world. It may be easier to effectively sanction Poland later when there is no war on its doorstep. My point was not “don’t do anything ever” but “maybe, in this case, don’t do anything now” if you care about refugee protection robustly over the long term.
Third, it should be noted that the time-sensitive approach proposed in this paper is one among many options. The selection of the most appropriate approach should depend on relevant empirical evidence from the same or similar cases of non-compliance on the part of member-states. The proposal outlined here is based on tentative assumptions regarding the impact of sanctions during turbulent times. There is much to be said about the right choice of policy instruments that best serve a set of normative desiderata. Epistemic uncertainty is inevitable and, hence, at the core of such policy choices. These worries are relevant here but go beyond the scope of this paper. The broader aim of proposing a time-sensitive approach was to show that the adoption of a robust conception of refugee protection may require making tough choices, including counter-intuitive policies if shown to be necessary, instrumentally speaking, for best-serving refuge protection robustly. A concern for robustness does not preclude the consideration of such choices. In fact, a concern for robustness necessitates the imposition not just of diplomatic pressure on non-compliant member-states but also of heavy sanctions if these are likely to be effective in increasing compliance with its desiderata, and even more so when such measures do not significantly undermine other central desiderata of EU policy such as security.
Conclusion
I have argued that the increase in the number of refugees admitted to the EU does not necessarily indicate an improvement in refugee protection and integration. This is because two conditions must be met to provide genuine refuge.
Firstly, the provision of refuge needs to be robust in the sense of making protection and integration equally available to all types of refugees in nearby worlds and, in that sense, to all those in need of refuge. This notion of robustness was argued to be violated when the admission of refugees is conditional upon their specific characteristics or types. Secondly, the provision of refuge requires a genuine disposition towards those in need of refuge. When refugees are admitted for reasons unrelated to the core purpose of refuge, such as geopolitical interests, this second condition of genuine disposition is not met.
The article highlighted that violating these two conditions of robustness likely renders high refugee protection numbers ephemeral and undermines the commitment to long-term protection and integration. When the provision of refuge is contingent upon geopolitical calculations, conditional on offers and threats, the commitment to providing refuge is compromised. Acknowledging the challenges posed by non-compliance in non-ideal conditions, the article recognised that satisfying these two conditions may be difficult. In such circumstances, it was argued that the path forward should prioritise minimising concessions to robustness, at least in the long run. This path might involve counterintuitive measures and, rather than acting against non-compliant member-states, may entail choosing inaction as a means to an end.
Notes
UNHCR Ukraine Situation Data and Syria Situation Data. The admission of refugees from Ukraine was made practically easier by the invocation of EU’s Temporary Protection Directive (TPD), a policy option that was never considered for the case of Syrian refugees. For the purposes of this paper, I use the term refugee in a generic sense that is compatible but not necessarily limited to the cases that fall under EU’s TPD and International Conventions. Hence, a refugee can be not just a person fleeing owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership in a particular social group or political opinion, is outside the country of his nationality and is unable or, owing to such a fear, is unwilling to avail himself of the protection of that country, but also a person fleeing a broader range of life-threatening conditions.
That is, member-states should not be permitted to trade their quota and compensate other states merely because of prejudice or self-interest.
A quota formula that gives almost exclusive weight to GDP is a better way to think about capacity, because it better reflects equal proportional sacrifice, especially in comparison to the current EU formula that gives population size equal weight (i.e. a composite measure including national GDP (40%), size of the population (40%), unemployment level (10%), the number of asylum-seekers already hosted (10%) (Bovens & Bartsch, 2015).
For a criticism of quota schemes see Anker et al. (1998).
In other words, potential refugees know in the case of states like A their direct protection and integration by them depends on whether the flee from a situation that constitutes a security threat to the EU as meeting the less stringent conditions of refugee states as specified by international conventions is insufficient to motivate states like A to protect and integrate them.
See Lippert Lippert-Rasmussen and Vitikainen (2020) for a discussion of impermissible state preferences when background injustices are present.
Discrimination in the form of arbitrary selection I take it to be here an expressive harm. It is tantamount to saying to an unwanted refugee: “your reasons for fleeing are no different to those of other refugees but nonetheless you are of lesser moral worth to me than they are” (Oberman, 2020: 707).
As we shall see below this is important because unwillingness could affect capacity. In other words, preferences are endogenous to achieving maximum capacity. Further, in some cases, politicians are not decisively constrained by such preferences of the electorate but rather actively seek to propagate the need for such constraints in their constituency.
Such a policy could take different forms. It could take the form of translating official documents to languages refugees understand best, but it could also take the form of adopting a lingua franca, such as English, as the second official language of the state for reasons of integration among others. Such measures do not challenge the dominance of domestic linguistic traditions. They only challenge linguistic monopoly within state institutions for the sake of robust refugee protection and integration.
Here I agree with Gibney (2015: 457) that more often than not the invocation of such considerations has more to do with an unwillingness than actual inability of a state to admit certain types of refugees. The case for positive preferences becomes stronger, but not necessarily decisive, when a state is asked to do more than what is due to do.
Hence, B exploits A if A only accepts B’s offer for reasons related to the unequal bargaining power between them for example due to inequality in GDP per capita (Vrousalis 2023). Below I outline policy measures that could ameliorate the power asymmetry of such exchanges.
Bigger awards might work better than smaller ones, especially when they do not convey a message of inferiority to the recipient.
This is not to imply that reasons surpassing those related to refugee status can never exist. The concern, instead, is that if a state can readily cite such overriding reasons, it might be tempted to use these otherwise valid reasons unrelated to refugee status without shouldering the burden of proof to demonstrate their relevance in the given context.
In other words, in some cases a more efficient allocation of resources is unjustifiable if it requires discriminatory policies disallowed by the robust conception of refugee protection (Oberman, 2020: 710).
This is not solely an issue with non-EU countries. EU member-states like Italy and Greece do not provide adequate protection and integration to refugees not only because of both an increasing unwillingness of their right-wing parties in office to do so but also because of chronic state corruption and institutional inefficacy. EU institutions and agencies have also tolerated this unwillingness on the part of member-states that clearly violate the principle of non-refoulment as well as often turned a blind yet on these issues (BBC, 2020).
References
Anker, D., Fitzpatrick, J., & Shacknove, A. (1998). Crisis and Cure: A reply to Hathaway (Neve and Schuck). Harvard Human Rights Journal, 11(1), 295–310.
Bauböck, R. (2017). European duties towards refugees. EU of EU-ENP Cooperation: Rational, bounded rational and ethical critiques’. Journal of Contemporary European Research, 9(1), 60–77.
Bauböck, R. (2018). Refugee protection and burden-sharing in the European Union. JCMS: Journal of Common Market Studies, 56(1), 141–156.
Bauböck, R., Mourão Permoser, J., & Ruhs, M. (2022). The ethics of migration policy dilemmas. Migration Studies, 10(3), 427–441.
BBC (2020). Pushbacks: Migrants accuse Greece of sending them back out to sea. https://www.bbc.com/news/world-europe-55231203 (accessed 30th May 2023).
Bovens, L., & Bartsch, A. (2015). Why the refugee quota system is unfair to poorer eastern and southern EU states. LSE European Politics and Policy (EUROPP) Blog.
Carens, J. (2013). The Ethics of Immigration. Oxford University Press.
European Commission (2016). ‘Relocation and Resettlement: Member States need to sustain efforts to deliver on commitments’. Press Release, 9 November. Available online at: http://europa.eu/rapid/press-release_IP-16-3614_en.htm.
Gibney, M. J. (2015). Refugees and Justice between States. European Journal of Political Theory, 14/4, 448–463.
Gilabert, P., & Lawford-Smith, H. (2012). Political feasibility: A conceptual exploration. Political Studies, 60/4, 809–825.
Gneezy, U., Meier, S., & Rey-Biel, P. (2011). When and why incentives (don’t) work to modify behaviour. Journal of Economic Perspectives, 25(4), 191–210.
Hathaway, J. C., & Neve, R. A. (1997). Making International Refugee Law relevant Again: A proposal for Collectivized and solution-oriented Protection. Harvard Human Rights Journal, 10(1), 115–151.
Lippert-Rasmussen, K., & Vitikainen, A. (2020). The Ethics of Refugee prioritization: Reframing the debate. Ethics & Global Politics, 13/1, 1–5.
Lovett, F. (2022). The Well-Ordered Republic. Oxford University Press.
Milazzo, E. (2023). Refugee Protection and Solidarity. Oxford University Press.
Miller, D. (2016). Strangers in our midst: The political philosophy of Immigration. Harvard University Press.
Oberman, K. (2020). Refugee discrimination – the Good, the bad, and the pragmatic. Journal of Applied Philosophy, 37, 695–712.
Owen, D. (2019). Refugees, EU citizenship and the common European asylum system a normative dilemma for EU Integration. Ethical Theory and Moral Practice, 22(2), 347–369.
Pettit, P. (2012). On the people’s terms: a republican theory and model of democracy. Cambridge University Press.
Rawls, J. (1999). The Law of peoples. Harvard University Press.
Ruhs, M. (2022). Who cares what the people think? Public attitudes and Refugee Protection in Europe’, politics. Philosophy & Economics, 21/3, 313–344.
Sandel, M. (2012). What money can’t buy. The Moral limits of markets. Alan Lane.
Simmons, A. J. (2010). Ideal and nonideal theory. Philosophy & Public Affairs, 38(1), 5–36.
Thielemann, E. (2017). Why Refugee Burden-sharing initiatives fail: Public Goods, Free-riding and Symbolic Solidarity in the EU. Journal of Cutaneous Medicine and Surgery: Incorporating Medical and Surgical Dermatology, 56(1), 63–82.
UNHCR - Ukraine Situation Data - https://data.unhcr.org/en/situations/ukraine.
UNHCR - Syrian Situation Data - https://data.unhcr.org/en/situations/syria.
Vrousalis, N. (2023) Exploitation as Domination: What Makes Capitalism Unjust. Oxford University Press.
Funding
This research is funded by the German Research Foundation (DFG).
Open Access funding enabled and organized by Projekt DEAL.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Declarations
There are no conflicts of interest regarding the research conducted for this article. The study does not involve, and therefore require, informed consent on studies with human and animal subjects. The study does not include primary data and hence data availability is not relevant to its publication. The research carried out in this article has been supported by founding provided by German Research Society (DFG). The article is single-authored.
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Efthymiou, D. Between meeting quotas and following the duty-bound heart: navigating the formidable dilemma of refugee protection in the EU. CMS 12, 26 (2024). https://doi.org/10.1186/s40878-024-00386-z
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/s40878-024-00386-z