Introduction

In the last seven years, over twenty thousand people have died trying to reach Europe by crossing the Mediterranean Sea (IMO 2021). Since the peak of what has become known as the ‘European migration crisis’, European citizens have put increasing pressure on the EU and European governments, which have in turn shifted their focus from rescue towards migration control. The EU’s border regime is rapidly expanding, in terms of both funding and integration. In its 2019 regulation, the European Border and Coast Guard Agency, Frontex, received a mandate to create its own standing corps (EU 2019). This has made it the EU’s first uniformed law enforcement agency. Recent controversies over Frontex’s complicity in human rights violations, stemming from its participation in Greek pushback operations (Christides et al. 2020), highlight existing worries about the agency’s practices (Fink 2020). However, the rapid expansion and integration of the EU’s border regime raise general moral questions about the legitimacy of an entity such as the EU, as distinct from a nation-state, coordinating and enforcing border control. In doing so, the EU clearly exercises power, not only over citizens of the Schengen Area Countries, but also over would-be immigrants. In this paper, we therefore ask: What are the legitimacy criteria for the EU’s border regime? Are there salient differences between the EU upholding a common border regime and each European state unilaterally enforcing border control?

These questions are not only of philosophical interest. They have urgent practical importance in light of recent attempts by European states to prosecute rescue missions enacted by dissenting EU citizens and by NGOs (see King 2019). If the EU’s border regime is legitimate, then arguably the rescuers are morally liable to such prosecution, despite any substantive injustice of the policies they are trying to combat. An agent has weighty reason to comply with an unjust decision so long as its source is legitimate. Illegitimacy, by contrast, undercuts such reason. Thus, if the EU’s border regime is illegitimate, then that would straightforwardly justify non-compliance with and even resistance to it.

The contribution of the present article is twofold. First, we offer a conceptual argument about legitimacy’s role in debates over border control. We give a general account of legitimacy as a justified authority and apply this account to systematise some arguments in the philosophical debate over the state’s right to exclude potential immigrants. Second, we offer a substantive normative argument about the legitimacy of the EU’s border regime. We argue that, even assuming a minimal legitimacy standard for the power that states claim over migrants, the structure of the EU’s border regime causes it to fail to meet important conditions of this standard. We argue that there is currently an unstable asymmetry between the EU’s competences in border control and its competences regarding immigration more generally. This creates conditions under which the EU wields unaccountable power over potential immigrants. Such unaccountability, we argue, corrodes the European border regime’s legitimacy.

The article is structured as follows: The first section introduces the philosophical debate on migration and borders, highlighting the somewhat underexposed difference between the concepts of justice and legitimacy in that debate. It provides a broad outline of the concept of legitimacy as justified authority. The second section then argues that accountability mechanisms are an important precondition for institutional legitimacy because of their role in securing robust rights protection. The third section outlines the core characteristics of the European border regime, highlighting its current degree of incorporation and operational capacity. The fourth section assesses these characteristics of the European border. It argues that the extent of the power involved in the EU’s efforts to control migration is not matched by mechanisms for holding it accountable and generates a moral hazard. The final section draws out the implications of our legitimacy argument for the obligations of EU citizens and for potential immigrants.

The Legitimacy of Border Control

Over the last couple of decades, the institution of border control has come under intense scrutiny from political philosophers. It is easy to see why: the modern world is characterised by a system of sovereign states with a vast degree of material inequality between them. Restrictive border regimes sustain and reproduce this inequality because they determine who has access to the institutions—labour markets, education systems, social security networks—that reliably give some populations more than others. Revoking or at least significantly reducing states’ capacity to exclude unwanted migrants would remove many of these institutional barriers and would thereby contribute to alleviate levels of global inequality (e.g. Oberman 2015). On the other hand, there are questions to be asked about whether significantly more open borders are compatible with valuable forms of community and, in particular, with the value of political self-determination (e.g. Miller 2016). Thus, political philosophers have scrutinised the institution of border control to evaluate whether it is compatible with the demands of political morality and, if so, what constraints those demands put on the shape and content of immigration regimes.

Justice and Legitimacy in Migration

These debates give rise to complex and multi-faceted issues. As is common in contemporary analytical political philosophy, these issues have normally been tackled through the lens of justice.Footnote 1 Hence, authors have asked: what does justice require when it comes to the design of immigration policy? However, an equally important but less systematically addressed question is: what standards must states meet in order to wield legitimate power in their setting and application of such immigration policy? There is widespread agreement that there is a difference between asking what makes a social order just and what makes the imposition of that order legitimate. This is what accounts for the conceptual possibilities of the illegitimate imposition of just law (e.g. a functioning welfare system in an authoritarian state) and the legitimate imposition of unjust law (e.g. a democratically authorised labour market which allows for exploitation of workers).

In order to understand what is at stake when debating the legitimacy of an institution, we need to have a closer look at the concept of legitimacy. So, what is it that needs to be legitimised? What is it specifically about the behaviour of institutions that needs justification? Legitimacy has often been characterised as the ‘right to rule’. But what exactly does ‘rule’ entail? First, in many traditional accounts of legitimacy, the concept applies to coercion (e.g. Locke 1980; Blake 2001). The reason why institutions that wield political power need to meet standards of legitimacy is grounded in the fact that this power includes the coercive means to enforce their directives. Legitimacy is thus necessary to justify the prima facie wrong of coercion. Despite the fact that international institutions, including the EU, do not generally exercise coercion, their influence and structures are nonetheless publicly and academically criticised. The core of these criticisms can be better grasped by a second concept of legitimacy as the right to rule that understands it as justified practical authority (Raz 1986, p. 21). Here, the question is whether compliance with an institution’s rules and directives is normatively demanded, irrespective of whether it has the coercive power to enforce them. Authority can be understood as providing content-independent and exclusionary reasons for compliance (e.g. Raz 1986).

From a normative point of view, a concept of legitimacy that focuses on authority captures particularly crucial questions about duties and compliance that also arise in relation to non-coercive institutions. In our view, the concept of legitimacy applies generally to political institutions, that is, institutions which seek to regulate a given domain by setting and applying rules. The focus on legitimate authority is well suited to grasp the issues that arise concerning the EU’s rules and decisions, as these seek to bind states and individuals. As discussed above, the normative issue of legitimate authority, and whether compliance is demanded, is not only created through state institutions with coercive power. It is the claim to authority and the demand for compliance that needs to be legitimised. In this sense, the concept of legitimacy as the right to rule, focusing on reasons for compliance, is also applicable to international institutions; indeed, recent contributions have started to explore this aspect of the issue (e.g. Buchanan 2010; Tasioulas 2010; Besson 2014; Scherz 2021). We will follow Buchanan’s understanding that legitimacy confers moral reasons for compliance and non-interference, leaving aside the issue of whether this amounts to a (moral) obligation to obey. Legitimacy assessments thus seek to establish whether we have moral reason to comply with a given institution (Buchanan 2018, p. 54).

The tasks of working out what makes a border regime substantially just and of identifying the conditions under which states may set and apply a border regime at all should be treated separately. The first task involves working out what a rightful distribution of burdens and benefits in global migration would constitute between the relevant actors, normally understood as migrants, host states and sending states (Ypi 2008). Having arrived at such a standard, we can evaluate immigration regimes by reference to whether they exclude only those individuals who lack justified claims for admittance. However, a standard of justice in migration does not settle the question of which states are entitled to enforce their border regimes in the first place (see Yong 2017, pp. 463–464; Ypi 2008, p. 406 note 38). The second task thus involves giving an account of when and why states are entitled to set, adjudicate and enforce their border regimes, so that these regimes generate reasons for compliance and non-interference, even when the laws of which they are composed deviate to some degree from substantial justice requirements. This is not to say that these tasks are wholly independent of each other, as they often rely on the same normative values. However, they seek to answer different questions about the distribution and justified exercise of authority.

The Legitimacy of Border Control

Most philosophical defences of the legitimacy of border control hold that it follows from a commitment to the value of self-determination. On these views, even if citizens are constrained by the demands of justice in migration, they have a collective right to make decisions about how to interpret and act on those demands. Mainstream defences of the legitimacy of border control share an approach that locates the grounds of its legitimacy in internal characteristics of the state, for example, in its national culture (Miller 2016), its relations of ownership (Pevnick 2011), or its democratic character (Song 2019). This internal focus gives rise to an explanatory problem, however. By enforcing border control, states do not only exercise power over their own citizens. They also, and primarily, exercise power over potential immigrants, whose movement they seek to control. Therefore, we need to know what standard states are required to meet in order to wield legitimate power over outsiders.

This question has received relatively little attention in the philosophical literature compared to the cluster of questions surrounding what constitutes justice in migration. There are important exceptions, however. In a seminal article, Arash Abizadeh (2008) argues that sustaining a unilaterally imposed border regime, backed up by force, constitutes coercion in the sense that it triggers a demand for democratic standards of legitimacy. On this view, a state’s claim to enforce border control would only be legitimate in the sense that generates reasons for compliance and non-interference if it were authorised democratically by a constituency with global scope. In response to Abizadeh’s challenge, authors have argued that states do not coerce potential immigrants (Miller 2016, pp. 70–75) and that sovereignty over immigration policy should be seen as part of the bundle of rights that are protected by international or external legitimacy, that is, the right to non-interference (Yong 2018).

External legitimacy is commonly assumed to obtain for internally legitimate states that do not engage in aggressive acts beyond their borders (van der Vossen 2012). This implies that a state’s treatment of outsiders is relevant for its legitimacy: a state that systematically violates the rights of outsiders would be liable to outside interference, even if its relation to its citizenry optimally satisfied (for example) the requirements of democratic legitimacy. Following on from this thought, Yong argues that border control is part of the bundle of privileges conferred upon the states that enjoy external legitimacy. As Yong argues, a plausible standard for determining the set of states that enjoy external legitimacy is respect for human rights. On this view, which is implied by the mainstream defences of the legitimacy of border control (e.g. Miller 2015, p. 392), states have legitimacy in their enforcement of border control provided that they refrain from violating the human rights of potential immigrants.

Border control is a special case for a theory of external legitimacy because it represents a genuine instance of a claim to authority by states over individuals who reside outside of their jurisdictions. This is one of Abizadeh’s core insights and it does not depend on his further claim that border control amounts to coercion in the sense that triggers a demand for democratic legitimacy. Of course, it is highly contested whether Abizadeh is right that states need to justify their actions to outsiders as well as insiders by the same standard (cf. Hidalgo 2015, pp. 459–466).

If we start from the concept of legitimacy as justified authority, it becomes clear that authoritative rules need to be justified to those they seek to bind in order to generate exclusionary content-independent reasons for compliance. Since border regimes seek to bind citizens and potential immigrants, they need to be justified to both of these groups. While the border regime of a particular state is justified towards its citizens if the state functions democratically, the same is not true for non-citizens. The main question is, therefore, how border regimes need to be justified towards potential immigrants to generate exclusionary and content-independent reasons for compliance. It is important to note that the lack of such reasons means that migrants are justified in avoiding and even resisting border control. In this resistance, they are not even constrained by the conditions of civil disobedience—such as publicity or non-violence—that apply to legitimate but unjust regimes. Regarding the rescue missions in the Mediterranean, a further question is whether the illegitimacy of the border regime that is not justified towards migrants also affects how those taking part in rescue missions (often EU citizens) need to comply with it.

Legitimacy Standards for Border Control: Accountable Human Rights Protection

For the sake of our argument in this paper, we will grant that states wield legitimate authority over potential immigrants provided that they protect core human rights in their enforcement of their border regimes. This includes respecting the rights of the Convention Relating to the Status of Refugees 1951 and 1967 Protocol, in particular the principle of non-refoulment. This principle holds that a refugee shall not be returned to a country where they face serious threats to their life or freedom. We take this to be the standard assumption under international law (Criddle and Fox-Decent 2021).Footnote 2 This legitimacy standard does not demand democratic inclusion of potential immigrants in the political decisions about a state’s border regime, since they are not subjected to the state in the same, all-encompassing way as permanent residents. Nevertheless, we will argue that this seemingly limited requirement has far-reaching implications when applied to the supranational organisation and enforcement of border control, such as the EU’s border regime. To satisfy a protection of human rights standard, institutions must include suitable accountability mechanisms. The rights of refugees and asylum seekers imply access to a fair and efficient asylum procedure. This procedure, and the rights upheld through it, should not just be fulfilled coincidentally. Instead, they should be robustly ensured through accountability mechanisms.

As we understand the term, accountability implies that an institutional agent is held to a specific standard by some external agent with the requisite power to impose sanctions upon them if they fail to discharge their responsibilities according to the relevant standard (Grant and Keohane 2005, p. 29). Accountability is important for legitimacy because it functions to constrain institutions from abusing the power with which they are entrusted. Accountability mechanisms always apply ex post. Once the relevant standard is violated, they ensure that potential abuses of power are sanctioned. This means that they will also have a deterring effect, thus granting accountability mechanisms a certain ex ante power (Grant and Keohane 2005, p. 30).

Why does accountability matter? In short: because it functions to ensure that institutional agents fulfil the standards that ground their right to rule. This is important both on instrumental and intrinsic grounds. Instrumentally, accountability makes it more likely that the protection of human rights is provided. It is certainly possible that institutions will act to satisfy the standards that ground their right to rule simply by themselves. However, the reality of politics means that institutions and their officers will be faced with incentives to act in ways that contradict the grounding of their mandate. For example, it is certainly possible that an institution will perform its tasks perfectly simply because each of its members has a strong conviction in its mission, but without any mechanisms for holding the institution to account, such a perfect performance is unstable. Intrinsically, accountability mechanisms are constraints that not only make it more likely that human rights are fulfilled but also that there is a certain control over their fulfilment. Thus, they are changing the mode of fulfilment from an incidental to a robust one (Pettit 2012). Without such insurance, migrants are subjected to the arbitrary power of border guards and immigration officers and, by extension, of the border regime these work to enforce (Sager 2017). Accountability mechanisms function to ensure that institutions can be robustly expected not to violate baseline constraints on their legitimacy—in the case of border control, these are core human rights.

At the domestic level, accountability is usually understood in terms of democracy. Given that political philosophers and theorists almost invariably hold that the power states exercise over their citizens is subject to a democratic legitimacy standard, it can appear paradoxical to demand that states should be accountable for their acting beyond their borders. If governments are accountable to their citizens, this line of argument goes, they cannot also be accountable to non-citizens outside of their borders. If citizens demand that their state acts in a certain way abroad, then this is all that matters (Lafont 2010, p. 196). Thus, to take the example of border control, it can appear incoherent to hold that states should be held accountable both to their citizens who demand strict immigration regimes and to the migrants whose human rights they should respect (cf. Nagel 2005). This conceptual objection to accountability beyond borders is mistaken, however. It turns on imposing a representative interpretation of democratic accountability onto the power states wield over non-citizens beyond their borders. However, states can be accountable to individuals without including them in the decision-making structures that authorise their actions. There are many sources of accountability other than democracy, including fiscal, legal and reputational accountability (Grant and Keohane 2005, p. 36).

Assuming for the sake of argument that border regimes are not subject to democratic accountability, they still require critical accountability mechanisms: first, transparency of processes and decisions; second, contestation rights to both legal remedies in the case of rights violations and the right to appeal decisions regarding asylum claims; and third, independent legal oversight. Transparency is an indispensable precondition for any form of accountability. Contestation rights are required as control over political authority that is less all-encompassing than the subjection to a state’s law but still affects important and sensitive areas of individuals’ lives (Scherz 2021). Finally, independent legal oversight is important for accountability to ensure that any complaints and challenges of decisions are taken seriously and that sanctions are implemented where necessary.

The EU’s Border Regime

This section outlines core characteristics of the EU’s border regime which distinguish its independence from the individual border regimes of the Schengen Area Countries. We highlight three such characteristics: (1) the high degree of integration of the EU’s border agency, Frontex; (2) the internal incentives for each Member State engendered by the Dublin Regulation; and (3) the externalised reach of the European border regime. We argue that in conjunction, these points suffice to show that the EU’s border regime is more than the sum of its constituent parts. We will show that the EU’s border regime creates accountability issues, perverse incentives, and externalises power.

Integration and Exclusion: The Case of Frontex

Despite the acceptance of the four freedoms within Europe, a core principle in the European integration process has been that EU members and associated countries retain full sovereignty over borders (e.g. Goodman and Schimmelfennig 2020). The main argument brought forward by the EU is thus that it lacks competences in the domain of immigration: it cannot impose any demands on its members or associates with regard to how they deal with non-European immigration, including refugees and asylum seekers. When it comes to the policy area of border control, however, the picture looks rather different. Here, the EU has imposed a series of reforms that unifies the European border regime to a significant degree.

This development was principally driven by the EU’s Border and Coast Guard Agency, Frontex. When the Agency was founded in 2004, its mandate was restricted and its responsibilities were strictly limited to coordination. Yet, in the wake of the European ‘migration crisis’, the Agency has grown rapidly in both size and competences. After its 2016 regulation, Frontex received increased competences vis-à-vis the Member States whose border regimes it is tasked with coordinating and strengthening (Fjørtoft 2020). Notably, Frontex was given powers to engage in ‘vulnerability assessments’ of Member States’ capacity to manage their borders, and (subject to a council decision) to intervene in the border regimes of Member States, raising familiar concerns of European integration granting the EU powers to interfere in domestic political affairs (Deleixhe and Duez 2019). More distinctively, however, Frontex’s 2019 regulations gave the Agency unprecedented powers in the context of European integration. In particular, the Agency was granted the power to build its own ‘standing corps’ of an initial 10,000 officers (EU 2019). This makes Frontex the ‘first uniformed law enforcement agency of the European Union’ (Frontex 2020). In practice, this means that the borders of the EU will increasingly be controlled—and the border law of its Member States will be enforced—by officers who derive their mandate from the Union itself.

The important point to note for our principled argument is that the border controlling part of the EU’s border regime is characterised by high degrees of integration. As such, in the EU’s competence in the domain of migration there is an asymmetry: whereas the EU has increasing powers in the area of border control, it has no powers in setting immigration policy, even for vulnerable migrants such as refugees and asylum seekers.

The Dublin Regulation and Perverse Incentives

One of the most important international treaties regulating migration is the Convention Relating to the Status of Refugees from 1951 and the 1967 Protocol expanding its temporal and geographical reach beyond the Second World War. These legal instruments define the term ‘refugee’, the rights and duties of refugees, and states’ duties to protect them. These duties include in particular the principle of non-refoulment: a refugee should not be returned to a country where they face serious threats to their life or freedom. However, with regard to the EU, there are further internal regulations to rapidly determine the Member State responsible for an asylum claim, the so-called Dublin Regulation. Chapter III of the Dublin II RegulationFootnote 3 determines the Member State responsible for a person’s asylum claim. It sets out several criteria, such as having family members in a Member State or a valid residence permit, but the criterion that is most often used for Dublin transfers is the principle state of first entry criterion in case of irregular border crossing (Article 10).

This criterion of the Dublin Regulation creates perverse incentives in several ways. First, it has not only led to an unjust distribution of refugees within the EU (Holtug 2016; Bauböck 2018), but in addition, given the unwillingness of other Member States to redistribute refugees or to share the costs, it has overburdened the asylum systems of ‘frontline’ states such as Greece and Italy. As a result, the system weakens effective rights protection. This is so when judged against not only an ideal standard but even against what would be possible if migrants could travel to their preferred state to launch asylum claims. Second, this criterion creates the incentive for frontline states to minimise the entry of migrants to their territory, the registration of persons, and launching of asylum claims. This leads to behaviour ranging from omissions of registering the arrival of a migrant in the country in order to let them travel to another county to launch their asylum claim, to so-called pushback operations, for example, by Greek border guards (Christides et al. 2020). Third, the effective breakdown of the Greek asylum system and related human rights violations led several states to suspend Dublin transfers to Greece after the ECtHR decision M.S.S. v. Belgium and Greece (2011). Moreover, the Dublin III RegulationFootnote 4 prohibits the transfer of persons to another Member State if there is a risk that they will be subjected to inhuman and degrading treatment. This also creates the perverse incentive that leaving their asylum system broken becomes the rational option for struggling states.

In September 2020, the EU introduced a Pact on Migration and Asylum with the explicit aim of rectifying some of these issues. The Asylum and Migration Management Regulation, which will replace Dublin III, introduces a ‘solidarity mechanism’ which can be triggered by the Commission to assist a Member State experiencing migratory pressure.Footnote 5 However, the obligations that the solidarity mechanism imposes on other Member States can be discharged along several lines, for example through ‘return sponsorship’ of migrants without asylum claims. Thus, they do not include mandatory relocation requirements. Moreover, the Regulation could possibly impose new obligations on frontline states to enact new pre-screening requirements. This means that the reform does not target the causes of the existing problems with the Dublin regulations. Because states can voluntarily choose which form of ‘solidarity’ they want to enact, they are still exposed to the three perverse incentives outlined above.

The Externalised Reach of the European Border Regime

In addition to integration and asymmetry, and the perverse incentives these engender, the EU’s border regime is also characterised by externalisation. Frontex is only one part of the EU’s ‘Integrated Border Management’ (IBM) strategy, which explicitly seeks to increase the efficiency of European border control by pushing it outward beyond the borders of its Member States. IBM aims to track ‘the movement of third-country nationals from the point of departure in countries of origin, all throughout transit, and up to their arrival in the EU’ (Moreno-Lax 2017, p. 3). At every opportunity, the EU deploys ‘remote control’ techniques to block unauthorised migration towards Europe. One central technique is the imposition of carrier sanctions on airlines and transport companies. Another is placing so-called immigration liaison officers—tasked with advising governments and transport companies on ‘European immigration priorities’—in third countries.Footnote 6

Perhaps the best-known strategy, however, is the creation of what are known as ‘migration compacts’. The first of these was concluded between the EU and Turkey in 2016 and it has since, despite the controversy surrounding it, become the model for how the EU intends to deal with migratory pressure (Reslow 2018). Migration compacts are bilateral agreements made between the EU and third countries, where financial and political benefits are exchanged for assistance in hosting refugees and asylum seekers heading towards Europe and promises to shut down active migration routes.

The EU is not alone in seeking to externalise its border regime. Most of the world’s most powerful states are relying on similar techniques to increase their border controlling capacity and, in particular, to avoid the costs associated with adjudicating asylum claims made on their territories (Shachar 2020). First, it should be noted that this does not make the practice any less problematic. Second, bilateral agreements between two states A and B do not usually establish the authority of state A over the citizens of state B and vice versa, because state B itself binds its citizens through the agreement. This is because states are generally presumed to act as trustees of their citizens. However, in circumstances where this relationship is clearly not functional, for example in failed or occupied states, the authority is exercised by both states (Ganesh 2021). This argument also applies to the EU migration compacts, if the states that are partners in such compacts are not accountable to migrants (even in the more restricted way suggested in this paper), over which they seek to exercise authority. The EU must have known about the worrying human rights conditions in Turkey’s asylum system, due to which Turkey is not a reliable trustee and does not exercise legitimate authority over migrants. Therefore, the EU itself exercises authority through its migration compacts over migrants in third countries such as Turkey.

Evaluating the EU’s Border Regime

Having outlined a conception of legitimacy as accountable human rights protection and given a brief outline of the current shape and independence of the EU’s border regime, this section evaluates that regime’s legitimacy. To recall, our argumentative strategy is to grant that the relevant legitimacy standard for each state’s claim to unilaterally control its borders is set at an accountable human rights protection. This allows us to better isolate whether the structure of the EU’s current border regime gives rise to particular legitimacy challenges. We find that it does. The asymmetrical nature of the EU’s border regime creates a state of affairs in which the EU wields unaccountable power over vulnerable migrants, which in turns creates a moral hazard that exposes migrants to increased and novel risks of abuse.

The creation of a European border regime has had a demonstrable effect on migrants. The decision to cut down on rescue operations in favour of more stringent border control measures has an impact on those migrants capable of entering European territory to claim asylum, instead of waiting (sometimes for years) in camps, and who can do so relatively safely. In particular, the externalisation involved in the EU’s integrated border management strategy does not solely reduce the number of migrants heading towards Europe, but also ensures that those who do still choose to go have to set out on more dangerous paths in order to reach European territory to claim asylum (FitzGerald 2019). It is clear that the EU’s border regime impacts potential immigrants by way of making the border regimes of each of its constituent states more effective. This is not problematic as such, as we have granted for the purposes of our argument that each state is entitled to uphold a border regime, given that they provide accountable human rights protection for migrants. Yet, we argue that the features of the EU, highlighted in the last section, systematically undermine this accountability and even lead to a moral hazard.

Accountability

Formally, of course, the EU is bound by human rights norms. Indeed, in all of its central documents for border management, as in its foreign policy more generally, the EU stresses that the human rights of migrants are central to its conduct (e.g. EU 2019). However, it has been demonstrated that the EU’s border regime is characterised by a series of features that make it unaccountable with regard to human rights protection. This unaccountability, we argue, is a serious issue for the legitimacy of the EU’s border regime. We should note again that the accountability at issue is not democratic accountability. As we take potential migrants to be subject to some of the EU’s power but not as fully as permanent residents, we are interested in accountability in the essential minimal sense of transparency, contestation, and legal oversight.

Activists and legal scholars alike have been worried over what the integration of European border control through Frontex entails for the protection of the rights of individuals who attempt to access Europe. Their critiques have emphasised how difficult it is to hold Frontex to account for potential rights violations. Since the EU has not ratified human rights treaties and thus cannot be held accountable by an external international court,Footnote 7 the only clear path available for migrants to contest any maltreatment they receive at the hands of Frontex officers is through a complaints mechanism that is internal to Frontex—and largely deemed ineffective (Fink 2020).Footnote 8 This complaints mechanism largely involves disciplinary measures against individual officers (EU 2019, annex V, Sect. 4). Many have justifiably worried that these measures are insufficient for the expansive kinds of power Frontex wields over migrants, especially after the 2019 regulations (Fink 2020). Consider, by way of comparison, military forces whose misconduct could only be sanctioned by internal disciplinary procedures. Hardly anyone would deem this sufficient to hold armies to account—hence, the need for institutions such as the International Criminal Court. Yet, no similar accountability mechanism exists for Frontex’s border guards, despite the fact that the 2019 regulation authorised Frontex’s standing corps to carry weapons (Regulation (EU) 2019/1896, annex V, Sect. 1).

The externalised nature of the EU’s border regime creates further accountability problems. By externalising its border regime, the EU not only makes its border regime more effective by way of administering migration flows ‘in the affected areas’, but also makes that administration far less transparent, since it is conducted outside of European territory and dispersed across a variety of different private and public actors. Under these conditions, it is much harder to uncover potential maltreatment of migrants, which is why activists and NGOs have seen monitoring missions as one of their most important strategies for reducing human rights violations in migration control (Fitzgerald 2019, pp. 54–55). Furthermore, and in addition to the problems associated with the EU’s unclear status before human rights courts, the externalised nature of the European border regime creates legal barriers between the EU itself and the actions performed on its behalf. The essential function of the externalisation of the European border regime is to frustrate migration before it reaches European territory where individual migrants can launch asylum claims (Moreno-Lax 2017, p. 3). Moreover, migration compacts essentially involve hosting refugees and asylum seekers on non-European territory. Thus, the externalised European border regime works to frustrate the movement of migrants who would otherwise have been able to move across the territories of neighbouring countries, but without incurring the international obligations that would have arisen were those migrants territorially present in a European state. We also know that these strategies have detrimental effects on the rights of those migrants (FitzGerald 2019; Moreno-Lax 2017; Schachar 2020). Yet, the EU is currently able to shift the blame and to allege that these effects are due to the negligence of their collaborators since the legal documents that form the basis for their externalising techniques specify that all actors must respect human rights.

These accountability problems are structural features of the EU’s border regime in its current instantiation. In the absence of that regime, or under a regime that better tracked authority as functional jurisdiction (see Shachar 2020, pp. 75–82), migrants would plausibly face less risk of rights violations. Given the risks associated with the accountability deficits of the EU’s border regime, simply abstaining from human rights violations is thus insufficient for conferring legitimacy onto the power involved in that regime. Rather, the regime should include accountability mechanisms that ensure that rights protection is robust.

Moral Hazard

One immediate objection to this accountability argument is that it merely points to contingent features of the EU’s border regime. According to this objection, it is of course detrimental to the legitimacy of the regime if its institutions actively violate human rights. However, there is no good reason to think that the current problems faced by the EU’s border regime are anything more than superficial and that they cannot be addressed by the EU itself—for example, by the Member States demanding more stringent supervision of Frontex officers. Thus, our argument that the EU’s border regime faces unique legitimacy challenges is undermotivated: if no migrants had their rights violated, then there would be no legitimacy worries facing the EU’s border regime. The EU is just like a state in all relevant respects.

However, there are strong reasons for thinking that such optimism is unfounded. As we have noted above, there is an asymmetry at the heart of the EU’s border regime. On the one hand, the EU wields significant power in border control whilst, on the other, having no competences over immigration policy. This asymmetry makes the risks generated by that regime robust, rather than contingent. This is because the asymmetry exposes states to a specific perverse incentive: governments keen to keep anti-immigration forces in their domestic constituencies at bay have everything to gain from delegating the enforcement of their border regime to an agent that cannot make them accept unwanted asylum seekers—even (and especially) when those asylum seekers have claims that are protected under international law. This incentive is likely to persist so long as the EU fails to erect institutional mechanisms which secure joint responsibility for vulnerable migrants. Thus, the failure of the new Pact on Migration and Asylum to address the issue of relocation, as well as its potential for introducing new obligations for frontline states, means that it is unlikely to effectively break the perverse incentive to minimise human rights obligations to which European states are currently exposed. Indeed, a recent report on the new Regulation argues that the option of solidarity through ‘return sponsorship’ will merely exacerbate the problems by creating an oversupply of return instead of (as is required) paths for the protection of vulnerable migrants (Carrera et al. 2021, p. 177).

This perverse incentive amounts to a moral hazard. The EU has strong and increasing powers to regulate migration, but no powers to guarantee the satisfaction of migration-related human rights. European states can thus secure their borders by delegating tasks and competences to the EU level, whilst avoiding the responsibilities that otherwise would follow those tasks and competences—especially the obligation to accept and adjudicate asylum claims. This unsurprisingly exposes vulnerable migrants to increased risks of abuse, since the logic underlying the system functions to avoid responsibility and accountability for their basic human rights. We acknowledge that the new Pact also seeks to enhance fair trial guarantees and includes a monitoring mechanism for ensuring respect of fundamental rights. However, here also there are worries that these measures are not sufficient, in particular as it remains unclear whether the monitoring mechanism will be independent and whether it will also apply to EU institutions (Carrera et al. 2021, p. 62). In conclusion, while the new Pact on Migration and Asylum seeks to address some of the issues we discuss, at this stage of its implementation it remains doubtful that the new regulations can overcome the fundamental problems of the EU’s border regime.

The Case of Non-compliance and Interference Reconsidered

What follows from the legitimacy deficits of the EU’s border regime identified here? In this section, we want to highlight two implications that warrant consideration: the weakened compliance requirements for (1) migrants and (2) EU citizens.

First, the authority of the EU’s border regime exercises over potential migrants outside of its territory, and within its asylum system, is highly problematic and has not been sufficiently legitimised towards them—even assuming a minimal legitimacy standard. Now, what does this mean for the appropriate behaviour of migrants towards this regime? One framework that is useful to answer this question is that of civil disobedience.Footnote 9 John Rawls’s liberal approach takes civil disobedience to be appropriate in nearly just societies with legitimately established democratic authority, in which serious violations of justice occur (Rawls 1999, p. 319). He defines ‘civil disobedience as a public, nonviolent, conscientious yet political act contrary to law usually done with the aim of bringing about a change in the law or policies of the government’ (Rawls 1999, p. 320). This means the legitimacy of the regime constrains permissible acts of non-compliance to civil disobedience, which must fulfil conditions of conscientiousness, publicity, and non-violence. This distinguishes acts of civil disobedience from other forms of resistance and revolution. Unauthorised immigration has sometimes been characterised as an act of civil disobedience (e.g. Cabrera 2010). However, if the EU border regime is illegitimate, as we have argued, it does not generate content-independent reasons for compliance. In this case migrants are not bound by the rules of this practice and are normatively not required to comply with them. Furthermore, the fact that the regime is not only unjust but illegitimate means that the requirements of civil disobedience, such as publicity and non-violence, do not restrict the permissible resistance—though violence may be prohibited on other content-dependent grounds. Thus, the border regime’s illegitimacy means that potential migrants can rightfully resist it and try to undermine it even without publicly aiming for social and political change. This also shows that the rhetoric often used against unauthorised migrants as lacking in respect, or even criminal, is inappropriate.

Second, the rescue missions in the Mediterranean have received considerable public attention in recent years. This is not only because they highlight the dangers of this migration route, but also because several of these missions have had brushes with the law, in particular when they were not given permission to enter ports but still did so. How should we regard those incidents from the perspective of legitimacy? Do the rescuers have exclusionary content-independent reasons to comply with these rules? We argue that EU citizens also do not need to comply with the rules of an illegitimate border regime. To support this argument, however, we need to address several objections: First, as these are national—e.g. Italian—laws and decisions, they do not really concern the EU border regime’s legitimacy. Therefore, rescue missions do not at first seem to capture non-compliance with EU rules but rather domestic rules. In response to this objection, it is also important to consider the position in which Italy finds itself through the Dublin Regulation. As such, it is a two-level issue: As the EU rules have changed Italy’s normative relationship to arriving migrants (thereby exercising authority), Italy has changed the normative relationship of rescuers (e.g. to enter Italian ports) at least partly as a reaction to the EU border regime. Since the EU and its Member States’ border regimes are intertwined in this and other complex ways, Italian border rules form part of what we have argued to be an illegitimate practice with which the rescuers need not comply.

Assuming that the EU and Italy have been democratically legitimised, one might also object that they hold legitimate authority over the rescuers (in particular if they are EU citizens). In response to this point, we argue that, while the EU overall might be legitimate, its border regime is not. This differentiation raises a difficult question in respect of how the authority of distinct issue domains can be illegitimate while the overarching institution remains legitimate. In particular, this differentiation may be problematic if we understand legitimate authority to be content-independent and therefore binding, even if it errs in individual points. To be sure, on the standard Razian account of legitimacy, domain and subject fragmentation is not a problem. The legitimate authority over a specific individual depends fully on how well an authority makes her comply with reasons in a given domain. This means that there are not only differences in the authority that, for example, a pharmacist exercises in the different domains such as drug use or jazz (domain fragmentation), but also variation in whom is subject to her authority due to differences in customers’ knowledge about drugs (subject fragmentation). In this sense, it can be the case that the EU has legitimate authority over other domains, even if its border regime is illegitimate. To be clear, we do not defend the stronger claim that the EU is illegitimate overall as this goes well beyond the scope of this article to evaluate the appropriate legitimacy standards for the EU.Footnote 10

In addition, one might object that if we accept domain fragmentation, we should also accept subject fragmentation. In other words, while potential migrants are not bound by the EU’s border regime, the democratic legitimation given to EU citizens does bind them. However, there are at least two good reasons to avoid subject fragmentation for political institutions: the reliability of law and mutual respect of those subject to the law. First, subject fragmentation generates a problem of reliability since it makes it unclear whether the laws apply to all whom they claim to bind. This means that subjects of the law cannot know whether others are bound in the same way as they are. For political institutions this is very problematic as it undermines the mutual insurance that binding laws are meant to create. Second, if some groups subject to an institution’s rule would remain bound by it, even though the institution violates the legitimacy criteria for another group that it claims authority over, this would mean disregarding the violation. In other words, the second group would be regarded as less important, thereby violating the fundamental requirements of equal respect. For example, within an illegitimate apartheid state, the state lacks legitimate authority over both its black and its white citizens, who do not owe compliance to the regime even if they themselves have the right to vote. In fact, the continued compliance of white citizens will provide stability to an illegitimate regime and therefore cannot be required. Nevertheless, there are differences in the permissible non-compliance of potential migrants and EU citizens. In particular, the privileged position of EU citizens, due to the border regime, means that only actions of non-compliance that seek positive social and political change of this regime or benefit those disadvantaged by it are permissible.

Conclusion

We have identified legitimacy deficits in the EU’s border regime and argued that these would persist even if the EU’s border guards, as well as those of its Member States, did not violate migrants’ human rights. At the heart of our argument is the unaccountable power on which the EU currently relies to enforce its regime. We have argued that unaccountability is explained by a perverse incentive engendered by the asymmetrical nature of the EU’s approach to migration. Rectifying the legitimacy deficits of the EU’s border regime will thus be a matter of making it more symmetrical so that powers to control borders better satisfy migration-related human rights and accountability standards. Our argument says nothing about the direction in which the asymmetry should be rectified. One solution could be to scale up the EU’s competences in setting immigration policy, for example by instituting a common asylum regime. Alternatively, the EU’s competences in border control could be scaled back. Given the politics surrounding migration in Europe, neither alternative currently looks viable. However, we contend that the legitimacy worries surrounding Europe’s approach to migration will persist until the asymmetry is resolved.

Finally, a larger question looms with regard to the potential spillover effects of an illegitimate (border) regime on the overall legitimacy of an institution, in this case the EU. Our argument suggests that a regime that is internally legitimate, but not externally so, may thus also lose internal legitimacy vis-à-vis its own citizens for a particular issue domain (see Nili 2018). This raises the further question of whether the negative legitimacy assessment of an institution in one issue domain can also delegitimise its authority over other domains. This depends on how important this domain is in general and how central it is to the purpose of the institution in question. The EU’s border regime thus provides a new locus for the central question of what the function of the EU should be (Sangiovanni 2019). For now, we conclude that it is unsustainable for the EU to take on a state-like role in one area without also having the powers to discharge relevant responsibilities in that area.