Steve Peers

Introduction

A recently published EU Regulation establishes new guidelines for maritime surveillance and rescue missions coordinated by Frontex, the EU’s border agency. This piece examines the potential impact of these rules on migrant deaths in the Mediterranean and the handling of asylum claims from those rescued at sea.

These new regulations are a direct response to ongoing criticisms that the EU shares responsibility for the migrant deaths in the Mediterranean. Member State authorities and Frontex have been accused of violent conduct and “push-backs,” forcing migrant vessels back to unsafe countries, a practice deemed unlawful by the European Court of Human Rights in 2012.

This Regulation replaces a 2010 Council decision concerning the Schengen Borders Code, which was overturned by the Court of Justice of the European Union (CJEU) after a procedural challenge from the European Parliament (EP). The Court ruled that an EU act addressing human rights and involving coercive measures required adoption through the EU’s legislative process.

Consequently, the European Commission put forward a legislative proposal in April 2013. Initially, a group of Member States resisted most search and rescue and disembarkation provisions, even after the tragic deaths of 300 migrants in the fall of 2013. These Member States eventually agreed, with the European Parliament advocating for the retention and improvement of the Commission’s proposal.

The new Regulation, adopted and coming into effect on July 17th, raises a critical question: will it effectively address the loss of life and push-backs in the Mediterranean?

This piece will examine these concerns and conclude by evaluating Frontex’s accountability. It is an updated analysis of the new rules.

Search and Rescue

The previous Decision included mandatory rules on interception and non-binding rules on search and rescue. However, the CJEU determined that both categories were binding. As mentioned, some Member States attempted to weaken these rules in the new Regulation but were unsuccessful.

New search and rescue rules incorporate the Commission’s comprehensive proposal, particularly the classification of vessels in a state of alert, uncertainty, or distress. If sinking vessels are identified promptly and these regulations are correctly applied, the Regulation should ensure the rescue of migrants from drowning in most situations.

Protection and Disembarkation Rules

Rules regarding the protection of migrants at risk of persecution, torture, or other ill-treatment in their home country (or another country) are more intricate.

At the heart of the new Regulation is Article 4 – safeguarding against non-refoulement (return to an unsafe nation) and upholding fundamental rights. Article 4(1) asserts that no one can be ‘disembarked in, forced to enter, conducted to, or otherwise handed over to’ an unsafe country. The EP successfully broadened the scope of the regulation by adding ‘forced to enter’ and ‘conducted to’, which explicitly covers push-backs.

Defining an unsafe country, Article 4(1) outlines two scenarios: (a) a substantial risk of the migrant facing the death penalty, torture, persecution, or other inhumane or degrading treatment; and (b) threats to the migrant’s life or freedom based on grounds outlined in the Geneva Convention on Refugees (race, religion, nationality, political opinion, or social group membership), including sexual orientation. Additionally, ‘chain refoulement’ is prohibited: transferring a migrant to a safe country that would then transfer them to an unsafe country is forbidden.

Compared to criteria in EU asylum law, the first category includes two grounds for granting ‘subsidiary protection’ (protection for those ineligible for refugee status under the Geneva Convention): the death penalty and torture or other degrading treatment. It omits the third category, concerning ill-treatment during armed conflict, but includes ‘persecution’ without specific definition. The second category aligns with Article 33(1) of the Geneva Convention but excludes the exception in Article 33(2) for individuals deemed security threats and includes ‘sexual orientation’ as grounds for protection. Notably, the CJEU has affirmed that homosexuals can constitute a ‘particular social group’ under the EU’s existing asylum law.

The 2010 Decision only referenced ‘non-refoulement’ without elaboration, while the 2013 proposal (and the Council’s stance) only cited the first category of grounds, excluding ‘persecution’. Clearly, the EP succeeded in strengthening this provision.

Article 4(2) addresses determining an unsafe country. It states that when considering migrant disembarkation in a third country, the host Member State must ‘take into account the general situation in that third country’ and cannot disembark or force entry into a country identified as posing a risk by the host Member State or other participating Member States. The EP ensured the inclusion of forceful entry, conducting, or handover.

Furthermore, the EP clarified that information sources – a ‘broad range,’ including other Member States, EU agencies, and international bodies – must be considered when assessing a third country’s safety.

Article 4(3) outlines procedural rights for migrants, stipulating that before disembarkation or transfer to a third state, Member State units should ‘use all means’ to identify migrants, evaluate their circumstances, inform them of their destination, and allow them to object based on the non-refoulement rule. These obligations are secondary to the safety of all involved (Article 3).

Operational plans must, ‘where necessary,’ include provisions for medical personnel, interpreters, legal counsel, and other relevant experts onshore. Annual reports from Frontex regarding the Regulation’s application must include details about disembarkation cases in third countries and the application of relevant criteria. The EP advocated for these stipulations, particularly referencing ‘legal advisers.’ However, the phrases ‘where necessary’ and ‘use all means’ allow for flexibility in implementation.

The regulation also restricts the exchange of personal data with third countries, mandates respect for human dignity, and sets standards for staff training.

Crucially, protection rules are intertwined with migrant disembarkation rules. Article 10 outlines three scenarios: Firstly, migrants intercepted within a Member State’s territorial sea or contiguous zone must be disembarked in that coastal Member State. However, there’s a significant caveat: a vessel reaching this proximity to a Member State could still be redirected elsewhere.

Secondly, migrants intercepted on the high seas may be disembarked in their assumed country of departure, contingent on the Regulation’s non-refoulement rules. If that’s not feasible, disembarkation ‘shall’ occur in the host Member State.

Thirdly, during search and rescue operations, migrants should be taken to a safe location. Failing that, they should be disembarked in the host Member State.

These provisions raise crucial questions. The precedence of the general non-refoulement rule over disembarkation in a third state should have been explicitly stated.

Furthermore, the Regulation’s preamble acknowledges that EU asylum law applies to anyone within EU territorial waters. This means that, per existing legislation, an asylum seeker within these waters cannot be transferred to a third state until their application is processed, with some exceptions. Consequently, asylum seekers within territorial waters must be disembarked within that Member State, as remaining on a vessel throughout the procedure is impractical. However, the Regulation’s main text, allowing for the transfer of individuals intercepted in territorial waters to third countries, contradicts this.

Another concern arises when migrants are disembarked in a Member State: is that Member State safe? Article 4(2) and (3) address the safety of third countries, while the general rule in Article 4(1) pertains to all countries. The European Court of Human Rights and the EU’s Court of Justice have both found Greece to fall short of Article 4’s standards, and legal action in some Member States is challenging Italy’s safety. This could lead to conflict between the non-refoulement rule and the obligation to disembark in a host, coastal, or safe Member State, or where applications were lodged in territorial waters.

For those intercepted on the high seas or in the contiguous zone (the Regulation overlooks interception in a third state’s territorial waters), most EU asylum legislation doesn’t apply. However, the EU’s Qualification Directive does apply, as its text doesn’t limit its territorial scope. The Regulation’s wording is unclear on this point, as it references general non-refoulement standards, which differ slightly from the Directive (e.g., omitting those fleeing conflict).

EU regulations concerning asylum procedures and reception conditions don’t apply to those intercepted on the high seas or contiguous zone; the Regulation’s rules would apply. The rules essentially establish a simplified ‘maritime asylum procedure’ for such cases. As mentioned, the phrases ‘where necessary’ and ‘use all means’ grant Member States significant leeway in implementing these rules, which are subordinate to ensuring everyone’s safety. This implies that if a safety risk arises before a non-refoulement assessment, migrants should be transported to a safe Member State to avoid jeopardizing the assessment. Once within a Member State’s territorial waters, full EU asylum law applies (arguably even if the application was submitted beforehand; if not, nothing prevents a renewed application once within those waters).

Since most EU asylum law doesn’t apply on the high seas, neither do the EU’s Dublin rules on asylum responsibility. Whether these rules would apply if an individual applied for asylum on the high seas and was transported to a Member State’s territorial waters during processing is unclear. To mitigate the burden on coastal Mediterranean states, EU Member States could establish new rules for asylum seekers applying on the high seas, potentially as an amendment to the Dublin rules, similar to the proposed amendment regarding unaccompanied minors.

At the EP’s urging, the Regulation clarifies (recital 7) that shipmasters and crews shouldn’t face legal repercussions for rescuing and transporting migrants to safety. While a step in the right direction, amending EU legislation to explicitly state that such actions aren’t criminally liable would be more effective.

Accountability of Frontex (and National Authorities)

Article 13, introduced at the EP’s insistence, mandates annual reports from Frontex on the Regulation’s application, including internal procedures, practical application, and ‘detailed information on compliance with fundamental rights and the impact on those rights, and any incidents which may have taken place’. This suggests that these reports must detail migrant disembarkation points and safety assessments of third countries (and Member States). Explicitly outlining these obligations would have been beneficial, but any accountability measure is a positive development. The EP also pushed for a preamble recital reiterating existing text in the Frontex Regulation about cooperation with third countries.

The 2011 amendment of the Frontex Regulation, at the EP’s behest, included provisions to ensure Frontex’s adherence to human rights standards. The European Ombudsman has since reviewed the implementation of these new rules. Frontex maintains its refusal to establish an individual complaint mechanism for migrants, arguing that its role is to coordinate Member State authorities’ actions, not to be held directly accountable for them.

While a dedicated complaint mechanism would be ideal, it’s accurate that Frontex primarily coordinates Member State actions. Nonetheless, these authorities also operate independently of Frontex. Since, as the new Regulation acknowledges, any EU external border control (including patrols beyond territorial waters) is tied to EU regulations, the EU Charter of Rights applies. There have been allegations of push-backs and mistreatment of migrants at external borders by Member State authorities.

Considering this, Frontex might have become a scapegoat for critics of EU border control policies, diverting attention from potentially problematic actions of Member State authorities. The European Court of Human Rights has interpreted the ‘right to life’ as necessitating an independent investigation into deaths potentially resulting from authorities’ actions. Therefore, the EU may be obligated, under the EU Charter of Fundamental Rights, to ensure that Member States investigate deaths linked to EU policy implementation, including external border control. These authorities should also be held accountable for alleged push-backs or mistreatment of migrants.

Therefore, the EU should establish a framework for independent investigations into such alleged abuses. The Commission should incorporate the findings of these investigations into its semi-annual Schengen system reports. Moreover, it’s crucial for the Commission to take a stronger stance and initiate infringement proceedings against Member States when substantial evidence suggests their authorities are responsible for push-backs or other violations.

Conclusions

This Regulation, with its valuable search and rescue rules, provisions on protection and disembarkation, and measures for Frontex accountability, is a positive development. However, the complex interplay between the Regulation and EU asylum law has resulted in a somewhat messy set of rules governing asylum applications from individuals rescued in the Mediterranean. The focus on Frontex’s accountability might be misplaced given the allegations against national authorities. The Commission’s reluctance to initiate infringement proceedings (even in the face of documented violations of other EU immigration and asylum laws) undermines the Charter’s spirit by fostering a sense of impunity among Member State authorities.

Further reading: Council of Europe Parliamentary Assembly report, June 2014

UNHCR comments on the Commission proposal

Barnard & Peers: chapter 9, chapter 26