EU-Tunisia like UK-Rwanda? Not quite. But would the New Pact asylum reforms change that?

EU law only allows sending asylum applicants to third countries if they can be considered safe and have a reasonable connection with the applicant. However, if the New Pact reforms go ahead, the safe third-country concept may be used to further shift protection responsibilities outside the EU. While the EU would be unable to conclude a UK-Rwanda-style agreement, the pressure for reforms, as well as the recent Memorandum of Understanding with Tunisia, signal the threat of diminished human rights protections and safeguards.

Following an agreement on the last piece of the reform package, the so-called crisis instrument, the Council and European Parliament are currently negotiating all proposals included in the New Pact on Migration and Asylum. While the focus on preventing irregular arrivals and increasing returns has long been a priority for the European Council and has been widely debated in EU migration policy, the possible impact of Pact reforms on the Safe Third-Country (STC) has gone largely unnoticed, partly due to its technical nature.

However, the use of this concept could be expanded, in line with the original Pact proposal by the Commission, unless the European Parliament’s approach prevails over the Council’s position. Should the Parliament fail to stay the course, the reforms could make it easier to transfer asylum applicants back to countries of transit, where they could be exposed to human rights violations. Recent developments point in this direction.

Faced with an increase in irregular arrivals, ahead of the February 2023 meeting of the European Council, eight member states called on the European Commission and European Council to “explore new solutions and innovative ways of tackling irregular migration” based on “safe third-country arrangements”. Reflecting an established trend of outsourcing to third countries migration management responsibilities, in July 2023, the Commission signed a Memorandum of Understanding (MoU) with Tunisia on border controls and returns in exchange, among others, for economic and trade benefits. This MoU, singled out as a blueprint for similar arrangements by von der Leyen in her 2023 State of the Union address, was concluded at the urging of the Italian government, as it wants to use the concept of STC to return those arriving by sea via Tunisia.

However, some member states aim to go further, with some commentators arguing that their goal is cooperation with third countries based on the 2022 UK-Rwanda deal. In this controversial deal, Rwanda committed to readmit asylum seekers arriving irregularly in the UK in exchange for development aid, regardless of whether asylum seekers had ever lived in or even entered the country before. Explicit bilateral declarations of support for this policy came from Austria and Italy. Denmark even negotiated a similar deal with Rwanda in September 2022. However, the Danish plans were put on hold, with the government expecting an EU-wide approach.

These developments raise several crucial questions regarding EU migration and asylum policies. Under EU law, are UK-Rwanda-like deals lawful? What difference, if any, could the Pact reforms bring in this respect? How should the MoU with Tunisia be seen in this context? And will the concept of STC be used to shift responsibility to third countries in the future, and with what potential consequences for human rights protections?

Externalisation: A U-turn on the concept of safe third-country

The concept of STC emerged at the end of the 1980s to ensure a fair burden sharing of responsibility for asylum applicants among members of the international community. Under its original conception, the states most impacted by refugee flows could send asylum applicants to another state on the condition that the receiving state could offer them adequate protection in line with international standards.

In recent years, however, restrictive immigration policies aimed at deterring irregular arrivals have proliferated, including outsourcing responsibilities to third countries. These policies, commonly referred to as ‘externalisation’, seek to transfer migration management and international protection responsibilities to countries already impacted by refugee flows, mostly countries of transit, in a spirit of burden-shifting rather than burden-sharing. These also come with diminished safeguards.

Notable examples from outside the EU include the controversial policy by Australia of setting up extraterritorial asylum processing in Papua New Guinea in 2001. In the European context, the idea of setting up extraterritorial processing centres was first suggested by the UK in 2003 and then by Germany in 2005. This idea regained attention in the aftermath of the so-called 2015 refugee crisis, with several heads of EU states calling for asylum external processing. However, legal hurdles and the lack of willing neighbouring countries prevented it from becoming  an EU mainstream migration control mechanism. The use of the STC concept to declare an application for international protection inadmissible without an examination of its merits prevailed instead, with the purpose of transferring the person who filed it to a third-country which is considered safe. Under current EU rules, the safety of the country is assessed on the basis of protection from persecution, serious harm, and respect for the principle of non-refoulement. However, this concept can only be applied individually, when there is a sufficient connection between the asylum seeker and the third- country, defined under national law, which makes it reasonable to return to that country.

The much-debated 2016 EU-Türkiye Statement is considered the blueprint of this approach. Under the statement, Türkiye  agreed to readmit all asylum seekers who crossed into Greece from Türkiye. In line with EU law, asylum applicants can only be returned after considering the time of stay there, access to work or residence, existence of social or cultural relations, and knowledge of the language, among others.

Unlike the EU-Türkiye Statement, the UK-Rwanda agreement does not foresee the need for any such connection. Under the deal, the UK would, in principle, be able to send to Rwanda Syrians, Albanians and Afghans without them having ever set foot there. However, as of October 2023, the deal remains on hold, as the Court of Appeal found that it overlooks serious deficiencies in the Rwandan asylum system, rendering it unsafe.

The expansion of safe third-country under the New Pact on Migration and Asylum

Considering the above, Denmark or other member states could not transfer asylum seekers to a third-country, regardless of their circumstances. As the European Commission and experts argued, this would only be possible if a reasonable connection with the applicant could be established.

Things may nevertheless change if the Pact negotiations were to proceed in accordance with the compromise reached by member states in June 2023. The Council’s compromise text keeps the reasonable connection requirement while also leaving to member states the exact definition of STC under national law, as in the current rules. However, two possible changes could broaden its applicability to facilitate returns.

First, albeit non-legally binding, the recitals introducing the Council’s agreed position explicitly mention that a mere ‘stay’ in a STC could be regarded as sufficient for the reasonable connection requirement to be met. If the changes were to pass in this form, it is not unforeseeable that some member states will push to return asylum seekers to countries of transit, even if they only have a tenuous link with them, although case-law of the EU Court of Justice points against this interpretation.

Second, the position agreed by the Council in June would allow member states to presume the safety of a third country when the EU and the said third-country have decided that migrants readmitted there will be protected in line with international standards. However, the mere existence of an agreement does not prove its safety in practice. For example, the recently agreed MoU with Tunisia was called into question for its human rights implications and overlooking abuses against migrants, arguably rendering it unsafe.

Despite Italian wishes to the contrary, Tunisia has so far only expressly agreed to readmit its own nationals. However, considering the expectations of EU states, the Council's position, and an uncertain future, the possibility of more comprehensive arrangements with other countries, such as Egypt, cannot be ruled out. While UK-Rwanda-style deals are and will remain a no-go for member states, an expanded STC concept could enable the EU to pursue its priority of implementing returns to countries of transit, a notable shift compared to current practices. The push for cooperation with Tunisia shows that this would come at the expense of human rights.

The negotiations with the European Parliament offer the opportunity to avoid this expansion and maintain stronger safeguards. However, considering the wider systemic weaknesses of the envisaged reforms – especially the loose solidarity obligations within member states – the EU will likely continue its burden-shifting efforts in the future regardless, whether through the STC concept or other types of arrangements with third countries.


Andreina De Leo is a PhD researcher at Maastricht University within the LIMES project and a former Junior Policy Analyst in the European Migration and Diversity programme at the European Policy Centre.

The LIMES project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 847596.

The support the European Policy Centre receives for its ongoing operations, or specifically for its publications, does not constitute an endorsement of their contents, which reflect the views of the authors only. Supporters and partners cannot be held responsible for any use that may be made of the information contained therein.