The EU’s new Return Regulation: rhetoric, reality and the limits of externalisation

On June 17, 2026, the European Parliament approved the new Return Regulation, intended to complement the Pact on Migration and Asylum that came into effect on June 7, 2026.

Returns are not a new ‘problem to be solved’. They have been at the forefront of the political agenda for two decades, and have long been part of the broader externalisation  agenda, i.e. the practice of shifting border control, asylum processing, and migration management outside a state’s own territory. The 2008 Return Directive (Directive 2008/115/EC) established basic minimum standards across EU member states, but in practice return orders were issued and then not implemented — for a range of reasons, including lack of cooperation of countries of origin as well as court interventions.

When the Commission proposed the new Return Regulation in March 2025, EU return rates of past years stood at around 20%. By 2025, the rate had risen to 28% — still the highest in the past 10 years, though the Commission itself notes that the number of effective returns remains too low.

Data from Eurostat indicates that in 2025 (Q1–Q4), 132,660 returns of third-country nationals (TCNs) took place across the EU, against a total of 492,580 return decisions issued that year.

Table 1: Top 5 EU Member State Returns, 2025

Top 5 EU Member State	Returns (voluntary & forced, Q1–Q4 2025)
Germany	29,295
France	14,935
Sweden	11,250
Cyprus	10,630
Poland	9,500

Source: Eurostat

It is worth noting that data on returns remains relatively weak in comparison to data on apprehensions. Most countries issue return orders even when they cannot be enforced (e.g. Greece) and there is no systematic data on how many non-enforceable return orders are withdrawn or suspended. The Commission estimates that 300,000 people cannot return from the EU every year. The new Regulation seeks to address this gap between orders issued and implementation.

The new Regulation expands the maximum detention period from 18 to 24 months, permits the detention of families with children, and imposes strict cooperation requirements on individuals,  with failure to comply resulting in reduced benefits or detention. Critically, there is no automatic suspensive effect on return decisions pending appeal, meaning individuals could be deported before a court rules on the risks they face. A further key element is the mandatory mutual recognition of return decisions across member states, replacing what had previously been a voluntary arrangement.

In line with the broader logic of the Pact on Migration and Asylum, the Returns Regulation turns to externalisation as a solution, proposing the establishment of ‘return hubs’ beyond EU territory — in third countries to which rejected asylum seekers and irregular migrants can be sent.

The idea is not new. In 2003, during a European Council meeting, the United Kingdom tabled a proposal for transit hubs in neighbouring countries where asylum applicants would be transferred pending deportation in the event of rejection. It was rejected. For the most part, however, past attempts have focused on offshore processing centres for asylum seekers. Australia had pioneered the approach through its Pacific Solution and offshore processing in Papua New Guinea and Nauru. The EU had attempted returns from the Greek islands to Türkiye under the EU-Turkey Statement of March 2016. Only 2,140 migrants were returned, with Türkiye suspending returns since March 2020. The UK-Rwanda scheme of 2023 sought the transfer for asylum processing of irregular arrivals to Rwanda. If an applicant received protection they would need to remain in Rwanda. The scheme was ruled unlawful by the UK Supreme Court in 2023 while in 2024 the scheme was abandoned by the new labour government. Finally, the Italy-Albania agreement of 2023 set up two detention centres in Albania for the processing of asylum applications under Italian jurisdiction. Following court rulings that blocked implementation, the Italian government revised the deal in March 2025, with the centres now being used for migrants whose application for asylum in Italy was rejected.

The above examples have been attempts by countries to reduce irregular arrivals through the setting up of deterrence mechanisms that would be deemed impossible until recently across the EU. What has made return hubs politically thinkable is a European landscape transformed by crisis, conflict, and the rise of the far right. The refugee crisis of 2015–2016, Russia’s invasion of Ukraine and the activation of the Temporary Protection Directive for millions of Ukrainians, continuous instability across the broader neighbourhood, and the rise of far-right parties and anti-immigrant rhetoric across the EU, all of this rendered the idea of return hubs not just thinkable but politically attractive.

In theory, return hubs would operate under bilateral agreements between EU member states and third countries, with migrants transferred regardless of any connection to the receiving country. The underlying logic, according to policymakers, is deterrence: if people know they risk being sent to a third country for deportation, they will not embark on an irregular journey. Those who do, and whose asylum claims are rejected, could — if they cannot be returned to their home country — be transferred to a hub where their stay would be indefinite, pending return.

Civil society and human rights organisations have repeatedly raised concerns: what happens to those who cannot be returned? Who bears responsibility for conditions in these centres? How will monitoring work, and who funds it? These are not abstract questions. History has shown a wide gap between what is agreed on paper and what unfolds on the ground — Tunisia being a prominent recent example.

One of the main reasons the return rate sits at 28% is difficulties in securing cooperation from third countries for the readmission of their nationals. Returns mean lost remittances; cooperation carries political costs; and in some cases, readmission agreements are filtered through the lens of postcolonial power dynamics.

It is difficult to see how return hubs alter any of this. They are far more likely to reinforce the perception that the EU is outsourcing its responsibilities to countries in the global south that already carry a disproportionate share of hosting migrants and refugees — while leaving the EU exposed to the instrumentalisation of migration by partners that are not always reliable, and that risk growing dependent on EU funding in exchange for acting as its gatekeepers.

Greece and returns

Greece has positioned itself in recent years as a strong advocate of deterrence policies and returns. On June 9, the Greek Parliament approved legislation to fast-track the deportation of rejected asylum seekers and allow their transfer to return hubs once bilateral agreements are reached with third countries. Greece is already in discussions with the Netherlands, Denmark, Germany, and Austria about setting up joint return and transit hubs. The preference for a joint initiative reflects Greece’s limited leverage with third countries: to attract a return hub agreement, a state needs to offer significant incentives that Greece, on its own, does not have.

The domestic rhetoric around return hubs has been forceful, and the speed of the parliamentary approval suggests a scale of arrivals and rejection levels that would make a dramatic increase in returns both necessary and urgent. The reality is rather different.

Greece’s returns record tells a different story from the political rhetoric. In 2024, 5,865 third-country nationals were returned; in 2025, 5,736 — figures that have remained broadly stable over the years. Of the 2025 returns, 2,464 were forced, 1,240 voluntary, and 2,032 implemented through IOM. As Refugee Support Aegean (RSA) has noted:

“Almost two out of three returns and deportations carried out in 2025 concern nationals of Albania (2,094) and Georgia (1,544) alone. Most returns were forced in the case of Albania, and most were voluntary returns implemented via IOM in the case of Georgia.”

These are nationalities that do not feature prominently in arrival statistics via Evros or the Aegean, and they originate from countries where existing cooperation frameworks already function.

The scale of irregular arrivals, while not insignificant, needs to be understood in context. According to official Greek data compiled by RSA, 52,180 people arrived irregularly in 2025, a 21% decrease from 2024, and far below the numbers recorded in 2015. In 2025, 55,383 initial asylum applications were lodged, with Afghanistan by far the main country of origin, followed by Syria and Somalia. Of those whose claims were assessed on their merits, the overwhelming majority were found to be in genuine need of protection. The overall recognition rate at first instance stood at 70.6%, with recognition rates over 90% for Afghans, Sudanese, Palestinians and Syrians.

In other words, the people arriving in Greece at the moment are not having their asylum claims rejected, nor can they lawfully be removed. At the same time., the people being returned are already being returned through existing mechanisms.

Return hubs would not change this equation nor are they the right response to a situation that even the numbers do notsupport. However, two variables should be taken into consideration. Firstly, for certain nationalities, like the Afghans, things are potentially changing at the European level, evident also by the recent meeting of EU officials and representatives from 15 member states with the Taliban to discuss deportations; initially at least for people with criminal convictions.

In early 2026, the overall acceptance rate for Afghan asylum seekers dropped to 69% in Greece, with Afghan men facing a noticeable increase in asylum rejections, as documented by Legal Centre Lesvos. In parallel, the Greek Ministry of Migration and Asylum reopened many Afghan refugee cases to reassess their protection status and examine the possibility of revocation. The emphasis on Afghans and Syrians is both a product of a broader exclusionary policy adopted in Greece but also a response to the expected increase in readmissions of recognised refugees from countries like Germany. Greece received 9,179 readmission requests in 2025 (vs. 2,468 in 2024), with Germany accounting for a 555% year-on-year increase. This is the second shift partly behind the urgency for ‘solutions’ around returns.

With a lower acceptance rate and prospects of returnees from other EU Member States, Greece’s priority will be on the implementation of returns that in turn require a return mechanism in place that enables large scale deportations but also functions as a deterrent. This renders the idea of return ‘hubs’ potentially “attractive” from a policy perspective.

Research shows that the prospect of detention or other punitive measures does not reduce the desire to migrate — but it does produce rerouting and more dangerous journeys, with detention-based deterrence generating deeper human costs without meaningfully reducing irregular arrivals. Deterrence also does not reduce smuggling operations. Rather, the higher the demand, the higher the profit.

For countries like Greece, where migration has remained predominantly in the field of domestic rather than foreign policy, the absence of legal pathways is a crucial obstacle in easing pressure on the external borders and enabling  safer, more orderly migration along the Eastern Mediterranean route. Similarly, in the broader European context, no deterrence policy will work if the strengthening of legal pathways and support for asylum, reception and integration systems is not prioritised. Returns remain a necessary part of a balanced migration system, but they are more effective when voluntary, humane, grounded in reintegration assistance carried out in partnership with civil society. This means balancing genuine labour market needs against a more sustainable migration framework, without resorting to measures that evidence shows do not reduce irregular migration, but do produce more dangerous journeys and deeper human costs.

---

The views and opinions expressed in this article are those of the author and do not necessarily reflect the official position of the Interdisciplinary Migration Research Unit (IMRU) or the National Centre for Social Research (EKKE).

---

---