Introduction

In the face of increased arrivals across the Central and Eastern Mediterranean since early 2023, political pressure exercised by Italy and the reinvigorated conflict in the Middle East, the Spanish Presidency circulated a discussion paper, ahead of the Justice and Home Affairs Council meeting on 19 October 2023, proposing a shift in the external dimension of migration from “a reactive to a preventive model”. The document does not appear to propose anything original that goes beyond existing measures. Rather, it calls on member States to double down on current policies and practices, including arrangements and agreements with third countries. The prioritization of the external dimension is strongly supported by member States along the Eastern Mediterranean route–particularly Greece and Cyprus. As the main recipients of mixed migratory movements, both countries have taken the position that irregular mixed migration should be prevented, and this can only be done in partnership with countries of transit and/or origin. Are such partnerships an effective way forward for EU migration management? The experience from the Eastern Mediterranean has shown that, in their current format, their effectiveness is patchy at best. It is a practical and essential approach that often falls short in terms of implementation. Responsibility for this can be attributed to the countries of origin and transit and their positions in the regional migratory movements, the foreign policy ‘weight’ of the main EU destination countries in the Eastern Mediterranean but also on the EU foreign policy on migration. The latter continues to be oriented towards a preventative model and has, since 2016, sought flexible informal arrangements that serve a performative role, but whose implementation and impact fluctuates depending on circumstances.

Who moves along the Eastern Mediterranean route?

The Eastern Mediterranean route refers to the passage from Türkiye to Greece, Cyprus[1] and to a lesser extent Bulgaria, with the latter being a relatively recent addition as a country of destination and transit. Prior to 2014, the main countries of origin along the route were Afghanistan, Palestine, Iraq, Pakistan and Bangladesh[2]. In 2015–2018, Syrians constituted the largest nationality in irregular entries across both land and sea borders, followed by Afghans and Iraqis. Restrictions imposed due to the pandemic reduced cross border mobility, but irregular migration has been gradually returning to pre-2015 levels since 2022. According to FRONTEX, around 42 800 irregular border crossings were detected on the Eastern Mediterranean route in 2022. The majority were made by Syrians, Afghans and Nigerians. One noticeable shift in the period since 2020 has been a diversification of nationalities. Data from 2023 (until 31 July) show that the most common countries of origin for migrants arriving in Greece by sea from Türkiye were Afghanistan, Somalia and Syria, but also Palestine[3], Eritrea, the Democratic Republic of the Congo, Sudan and Yemen. Most of these nationalities receive international protection in significant percentages in Greece and in most EU member States. In 2022, most of the migrants awarded international protection in Greece originated from Eritrea, Palestine, Yemen, Sudan, Afghanistan, Ethiopia, Türkiye, Iraq, Syria and Iran. The overwhelming number entered from Türkiye. This gives a clear indication of the fundamental issue with the route: namely, that many of the countries of origin are deemed either entirely or partially unsafe for individuals or specific groups and categories of people on the move (e.g., Hazaras from Afghanistan). To these, we should also add those migrants from Pakistan and Bangladesh, who receive subsidiary protection at a reduced rate, while many are also rejected by the asylum process but nonetheless remain in the territory – often undocumented. The particularities of the route render it distinct from the Central Mediterranean. Unlike the Central Mediterranean, where colonial histories and trade ties have allowed for the growth and maintenance of relationships with multiple countries along the route including countries of origin, the Eastern route offers limited possibilities, with Türkiye being the main partner when it comes to migration management.

Tracing the external dimension on the Eastern Mediterranean route

To date, the external dimension of migration has been based largely on diplomacy and the utilization of soft power tools, ranging from legal migration options to humanitarian aid, development assistance and—potentially, at EU level—visa facilitation. The initiatives taken include a combination of regional processes, mobility partnerships and readmission agreements (EURA). This renders countries like Greece and Cyprus (along with most other EU member states) reliant on the EU as a collective body to push for and negotiate EU-wide arrangements and agreements, while the individual countries can in some cases offer complementary bilateral arrangements to specific countries of interest. Thus, third countries can receive assistance ranging from training to development, humanitarian aid, and—potentially–legal migration options such as student visa schemes, labour schemes, Talent Partnerships, and visa facilitation—though only after a lengthy process, in the case of the latter. Partnerships remain centred on achieving, first, a reduction in the number of irregular arrivals and secondly returns. This is to say that their aim is mainly preventative. Conditionality is attached; thus, the third country must do what is required to secure the EU’s cooperation on financial assistance, training, and mobility opportunities. In contrast, almost all of the incentives are optional for the member States; this means that, apart from the immediately interested party (e.g., Italy in the case of Tunisia), the other member States can opt in or out of most of the arrangements. Significant effort has gone into developing partnerships along the southern and western Mediterranean route, largely due to its proximity with the external borders of Spain and Italy. It is through the foreign policy in the Central Mediterranean also that migration from countries like Eritrea, Nigeria or even Syria is addressed.  On the Eastern flank, the focus has been on Türkiye (and recently also Iraq), along with Pakistan, Bangladesh and Afghanistan for both formal agreements and informal arrangements[4]. Since 2016, the European Commission has negotiated six legally non-binding arrangements for returns. Of those, the arrangements with Afghanistan and Bangladesh are relevant to the Eastern route. To this we should add the EU-Turkey Statement of 2016, which was meant to complement the existing EURA from 2014. Only two are public—the arrangements with Afghanistan and with Türkiye—, which highlights one of the biggest issues with informal arrangements: the lack of accountability.

Lessons from Afghanistan and Türkiye

Despite the different countries of origin along the route, there are some common themes that emerge from the various arrangements sought by the EU; these are clear in the arrangements with Türkiye and Afghanistan, the only two deals whose texts have been made public. A common thread in both is the focus first on returns and second on capacity-building to reduce departures. Fragile deals In 2015, Afghans constituted the second largest group of migrants and asylum seekers in the EU and one of the main groups of unreturnable migrant populations. The Joint Way Forward with Afghanistan (JWF)[5] set as a target around 80,000 annual returns to the country in exchange for capacity-building projects and more aid. Despite the deteriorating situation in Afghanistan, the EU concluded its Joint Declaration on Migration Cooperation (JDMC) with the country just months before the Taliban’s return to power. Made with a highly unstable political partner, the arrangement focused on preventing irregular migration, countering migrant smuggling and trafficking in human beings, and facilitating the return process for Afghan nationals who would not qualify for international protection. However, no returns have taken place since the Taliban take-over in August 2021, and none are unlikely to do so in the near future. Even before, however, the return rate was low. In the period 2014–2018, approximately 30,000 return decisions were issued annually for Afghans, with roughly 3,924 annual returns taking place across the EU. Between 13 September 2016 and 30 March 2020, 1,844 Afghan nationals were forcibly returned from EU Member States by charters organised by the European Border and Coast Guard Agency (Frontex). Individual member states—Greece among them—have also undertaken returns and bilateral arrangements. According to EUROSTAT figures, Greece has zero assisted forced returns in 2021–2023 (1st quarter), with the total number across EU27 for the 1st quarter of 2023 standing at 410 persons, with over 50% of those coming from Croatia. This shows the serious problems encountered when attempting to move forward with arrangements—even informal ones—with countries that are politically unstable, and economically entirely reliant on international assistance. The EU-Turkey[6] Statement of March 2016 was also a reaction to the ‘crisis’ of 2015, when some 885,000 people arrived in Greece, intended to curb arrivals. The EU-Turkey Statement was an informal arrangement that set the stage for the management of refugees and irregular migrants between Türkiye, Greece, and the EU. The EU had signed a formal EURA with Türkiye in 2014 (with provisional application of the third-country-national clause post-2016), and Greece had a bilateral readmission arrangement in place since 2002. Neither was deemed sufficient to facilitate returns at the required level. Neither was considered effective in deterring large-scale movement to Greece. The Statement is an example of an informal arrangement that seeks to complement existing legally-binding agreements while also serving a preventative role. Designed along the traditional conditionality format of ‘more for more’, it sought to leverage visa liberalisation and financial assistance in exchange for border controls and refugee hosting. The deal has been often called a success in terms of reducing the numbers of arrivals in Greece, and it did reduce the numbers for a while. However, attributing reduction over time to the Statement is problematic; it fails to acknowledge the broader context, namely the closure of the Western Balkan route, which was the main transit point for people on the move; or the role conditions in Greece played as a barrier to mobility, and particularly the prospect of remaining for months—if not years—in the hotspots awaiting an asylum application decision. Beyond the partial reduction in arrivals, the deal aimed at returns, and on this the data paint a gloomy picture. Only 2,140 people have been returned from Greece to Türkiye under the deal. This is partly because, in the early period (until 2018), Greek courts found that in many cases Türkiye was not a safe country to send people back to. In addition, Türkiye was reluctant and eventually unwilling (post-2019) to accept returnees, despite Greece’s repeated requests for returns. In parallel, Türkiye remains unwilling to implement the EURA as regards the third-country-nationals clause “until it is exempted from the Schengen short-stay visa obligation. In addition, Türkiye does not implement the agreement towards Cyprus concerning the prevention of irregular departures.” The issue of returns suggests different perceptions and priorities between the EU (and Greece) and Türkiye that will continue to impact existing arrangements as well as any future deals. In Türkiye, as the Syrian refugee population remains large and the economic situation continues to deteriorate amidst political instability, the temporary usefulness of the deal is reduced. The deal served a political and transactional purpose for Türkiye, while producing an imbalanced partnership with the EU and negatively impacting both migrants and the EU’s normative role in the region. In fact, Türkiye’s unilateral suspension of the bilateral agreement with Greece and refusal to accept returnees since 2020 was compounded by its “instrumentalisation” of migration at the land border in Evros in February 2020. Aware of the importance of migration for the EU, it could and still does leverage its position to request additional gains in other areas of cooperation. Absence of burden-sharing The majority of countries of transit and origin along the Eastern Mediterranean route are themselves hosting countries for refugees. Türkiye currently hosts 3.4 million Syrians, while Iran remains the largest host country for Afghans of different statuses (registered, undocumented), with the UNHCR registering 3.4 million refugees being hosted in the country at the end of 2022. Pakistan currently hosts 1,333,749 Afghans and has announced recently it will begin large scale deportations to Afghanistan for those without documents. All this, on top of the 960,000 Rohingya refugees displaced in Bangladesh and broader irregular forced movements of different groups in the region. In other words, at least three of the countries on which the EU and member States had focused for returns—Afghanistan, Pakistan and Bangladesh—and would like to or have established formal and/or informal partnerships are already shouldering a disproportionally high share of the responsibility when it comes to hosting refugees. And yet burden-sharing mechanisms are almost never envisioned in the partnerships. In the EU-Turkey Statement, a rare exception in this, the commitment for 40,000 places for Syrians resettled to the EU was a positive but limited step. Capacity-building measures and financial assistance do not compensate for the reduction in global resettlement figures or for Europe’s hesitancy, when it comes to establishing an obligatory European resettlement programme for all member States. Using visa policy as either a carrot or a stick does not produce sufficient legal mobility to compensate for the restrictive measures most of the deals require.

Short-term gains or long-term prospects?

If the main aim of partnerships with third countries is preventative, through either ensuring returns and/or establishing border controls, then they are bound to underperform. Border controls can partially reduce numbers, but cannot reduce irregular migration as such; they have proven consistently ineffective as a deterrent to migrant journeys. Returns, too, have proven problematic for multiple reasons including the shortcomings of the internal dimension, third countries increasingly realising that dependencies are bidirectional and can be leveraged, the fact that many countries are not safe—for everyone, at least—, but also because transit countries often shoulder an unequal burden. Even where they work as planned, they still fall short of their envisaged aims. Bangladesh, a country that signed a EURA in 2016 and has been pressured into collaborating on returns, is a case in point. In 2022, a non-paper on a Strategic Approach to the Readmission Agreements and Arrangements issued by the Commission services notes in relation to Bangladesh that: “It is challenging to measure the efficiency of readmission agreements and arrangements through quantitative indicators only […] As a way of illustrating this complexity, we can take the example of Bangladesh—with a return rate of 9% in 2019 and 5% in 2020, and an issuance rate of 49% and 31% respectively. In 2020, 9,370 persons were ordered to leave, and only 691 readmission requests were submitted by Member States. This indicates that, even with a 100% issuance rate indicating perfect cooperation, the return rate would still be relatively low”. What the current arrangements fail to account for is whether collaborating with the EU is in the best interest of the countries concerned in the long run. For most, the income received from remittances far outweighs any financial support received from the EU. This is where legal migration partnerships could come into play and complement informal arrangements, and this is where the emphasis needs to be on the Eastern route. Greece’s agreement with Bangladesh functions as one such example: designed exclusively for labour migration, it complements a broader EU cooperation while also offering tangible mobility options for Bangladeshi workers. The arrangement with Egypt and preliminary discussions with Pakistan are similar in this respect. More such agreements are needed and larger in scale in terms of quota included.

How to render partnerships effective?  

There is no doubt that partnerships are needed, and informal arrangements allow for flexibility, which is preferred by third countries. Nonetheless, it is this informality that also renders them fragile and potentially unsustainable. The lens through which these deals are seen from the EU perspective remains problematic. There remains a disproportionate focus on prevention through returns and boosting border controls, with less attention paid to the need for a more equitable sharing of the “burdens”. Factoring in the complexity of countries of origin and transit along the Eastern Mediterranean route, it is important for countries like Greece and Cyprus to push for more equitable arrangements between the EU and third countries, which could ensure a more sustainable implementation of deals. Burden-sharing mechanisms should be integrated particularly with countries that are already hosting refugees and forced migrants. As Greece restarts cooperation with Türkiye on migration, this is worth remembering. Any discussion relating to the implementation of the Statement needs to set realistic expectations vis-à-vis returns and the effectiveness of border controls as well as accounting for fluctuations in implementation stemming from Türkiye’s foreign policy responses to Greece and the EU as well as the broader geopolitical situation. In other words, the deal cannot guarantee that arrivals will remain low continuously, so Greece and the EU should aim for a long-lasting cooperation, but be prepared for a transactional and temporary one. This means adjusting responses on the EU side, too. A more inclusive approach is needed which factors in the long-term, as well as the short-term, impacts of partnerships. It is important to move beyond a ‘problem-solving’ approach, which only looks to reductions in numbers. Migration partnerships can achieve much more in terms of strengthening local institutions, facilitating resilience, and creating corridors of cooperation that move beyond the logic of deterrence. In this, legal mobility—the biggest carrot in the toolbox—remains the least exploited, also in the case of the Eastern route. Whether they relate to complementary bilateral arrangements for labour migration, EU-wide resettlement for those in need of international protection, student visas etc, the numbers need to increase, and more member States need to participate. Countries like Greece, a crucial part of the route, are taking the lead in developing legal migration pathways, even if it is for fewer numbers. By strengthening positive cooperation, they can also acquire a stronger voice in the shaping of partnerships, which is currently dominated by a more traditional externalisation model that has historically yielded fewer and less sustainable results. More needs to be done at the country level to develop comprehensive legal pathways for labour migration and people in need of international protection (in which the labour dimension could also be integral). This is an area where support is needed also by the European Commission and member States. Since the beginning, the external dimension has been strongly dominated and shaped by internal concerns and priorities, that are often grounded in misconceptions around migration and unrealistic expectations. Nowhere is this more evident than on the focus on returns. The language of EU migration management with third countries promotes a ‘win-win’ approach but the reality is far different. The external dimension is seen as a solution to the ‘migration problem’, suggesting a simplification of a complex issue, and a negative (rather than positive) approach in place.   Partnerships are instruments of foreign policy which means they are made up of compromises, imperfect solutions, promotion of norms and well-chosen value frames, and often allowing foreign policy priorities to also shape domestic agendas. Going forward, and particularly as regards the Eastern Mediterranean, a different, more complex approach is needed; one that begins with a mental shift at EU level, a willingness from member States to adapt and change domestic policies where needed (e.g., labour demands and supply), as well as developing and/or utilising bilateral, regional and international actors and instruments in establishing and maintaining cooperation with third countries.


[1] In 2019, Cyprus was the country with the highest number of asylum seekers in relation to its population, with most of the former coming from Syria, India, Bangladesh and Pakistan. Asylum seekers now make up over 5% of the population and are mainly nationals of Syria, Nigeria and the Democratic Republic of Congo. [2] In the case of Greece, Albanians remained highly prevalent in apprehension figures. For the percentages of the different nationalities, see Frontex, Risk Analysis 2016; Frontex, Annual Risk Analysis 2015. [3] This is prior to the eruption of the war between Israel and Hamas in October 2023. [4] When referring to legally binding deals, the term ‘agreement’ is utilised. When referring to informal deals, the term ‘arrangement’ is used, per European Court of Auditors [5] No EURA has been signed with Afghanistan, though bilateral agreements are in place with Member States and the Government of Afghanistan. The JWF does not supersede any bilateral agreements in place between Afghanistan and EU member States, allowing Afghanistan to engage both with the EU as a whole and with individual member States. [6] Since the official name of the deal is “EU-Turkey Statement”, it has been retained. When the reference is to the country, the official designation Türkiye is now used.

Asli Selin Okyay Yet another ‘migration crisis’ in the Central Mediterranean: Seeking European answers to a European challenge outside Europe