The Multi-level Governance of Asylum in Italy: Understanding Eritreans’ Secondary Movements in Search of Relocation within Italy

Abstract

The migration policy, securitization and governance of the asylum landscape has greatly changed in Italy over the past five years. The implementation of the hotspot approach and the European Union relocation programme has characterized two of these significant changes. This article examines how recent (2016–17) Eritrean arrivals in Italy experienced and responded to the asylum system. The analysis reflects, first, on the multi-level governance of the interplay between the institutional level of the European Union, national policy and local municipal policies within Italy. Second, this article shows how Eritrean asylum seekers have been refused access to the relocation programme by local authorities in Italy and how Eritreans react to this bureaucracy by engaging in secondary movements within Italy. The results provide an analytical critique regarding the governance shortcomings of the relocation programme in Italy that is important for consideration in future solidarity and responsibility-sharing initiatives within the European Union.

Introduction

In recognition of the discrepancies between Member States, the European Commission announced the relocation programme in 2015 as a solidarity mechanism to relocate asylum seekers away from frontline Member States that were bearing the burden of unprecedented flows in 2015. Announced in May 2015, the programme initially intended to transfer 40,000 people from Greece and Italy to other European Union Member States and this goal was revised in September 2015 to 120,000 people. To be eligible for relocation, the individual must be from a nationality that has a 75 per cent or higher average recognition rate for international protection across the European Union. Since the inception of the relocation programme, this has included Eritreans, who have been the primary nationality eligible for relocation in Italy. Eritreans therefore make a natural choice for a case analysis of the relocation programme in Italy.

This article examines Eritrean asylum seekers’ secondary movements within Italy. Secondary movements can occur within a single nation state or across nation-state boundaries in the European Union. Most research on secondary movements focuses on movements across and between Member States, as this creates a condition of irregularity (Brekke and Brochmann 2015; Fontanari 2016; Schapendonk 2018), either when within the asylum-seeking process or when granted a temporary protection visa. Focusing on the refugee context, Zimmerman (2009) has termed these movements as ‘seeking refuge’ and ‘seeking asylum’. Less attention has been placed on secondary movements within one state. Research on asylum-dispersal programmes in countries such as the United Kingdom is of relevance here (such as Bloch and Schuster 2005); however, these movements are not necessarily referred to as internal secondary movement. It is evident that further research is necessary to understand these internal movements, and this article seeks to contribute to this research gap.

The Italian reception regime has been characterized by a state of humanitarian emergency since 2014 when arrivals reached an unprecedented high. Arrivals have continued to be high since 2014, which has led to a continuing state of humanitarian emergency within the reception regime. The result has been a system in disarray with not only significant variation in standards and asylum seekers’ experiences, but also in access to relocation procedures. Since the inception of the relocation programme, the European Commission has been calling on Italy to improve information dissemination of the programme to local authorities in an effort to increase registrations and access to the programme (European Commission 2017).

This article shows how Eritrean asylum seekers have been refused access to the relocation programme by local authorities in Italy and resultantly engaged in secondary movements in order to find access to the relocation programme. The findings illustrate the differences in reception, legal access and relocation procedures within Italy and analyse how the discrepancies within the Italy reception regime impact asylum seekers’ strategies in terms of secondary movements within Italy. The analysis reflects first on the multi-level governance of the interplay between the institutional level of the European Union, national policy and local municipal policies within Italy and, second, illustrates how Eritreans react to the bureaucracy by engaging in secondary movements within Italy.

Eritreans were one of the largest asylum-seeking groups in Italy in 2015 and 2016. Eritreans have been fleeing their country since 2000 and, as a former colony of Italy, Italy has an established Eritrean diaspora, particularly in Milan and Bologna. This established group is, however, distinct from the new arrivals and contestation occurs between the two groups (Martignoni 2015). The new cohort of Eritrean arrivals to Italy do not intend to stay in Italy as their destination and seek onwards migration movements within the European Union (Belloni 2015; Brekke and Brochmann 2015). The main reasons for wanting to move beyond Italy include a lack of economic opportunities and reunification with family and friends in other countries.

This article is based on interviews conducted with 35 Eritrean migrants in Italy in 2017 in Milan and Rome and 18 key stakeholder interviews conducted with local authorities, the European Asylum Support Office (EASO) and other researchers. The article is divided into the four following sections: first, a brief overview of the Italian reception regime; second, a presentation of the methodology and research design; third, the results are discussed; and, finally, a discussion and conclusion.

Italian Reception Regime: From Transit to Destination and Relocation

The policy and migration securitization landscape has significantly changed for arrivals in Italy over the past five years. In 2014, Italy experienced an over 400 per cent increase in arrivals from 2013 and a just under 300 per cent increase in arrivals from the previous peak in 2011. It is thus understandable that 2014 was characterized as a time of emergency and crisis in Italy. The priority of Italian officials in 2014 was on medical needs, food and shelter, meaning that fingerprinting at arrival and registering asylum claims was deprioritized. Refugees and other migrants were therefore able to easily move beyond Italy to claim asylum in other European countries without the fear of return to Italy under the Dublin III Convention. As a result, and as shown in Table 1, the number of asylum claims filed in Italy in 2014 was relatively low compared to the number of arrivals at only 38 per cent. (Note: these percentages are mainly for comparative purposes, as they represent an imprecise calculation since individuals may arrive and apply for asylum in different years.)

Table 1

Arrivals and Asylum Applications in Italy per Year

	2014	2015	2016	2017
Arrivals to Italy	170,100	153,842	181,436	119,369
Asylum applications in Italy	64,625	83,540	122,960	128,850
Percentage of arrivals compared to applications	38	54	68	108
Eritrean arrivals	34,329	39,385	20,718	7,052
Eritrean asylum applications	480	695	7,395	6,370
Percentage of arrivals compared to applications	1	2	36	90

Source: Author’s own compilation based on Eurostat and UNHCR data.

This is especially the case with Eritreans wherein only 1 per cent of asylum applications were filed by Eritreans as compared to the number of arrivals. In 2014, the countries in the European Union with the highest number of Eritrean asylum claims were Germany (13,255), Sweden (11,530) and Switzerland (6,920) (Eurostat 2017). The implementation of the hotspots and increased border controls to prevent onwards migration in Northern Italy significantly changed the number of asylum seekers in Italy as compared to arrivals. In 2017, there were more asylum applications made in Italy than arrivals (108 per cent) and for Eritreans this was 90 per cent.

In 2014, an infrastructure was developed across Italy to support the role of being a ‘transit’ country. In Rome, non-governmental organizations (NGOs) provided ‘transit’ backpacks to migrants with the provisions that they would need for their onwards journey. In Milan, the Hub Sammartini (referred to as ‘the Hub’) was opened as an overnight shelter north of the train station to prevent people from sleeping in the station and on the streets. Upon arrival at the Hub, refugees and other migrants were asked what they were planning to do and, if they wanted to migrate onwards, they were given shelter and food until they chose to leave. At that time, Italy’s northern borders were open as per the Schengen agreement and migrants could move onwards.

The steep rise in asylum seekers in northern European countries combined with the lack of fingerprinting led to tensions between the European Union and Italy. Under the Dublin Convention, Italy is formally required to document irregular arrivals and accept back such arrivals if they migrate onwards within the European Union. In May 2015, in the European Agenda on Migration, the ‘hotspot’ approach was initiated. In September of the same year, the first hotspot opened in Lampedusa, with the initial hotspots opening in southern Italy in late 2015 (Tazzioli 2018). The hotspots were administered by local authorities and work with a consortium of European Union agencies (EASO, EU Border Agency (Frontex), EU Police Cooperation Agency (Europol) and EU Judicial Cooperation Agency (Eurojust)) with the goal to ‘help [frontline Member States] to fulfil their obligations under EU law and swiftly identify, register and fingerprint incoming migrants’ (European Commission 2015: 1).

Tazzioli (2018) has argued that the hotspot approach has been a containment measure to control unwanted migration movements beyond frontline Member States and to ensure identification for the swift return of unwanted migration movements back to Italy. According to a report from the European Commission, since the establishment of hotspots in Italy, the proportion of new arrivals being fingerprinted increased from 36 per cent in September 2015 to 87 per cent in January 2016. Hotspots have been criticized for breaching fundamental rights of migrant arrivals by sorting arrivals seeking protection based on nationality, not individual circumstance, and immediately removing and deporting arrivals from nationalities not deemed eligible for protection (de Vries et al. 2016).

A second component to the European Agenda on Migration was the implementation of the relocation programme. Eritreans have been the primary nationality within the Italian relocation programme. At arrival, Eritreans are sorted by the hotspots into a protection and relocation stream. The intention is that, once placed in accommodation, like all other asylum seekers in Italy, Eritreans are registered at the local questura (police headquarters in charge of passport control and immigration). At questura, a form is completed for the permission to stay in Italy, called a C3 form. Within this form, there is a box to check for relocation. When the relocation box is selected, the case is then sent to the Dublin office, which is also in charge of relocation. The Dublin office assesses the relocation request and, upon approval, starts the process for relocation. This includes matching the asylum seeker to a receiving state and making all necessary arrangements between the questura, Dublin office and the receiving state.

Initially, it was intended that all individuals eligible for relocation were transferred to Villa Sikania near Agrigento to wait for the relocation process (AIDA 2015). However, due to large numbers, Villa Sikania quickly became full and individuals for relocation were then supposed to be allocated to regional relocation hubs (roughly nine in total). Hubs were centres that had cultural mediators and staff knowledgeable of the relocation process that could register individuals. Due to high numbers, the hubs were also quickly filled and Eritreans desiring relocation were moved into the national distribution mechanism for asylum procedure. EASO established a mobile unit with the purpose of travelling to relocation applicants that were not located in hubs, although the mobile unit was also under strain due to the large caseload.

Italy has a voluntary national distribution mechanism that municipalities can opt in or out of. In 2017, 3,200 or 40 per cent of municipalities in Italy were involved in the reception system (ANCI et al. 2017). First reception varies significantly in Italy, with a default two-tier system. Asylum seekers are supposed to be sent to a first reception centre from the hotspot; however, at the end of 2017, only 10,738 places were available across Italy (AIDA 2018). These centres include the old Governmental Centres for Accommodation of asylum seekers (Centri di Accoglienza per Richiedenti Asilo e rifugiati (CARA)) and accommodation centres (Centri di Accoglienza (CDA)).

In response to the critical influx in 2015, a parallel system of ‘extraordinary reception systems’, termed CAS (Centri Accoglienza Straordinaria), was established. The standards for the CAS were far lower than for the normal reception system, which was aimed to be a rapid, interim solution (D’Anegelo 2018). However, in effect, most asylum seekers in Italy have ended up being hosted within the CAS system at 81 per cent of the asylum-seeking population at the end of 2017 (AIDA 2018). The CAS now functions as both a first and a second reception system. According to the National Asylum Programme, second reception is supposed to be within the System for the Protection of Asylum Seekers and Refugees (SPRAR) centres (in Italian: Sistema di Protezione per Richiedenti Asilo e Rifugiati). These are integrated centres with boarding, legal assistance and social advice that are nationally monitored and implemented at the local level (D’Anegelo 2018). The SPRAR capacity, however, is much smaller than the annual flows and, in 2017, a total of 34,039 people were hosted in SPRARs (ANCI et al. 2017).

Reports have illustrated that conditions within the CAS structures vary widely (Cittadinanzattiva 2016; D’Anegelo 2018), with some lacking basic health and safety standards. Melchionda (2016) reported that, for many CAS operations, this was viewed as a business opportunity. Reception across Italy thus varies considerably. First, this is due to an integrated approach of the local municipalities and central government wherein local municipalities manage and contract reception centres. Different municipalities therefore have different approaches. Second, due to the lack of regulation of CAS facilities, there is significant variation in the implementation and management of the facilities. Migrants can have completely different experiences in centres located close together due to the autonomy of the CAS system. The cumulative result of this system is that individuals have highly varying experiences in reception across Italy, arguably creating a class of winners and losers.

The Italian reception regime is thus influenced by policies at the European Union, national and local levels, creating, first, a complex system of multi-level governance and, second, a stressed system still operating in an emergency approach that has not been able to adjust to the high number of arrivals. Based on the above complexities of the high variations in the CAS system and Eritreans for relocation being unexpectedly placed into the national distribution programme, it is unsurprising that the European Commission regularly reported in their relocation and resettlement reports to parliament that a central priority was for Italy to increase identification and registration of eligible applicants. The 15th relocation and resettlement report specifically identifies that ‘Italy supported by EASO must increase awareness about the relocation scheme among the local authorities and in the reception centres’ (European Commission 2017: 5), stressing that a lack of knowledge of the programme is a critical hindrance to implementation. It is evident that structural constraints within the Italian reception system presented a barrier for Eritreans in accessing the relocation programme. The rest of this article will examine the experiences of Eritrean migrants and how they navigated the Italian reception system to receive relocation.

Methodology

This article examines the experiences of Eritrean migrants that arrived in Italy between 2016 and 2017. The article draws from 35 interviews conducted with Eritreans between January and June 2017 and 18 key stakeholders. The interviews with Eritreans were conducted in Rome (11) and Milan (24). An Eritrean research assistant was recruited in Rome that arranged interviews with respondents via assistance from an NGO and all interviews were carried out in a community centre. The interviews with key stakeholders included researchers with direct experience on conducting research with Eritreans in Italy or the Italian reception system, and representatives of NGOs, local governments and EASO.

In Milan, permission was received from the city to conduct the interviews. The Comune di Milano (city of Milan) has a different structure than other parts of Italy wherein the comune has its own migration and refugee team that is separate from the Prefettura and national-level administration. The comune therefore manages reception of asylum seekers arriving in Milan outside of the national dispersal programme. The comune provided permission and support for this research and coordinated an introduction to two NGOs (Fondazione L'Albero della Vita and Fondazione Progetto Arca onlus) that are contracted by the comune to provide reception assistance. Both NGOs provided support to the project by inviting the interviewer to their centres and informing their beneficiaries of the interviews. Separately to this, an independent Tigrinya translator was hired for the interviews. Finally, during the fieldwork in Milan, policies and practices were changing within the city, and some Eritreans were also approached on the street for interviews.

At the end of each interview, respondents were asked whether it would be all right for the research team to contact them again in one month’s time. If they responded ‘yes’ (in all cases except one), their phone number was requested. After one month, the respective translator (separate for Milan and Rome) phoned the respondents to see how they were doing and ask about the progress on their case. For the interview respondents in Rome, these follow-up calls were conducted at one point in time in May 2017. For the interview respondents in Milan, they were conducted in May, June and July 2017 and, between September 2017 and January 2018, four additional follow-up interviews were conducted with the Milan interview respondents. Through these follow-ups, it was confirmed that 15 of the 34 respondents had successfully been relocated.

All interviews began with a detailed verbal informed consent process. The interview methodology used a lifecycle approach, beginning with leaving the country of origin, experiences in countries of first reception, decision-making processes, coming to Italy, initial arrival, current situation and future plans. The interviews generally lasted from 30 to 60 minutes. All interviews were transcribed and interviews conducted via translation were simultaneously translated and transcribed directly from the respondent. The analysis was conducted by first reviewing all transcripts for content understanding and to identify different experiences of secondary movements. Second, themes were developed and the interviews were then coded using qualitative software Nvivo. Third, additional themes were added as they arose in the transcripts. Finally, different cases were identified that could represent different scenarios for the analysis.

Almost all respondents were living in a reception centre at the time of interview, with the exception of a few Eritreans in Milan who were currently living on the street. Nearly half of the respondents were female (47 per cent) and the average age was 25 years old. Respondents were from different parts of Eritrea, with a larger percentage originating from the southern region. Respondents had various durations of their migration journeys ranging from as long as six years to only a few months. All respondents had arrived in Italy recently, ranging from as short as six weeks ago to as long as 18 months.

Eritreans’ Secondary Movements within Italy: The Search for Relocation

Table 2 provides an overview of when respondents first learned of the relocation programme, which will be discussed throughout the ‘Results’ section. It is important to stress that most respondents (57 per cent) first learned about the programme at disembarkation, which is essentially the intention of the current governance approach. Some learned of the relocation programme (20 per cent) prior to arrival in Italy via their transnational networks. The remaining respondents learned of relocation either in first reception or upon arrival in Milan (23 per cent).

Table 2

When Eritrean Respondents First Heard about the Relocation Programme

	Number of respondents	Percentage
Prior to arrival	7	20
Disembarkation	20	57
First reception	5	14
Milan	3	9
Total	35	100

Source: Author’s own calculations.

The results are presented in the following five sections: ‘Disembarkation and Fingerprinting’; ‘First Reception and Accessing Relocation’; ‘Drivers of Secondary Movements’; Changing the Initial Decision to Relocation’; and ‘Secondary Movers’ Reception in Milan’.

Disembarkation and Fingerprinting

All the respondents in this study arrived in Italy from Libya on boats that were rescued at sea. At arrival, disembarkation was thus organized (to a degree) as the arrivals were expected. In most cases, respondents reported that a Tigrinya-speaking cultural mediator was present upon arrival, whereas, in other cases, there was only a cultural mediator who spoke English and, in a few cases, no cultural mediators were present.

Immediately upon disembarkation, fingerprints were taken. Eritreans know the consequences of having their fingerprints taken and several respondents reported that other Eritreans they arrived with fled as soon as they disembarked to avoid fingerprinting. Cultural mediators communicated in Tigrinya upon arrival the process of the relocation programme and to accept the fingerprinting. Sesuna stated:

When we arrived there was a cultural mediator and they told us about relocation and that we are going to apply, but first we have to take fingerprints. They also told us that giving fingerprints is just for safety and that it will not affect anything in the process or once you get to other countries.

For most respondents, this was the first time that they learned about the relocation programme. Respondents reported feeling happy that there was a new legal option for onwards migration. The option for legal onwards migration also led deserters to come back and join the process for relocation. Yonas explained:

Most of the people who came with us don’t like the fingerprint record to be taken in Italy, so they run away as soon as we arrived. They told us [cultural mediators] that we do not have to run and that we can apply for relocation and go to other European countries legally. Some of them had run before they heard this. We called some of them and they were able to return.

This is a central finding that the relocation programme led to changes in respondents’ decision-making for irregular onwards migration from Italy. This will be discussed further later in the article and in the conclusion.

Respondents also stated that they were told by the cultural mediators that the process would take three to six months and, after this time, they would be relocated to another country. This created an expectation from arrival for the anticipated relocation process.

It is important to stress that none of the respondents in this study reported refusing or challenging to be fingerprinted. The cultural mediators at arrival stressed the importance of the fingerprinting and this seems to have contributed to resulting in a smooth process. The biggest complaint was the long line and wait for fingerprinting, which sometimes occurred without food being first provided.

First Reception and Accessing Relocation

After disembarkation and fingerprinting, respondents were transferred to reception facilities across Italy. It is unclear what type of facilities respondents were transferred to, as often respondents were not sure of the town/ village name, yet alone what type of structure the facility was within the Italian reception system.

Respondents had a wide spectrum of experiences upon arrival at the reception centres. By the time of arrival to the first reception centre, most respondents were aware of relocation and had made the decision to apply for this process. However, this was not always possible at the local questura. As discussed earlier in the article, applicants for relocation were supposed to be sent to the ‘relocation hubs’ that understood the process and had translators to assist with the relocation application. Respondents who were thus sent to smaller centres and not relocation hubs reported that they were unable to apply for relocation in these centres. In these cases, respondents reported that no cultural mediator was available. Some respondents tried to communicate with the little English they had, via another asylum speaker with some Arabic, or through whatever mechanisms they could use. Mariam explained:

We told them that we have siblings in other parts of Europe. The guy who is writing our relocation form told us OK, but instead he wrote it as we don’t have any siblings or any family. He wanted us to have an Italy paper. We refused to sign the paper because the information was all wrong. The guy told us that there is no relocation program here in [town a]. We waited in [town a] for 3 months with no purpose. Finally, after waiting for 3 months they wanted to give us the permit to stay in Italy. We contacted the legal adviser. He promised us to fix everything. So, we trusted him but he failed us. They told us to go to another camp and ask for relocation. After that we asked for a paper that stated the farewell from this camp and to be welcomed to another camp. We have been denied that paper. They insisted that we should go by our own.

It was unclear to the respondents why they were unable to access relocation in the centres for first reception. At the time of the interviews in Rome and Milan, it was also a learning process for the centres assisting the respondents in understanding their different complex cases. In both Rome and Milan (both of which are relocation hubs), the relocation process was known and understood, so it was a learning process for the cities and centre staff to understand that other parts of Italy were not properly administering the relocation procedures.

In other cases, respondents were less fortunate in that they thought they were applying for relocation, but this was not the case, as explained by Fatimah:

I asked them [about relocation]. They replied by saying there is no such relocation program here. People in the camp [other migrants] don’t really know that there is no relocation there. They gave me a permit to live because I stayed for 6 months there. I guess if you have that permit you can’t apply for relocation. When I arrived to Milan, I told them I was fooled and don’t know about the system.

As shown by Fatimah, there was a lot of confusion amongst respondents regarding the rules for applying for relocation. To qualify for relocation, one must actually first apply as an asylum seeker in Italy. As mentioned previously, in Italy, the form that is completed for asylum seekers at the questura is the same form as for asylum and relocation (C3 form) and there is simply a separate box that must be checked for the relocation procedure. When this box is not checked, the relocation application is not initiated, although, according to an EASO representative interviewed, all Eritrean asylum applications are sent to the Dublin unit to be assessed for relocation. This may have been a new procedure, however, implemented to increase registrations of Eritrean asylum seekers in the relocation programme. When questura officials are unaware of the procedure, the box is not checked and respondents fear that they will be forced to stay in Italy. Several respondents had the impression that, if they received a permit to stay in Italy, they were not eligible for relocation, which caused high distress and confusion.

Senait reported that the centre she was at did not have the relocation programme and that, one day, EASO came to the camp and she was able to speak with them. EASO has a mobile unit that comes to process relocation claims for individuals not residing in relocation hubs. She interpreted her situation as ‘luck’ that EASO came and she was able to process her relocation application.

In some cases, the staff at the reception centres were clear with the respondents upon arrival that they did not engage in relocation procedures. Amanuel explained that he was sent to a camp via the dispersal mechanism and, upon arrival, the staff told him that they do not offer relocation. The staff told Amanuel to go to Milan to receive relocation and gave him bus money so that he could take the bus to Milan and apply for relocation.

Drivers of Secondary Movements to Rome and Milan

Most respondents who arrived spontaneously in Rome or Milan came in search of the relocation programme. The decision to leave the original centre was influenced by three main factors: duration of time in not receiving relocation, experiences at the questura and advice from family or friends.

Upon disembarkation and in the explanations of the relocation programme by cultural mediators and on leaflets received by some respondents, Eritreans were told that the relocation programme takes three to six months. Therefore, expectations formed in the respondents’ minds regarding when they should have updates on their cases and receive relocation. The reality was that the process took much longer—more commonly at least 9–12 months or more. The delays in the programme appeared to be due to a combination of delays in receiving Member States’ processing of applications, a lack of pledges from receiving Member States, and registration and identification delays in Italy. Nearly all respondents expressed concerns regarding the wait time for the relocation process. Mariam stated: ‘The waiting is too long. It angers me that you have to do nothing, your power is only to wait. It feels like we are wasting time here.’ Research has illustrated that feelings of wasting time are prominent amongst asylum seekers and slow procedures create anxiety and depression (Griffiths 2014). At the time of interview, respondents had been waiting from one month to over a year in some cases. Respondents expressed feelings of fear and anxiety that their case was not being processed. The long waiting led several respondents to believe that the relocation process was not progressing where they were; therefore, they left the centre and went to the cities in search of accelerating the process.

Second, as described in the first section of the results, respondents were experiencing difficulties at questura in understanding and registering for the relocation programme. It was evident during the research that, when discussing their retrospective decision-making, respondents did not have a clear understanding of the relocation process when making these decisions. Bad experiences at the questura such as not having cultural mediators to communicate properly and not receiving confirmation or information regarding their relocation application prompted decisions for secondary movements.

Third, networks were important in influencing the decision for secondary movement. Eritreans are highly connected, with most respondents having family or friends living in other European countries. In addition, respondents were frequently in touch with people from within their journey whom they arrived with or met in Libya and were ahead of them in the journey. Through these connections, respondents were able to assess and compare their current situation to others, seek advice and make decisions. Faven’s sister had applied for the relocation programme while Faven was in Libya and had been sent after four months to Finland. Upon arrival in Italy, Faven was allocated to a centre via the dispersal mechanism and, once at the centre, she called her sister in Finland for advice:

While we were talking, she asked me how many of us [Eritreans] were there. When I told her we were only three she told me I have to get out of there and take my chances because they might not take us seriously and make us stay in Italy. Then we ran out and came to Milan. On our way here others changed their minds and went to Germany, but I chose to stay here.

Advice and financial support from family and friends were important in making the decision for secondary movement to Rome and Milan. Faven also stressed that being in a group of other Eritreans is important, and several respondents reflected on this. Although not the primary reason for secondary movement within Italy, the respondents clearly wanted to be in locations with other co-ethnics and felt more secure when they were together. Being in a group enabled them to regularly discuss their cases together and understand that other people faced delays as well, while still others were receiving their relocation. Knowing this information and seeing people they knew finishing the process seemed to bring them comfort that the process was working and they just had to continue to wait. Being isolated and removed from the process, on the other hand, created anxiety that the process was not working.

Changing the Initial Decision to Relocation

As illustrated in Table 2, three respondents did not learn about relocation until their arrival in Milan. These respondents had a plan prior to arrival to migrate onwards beyond Italy as soon as possible. They believed, as illustrated in research with Eritreans (Belloni 2015; Brekke and Brochmann 2015), that they would have better opportunities in other countries.

In line with their plan to move onwards, these respondents left their place of first reception immediately upon arrival and went to Rome or Milan. All the respondents in this study had been fingerprinted, so this group does not represent those who may have been able to evade fingerprinting at disembarkation. In Milan, the first point of arrival was the centre called the Hub. The Hub was initially opened by the Comune di Milano in 2013 to respond to large flows of unregistered Syrian arrivals at the Milan train station. These were individuals who were primarily not registered at any municipality in Italy at that time. They were asked whether they wanted to stay in Italy or move onwards, and were provided with food and temporary shelter if necessary.

The role of the Hub changed as the policies in Italy changed. In May 2015, the Hub began to take on a central role for registering new arrivals to the city. At that time, the majority of arrivals were still not registered in other comunes in Italy. The Hub was a first point of arrival for all secondary movers to Milan, as well as for Dublin returns. Individuals that wanted to stay in Italy would arrive at the Hub for registration and wait to be placed in a CAS. The Hub was never intended to act as a shelter itself but, when reception places were unavailable and people needed shelter to sleep at night, cots were brought out to provide people with a bed for the night. At peak times, such as in summer 2016, the Hub was feeding up to 700 people per day.

From the migrants’ perspectives, the Hub was quite well known and became a central meeting place for new arrivals to the city from a variety of statuses. Tesfay explained:

Respondent: I was planning to go [onwards] by myself before I knew the [relocation] programme existed.

Interviewer: What was it that changed your mind?

Respondent: I got advice from the ones that have been rejected from trying to enter to other European countries on their own. They advise me that I should try legally.

It is evident that, by bringing together both Dublin returnees and new arrivals at the Hub, respondents’ decision-making changed from seeking to migrate onwards irregularly to choosing for relocation. Although most respondents had received information regarding the relocation programme prior to arriving in Milan, speaking to Dublin returnees had a different impact of understanding the possible consequences of moving onwards without status. Ngiste stated:

At first were trying to start our journey to escape to Switzerland. But once we got to Milan we heard about the relocation programme and we knew it was the best option to go by relocation. We also knew your acceptance to enter to another country is best if you go by relocation.

It appears that locations like the Hub are unique and clearly play a very important role for asylum seekers and other migrants by bringing together individuals of different statuses.

Secondary Movers’ Reception in Milan: A Changing Space

Respondents received a different reception in Milan and Rome. As already mentioned, the Hub was a well-known location amongst the Eritrean networks and provided the initial reception to Milan. Officially, asylum seekers registered with the national asylum system were supposed to return to the municipality where they were registered and not be accepted in Milan. However, there was a quiet tolerance in Milan experienced by Eritreans who came to the city seeking assistance with their relocation claims. Milan has had a long reputation of being a welcoming city to asylum seekers and migrants in Italy (especially as compared to Rome). Upon arrival at the Hub, respondents were required to go to questura to register as asylum seekers in Milan. At that time, the questura in Milan accepted registering individuals and respondents were assisted by the staff at the Hub to receive a place in a reception centre (most commonly within a CAS). Experiences varied, with some respondents having to stay in the Hub for up to two months, while others were relocated to a reception centre immediately and this primarily depended on space available and immediate needs. In the CAS, respondents were given legal advice and access to health care. They were assisted with their relocation claims and were able to move their case forward. In complex cases, respondents were assisted in having their cases transferred from the original municipality to Milan.

During the fieldwork, on 27 April 2017, the approach in Milan changed and the Hub was recast as a CAS, meaning that new arrivals to Milan could no longer arrive at the Hub to receive immediate assistance. This decision was made by the comune to improve safety and to better respond to the needs of migrants and the city residents. The purpose was to still meet the needs of migrants and to separate the registration process for migrants from the shelter services that are provided by the comune. Within the new process, new arrivals at the Hub were sent to the Asylum Seeker and Homeless Resource Centre (CASC) Centro Aiuto (literally meaning Centre Help), which provides emergency shelter for all homeless in Italy. At Centro Aiuto, they could request emergency shelter. In addition to going to Centro Aiuto for shelter, they were advised to go to the questura to register as asylum seekers in Milan (as per the former process). Upon receiving documents from the questura for their registration, they could be moved to a CAS within the city. However, in addition to changing the process for the shelter, the comune instructed the questura to refuse asylum seekers trying to transfer their case and to instead advise them to go back where they were originally registered. Figure 1 shows how entrances to a CAS within the city of Milan drastically reduced after 27 April 2017.

Figure 1

City of Milan: Number of Asylum-Seeker Entrants to a CAS in Milan. Source: Department for Social Emergencies, Rights and Inclusion, Municipality of Milan, May 2018.

Samuel and Jamila were met leaving the Centro Aiuto one morning in late May. They were cousins who had travelled together from Eritrea to Italy. Upon arrival in Sicily, they were separated and sent to different reception centres. Jamila spoke with other Eritreans in her reception centre and was told that no one had received relocation yet. She spoke with family in Sweden who advised her to go to Milan. She stayed for a month at the reception centre in southern Italy to save her weekly allowance to pay for a ticket to Milan. Upon arrival at the Hub, Jamila experienced a new reality in Milan. The Hub was no longer accepting people inside and she was told (as per the previous scenario) to go to guestura. However, the questura refused to register her in Milan and told her to go back to where she had just left. She told the questura that she did not have any money to return and they told her ‘however you came here, to go back how you came’. Resultantly, she went to Centro Aiuto, but to no avail, as they were not able to provide her with shelter. When we met her, she had spent the last 10 nights sleeping on the street under a bridge near the Hub. Jamila was distressed when we met and uncertain of what to do. Samual, her cousin, had been sent to a different reception centre, where there was no language assistance and, when he asked at the questura for relocation, he said they did not understand him. He came to Milan to connect with his cousin and was in the same situation.

Jamila and Samual were deciding between what they viewed as three future options: first, to stay in Milan and try to see whether they could eventually get help, but they were increasingly feeling that this option was futile; second, return to where they had been assigned, which felt difficult to them, as they were not sure how to get back without any money, whether they would be accepted back into the centres and whether they would be able to apply for relocation if they returned; and, third, they were considering going to France irregularly. They had the impression that people were not returned from France, only from Germany, Switzerland and other countries to Italy. France was viewed as open for asylum seekers and they would be welcomed there. They were told by others on the street that they could just keep taking buses towards France. They may be kicked off the bus for not having a ticket, but could then get on another bus.

In follow-up instant messages a month later, the cousins had separated and Jamila had returned to the town of her initial claim and was trying to get assistance from the initial reception centre. Samual had decided it was better to go to France and had left on his journey to France. He did not have a mobile phone, so Jamila was unsure of how his journey was progressing.

This change in policy implementation by the comune of Milan resulted in both of these respondents entering into irregular situations. For Jamila, this was because she had not been officially accepted back into the initial reception centre and was struggling to find shelter in the city she had been sent back to. For Samual, this was because he decided to engage in irregular secondary movement beyond Italy. It is important to stress that both Jamila and Samual wanted to receive relocation to join family in Sweden and, had they been able to receive assistance and apply for relocation in Milan, they would have done so.

Discussion

This article has sought to illustrate the interplay between the changing multi-level governance in the Italian reception regime and how Eritrean asylum seekers react to this system. Each of these will be examined further.

Changing Multi-level Governance in Italy

From 2014 to 2016, Italy was a changing environment for asylum seekers, moving from a country of transit to containment. The implementation of European Union-level policies, such as the hotspots and the increased border controls between Italy–Switzerland and Italy–Austria, has led to a reduction in secondary movements from Italy to other European Union Member States. The result has been a rapid increase in the number of asylum seekers in Italy and the resulting need for a higher number of reception spaces. For Eritreans, the high number of arrivals has meant their inclusion within the national distribution mechanism, which was not intended for relocation applicants. Dispersal mechanisms create marginalization for respondents sent far from kinship networks (Bloch and Schuster 2005) and the humanitarian emergency approach in Italy has led to significant variations in the quality and conditions of reception.

Milan can be viewed as a microcosm within Italy for exemplifying this shift from transit to containment. As the natural ‘gateway to Europe’, Milan was the key transit destination for secondary movers going to other European Union Member States. This role shifted to being a central destination of secondary movers who now sought sanctuary in Milan, placing a heavy and unequal burden on the local reception system. In effect, Milan quickly became accountable for a much higher number of asylum seekers than its allocation under the central distribution mechanism. The profile of individuals coming to the city also changed; in the past, the spontaneous arrivals had not previously been registered in Italy, whereas, in 2017, the spontaneous arrivals were registered in another municipality that was responsible for their case.

Resultantly, Milan changed its practices to enforce national laws for registration at the initial questura. It is important to stress that this was not a policy change, but a change in the implementation of an existing policy. The result, in effect, was an Italian system similar to the Dublin regulation wherein secondary movers were sent back to the site of initial reception. In a country the size of Italy, these could be significant journeys wherein assistance was not offered for the return back. Furthermore, upon return, it was uncertain whether asylum seekers were accepted back to the reception centre.

Eritrean Asylum Seekers’ Responses to the Asylum and Relocation Bureaucracy

For Eritreans, Italy was never the intended destination and the goal was always to move beyond Italy. Upon learning of relocation and, for some, the Dublin regulation, relocation became the central goal and decision-making was often changed towards relocation. The central challenge for Eritrean asylum seekers was ensuring registration and processing of their relocation case. Although the official position is that relocation is accessible across the country and equally applied, the results of this research and European Commission documentation show that this was not the case. Relocation was not well known by local authorities in remote locations that were not relocation hubs and therefore access to relocation procedures was thwarted.

Through their networks and experiences, Eritrean asylum seekers identified these discrepancies within the system and engaged in secondary movements to access the relocation procedures. Secondary movements within Italy can be viewed as an act of resistance against the reception regime in three ways: first, as an act against the juridical competencies for relocation, language support and legal understandings that are present in the Italian reception regime and especially the CAS centres; second, as resistance to feelings of ‘waiting’ and feelings of ‘lost time’; third, as resisting the marginalization of the dispersal policies from kinship networks and engaging in secondary movements to find other Eritreans.

Finally, the instigation of secondary movements was not only led by the asylum seekers but, critically, is also instigated by members of the reception regime, such as staff within centres that openly acknowledge barriers to access relocation in their jurisdictions. This is important as an influencing factor within these decisions.

Prior to May 2017, the result of the secondary movement was generally positive for Eritrean asylum seekers, as they were able to receive support in Milan and Rome for their case. This critically changed in May 2017, when the comune of Milan opted to change their implementation of the existing policies to reduce the unequal hosting burden on the city. The result found in this research was a highly negative impact of pushing Eritrean asylum seekers out of the relocation programme and into irregularity. However, it must be noted that the fieldwork ended in June 2017 and follow-ups indicate that the system stabilized as asylum seekers came to understand the new policies.

Conclusion

The Italian reception regime has been continually changing and shifting through multi-level governance goals. From the European Union, this was the implementation of both the hotspots and the relocation programme; from the national government, the changes in the reception centres to cope with the high caseload; and, at the local level, the changes in the implementation of policies to a shifting caseload and environment. Through these changes, asylum seekers are constantly trying to re-evaluate and understand their position and make decisions for their future.

This article makes a unique contribution by examining the relocation programme through the experiences of Eritrean asylum seekers in Italy. The results reveal that, first, the relocation programme was viewed as highly desirable amongst the respondents. Attraction to the programme was very strong and led respondents to change their initial decisions of irregular secondary movements beyond Italy to choose the relocation programme. Second, the uneven implementation of the relocation programme across Italy created fear and anxiety amongst respondents that they were going to lose their opportunity for relocation and be forced to stay in Italy. This fear, combined with advice from kinship networks, prompted onwards secondary movements within Italy in search of relocation.

For most respondents in this study, the decision for secondary movement in Italy worked out positively in their favour. It is known that at least 15 respondents have received relocation. In Milan, respondents were highly satisfied regarding the services they were receiving, including that they understand whether their relocation application had been processed and, although they were frustrated with the waiting, they knew and accepted that the waiting should result in their relocation, as they saw others receiving relocation and moving onwards. Due to changing policy implementation, for arrivals in Milan after May 2017, the result was very different.

Finally, the relocation programme has been heavily regarded as a failure within the European Union due to a lack of pledges by receiving Member States, slow implementation and a significant gap between the final number of individuals relocated and the programme goals. In contrast, this article shows that the programme had several positive effects on Eritreans in Italy, including increasing willingness for fingerprinting at arrival, reducing decisions for irregular secondary movements beyond Italy and satisfaction with having a legal opportunity for onwards movement. Improvements in implementation are clearly required in future programmes, although, for eligible asylum seekers in Italy, the programme appears positively regarded and should not be considered a failure. Relocation efforts are important for creating an equitable distribution of asylum seekers within the context of a European Union Common European Asylum System and are desirable from an asylum-seeker perspective. It is clear, however, that such efforts require greater solidarity and efficiency across European Union Member States to implement such programmes.

Acknowledgements

The author expresses gratitude to the Comune di Milano, Fondazione L'Albero della Vita and Fondazione Progetto Arca onlus for supporting and assisting in this research. This research was funded by a Rubicon grant of the Netherlands Organisation for Scientific Research (NWO) and the research was conducted whilst a Research Fellow at the Global Governance Programme, European University Institute. I am grateful to Anna Triandafyllidou for valuable comments on a previous version of this article and being a mentor to this research project and to Yordanos Mehari for excellent research and translation assistance. Pseudonyms are used throughout this article to maintain the anonymity of respondents.