Governing Asylum without ‘‘Being There”: Ghost Bureaucracy, Outsourcing, and the Unreachability of the State

Abstract

When, where, and how do asylum seekers encounter the state? Anyone seeking asylum in the Global North might meet state authorities of the country where they want to apply for international protection long before arriving at its borders. However, if the state often becomes “very present” by transcending its geopolitical margins in border control, once asylum seekers have managed to cross into national territory, the state frequently vanishes. Insufficient information, opaque proceedings, difficulties in reaching state agencies, which dramatically increased with the COVID pandemic, often translate into a denial of asylum seekers' rights and their exclusion from welfare programs. Moreover, following a widespread tendency to outsource public services, access to asylum and related welfare programmes are being increasingly mediated by a range of nonstate actors (such as NGOs, activist groups, companies, and individuals) acting as state agents. Drawing on the analysis of ethnographic results from Spain and Italy, this article proposes the concept of “ghost bureaucracy” to theorise the street-level bureaucrats from their absence and explore asylum seekers’ encounters with a seemingly powerful and omnipresent but unreachable state through closed offices, digital bureaucracy and third-party actors.

1. Introduction: Governing through Absence

The study of state governance through administrative practices usually focuses on the encounters between street-level bureaucrats implementing public policy intention and their clients, particularly in their ability to make decisions within the “implementation gap” (Lipsky 2010; Hupe 2019). Migration and asylum bureaucracy present a very fertile terrain for observing this phenomenon since noncitizens are particularly subject to constant administrative requirements to demonstrate the legitimacy of their permanence in another country. Paperwork is a significant part of noncitizens’ life: migration and asylum-related procedures have carried more and more weight within the vast realm of bureaucracy as nation-states in the Global North deploy their legal armoury to select wanted migrants and seal up their borders to the undesired ones (Mountz 2010).

Indeed, numerous studies have highlighted how bureaucracy has become one of the main aspects of the migration and asylum infrastructure regulating entry, permanence, and access to welfare in receiving countries (Artero and Fontanari 2021; Borrelli and Andreatta 2019; Darling 2022; Zetter 2007). The more restrictive the laws, the more complex and challenging it is to navigate the administrative and legal apparatus, and the longer the paper trails noncitizens have to build along their life trajectories become. Anne Tuckett (2018, p. 4) speaks of a documentation regime as “the nexus of documents, paperwork, and legal and bureaucratic processes that migrants must engage with in their efforts to become and stay ‘legal’, to bring family members into the country, and to attain citizenship”. Her account of migrants’ encounters with Italian immigration bureaucracy shows how such a regime produces uncertainty, frustration, and anxiety, contributing to their long-term marginalisation, but also offers some margins of manoeuvre, especially when laws and regulations are ambiguous and contradictory. They leave grey areas and loopholes, where street-level bureaucrats can exert some degree of discretionary power (Cuadra and Staaf 2014; Giacomelli 2021) and noncitizens can find “opportunities for self-fashioning, social mobility and forms of cultural citizenship” (Tuckett 2018, p. 19). Therefore, the human factor appears to be a pivotal element in disentangling intricate and obscure “webs of administrative regulation” (Reeves 2013, p. 511), as in the case of migration bureaucracy in general and asylum procedures in particular.

In this article, however, the focus is not on implementation gaps in public policy. Drawing on fieldwork in Spain and Italy, we propose to account for the experiences of asylum seekers and other nonstate actors in their—often failing—attempts to secure asylum from street-level bureaucrats. Our first-hand observations revealed how opportunities for face-to-face interactions in administrative contexts are quickly decreasing as the digitalisation of all kinds of services catches on, a process accelerated by the COVID-19 pandemic (Cortés Abad 2020). Therefore, we propose to look closely at what happens when encounters between the state, impersonated by street-level bureaucrats, and its clients (in this specific case asylum-seekers1), do not take place or take place through intermediaries. This allows us to observe how access to asylum-related rights and welfare is denied in practice by the situation of abandonment that the selective retraction of the state creates. Here, the adoption of the concept of “ghost bureaucracy” provides us with a useful analytical tool to conceptualise street-level bureaucracy in its acts of absence as much as in those of presence. It also opens up future possibilities for theorising the absence of the state. By this, we are not suggesting that the state has no agency or interest in the processes we observe. This article shows how, through the outsourcing of policy implementation, the state still exerts substantial control over noncitizens, but it does so by proxy. At the same time, however, by making itself unreachable, it loses grip over policy implementation: delegating certain functions to other actors entails avoiding accountability for the dirtiest part of the job but also becoming, in a way, more vulnerable to subterfuge and ingenuity by those subject to its rule (Abarca and Coutin 2018).

2. Methodological Notes

The present article is the product of a collaborative effort of three ongoing research projects focused on the Spanish and Italian asylum and reception systems2. All three studies rely on methodologies combining policy analysis with the ethnographic method.

Caterina Borelli’s analysis of asylum seekers’ relationship with state bureaucracy derives from her experience as a social worker and coordinator of an asylum seekers’ reception centre in north-eastern Italy. Between 2020 and 2021, she assisted 26 individuals daily (all males, aged between 20 and 50 years, mainly from western African countries) with their necessities within the reception program and their administrative and legal procedures. The methodology heavily relies on participant observation (full-time during her commitment as coordinator and part-time as a volunteer after being appointed as a Marie Curie research fellow).

Arnau Poy’s research examines asylum seekers’ access to the asylum procedure and the state reception program in Spain. To this end, he has conducted multi-situated ethnographic fieldwork in diverse localities of Spain, mainly in Catalonia, mobilising different methods of data collection. From July 2021 up to now, he has conducted 14 in-depth interviews with asylum seekers (eight males, six females, aged between 20 and 40 years, mainly from South and Central American countries such as Venezuela, Colombia, Peru, Honduras, Cuba) and 37 with the main actors of the asylum and reception system (civil servants, NGO professionals, lawyers, activists, volunteers, and experts). Additionally, he held two focus groups with NGOs that manage the entrance to the state reception program, carried out document analysis and conducted participant observation in a civil society organization. Lastly, some results presented in the article drive from a previous study (Poy 2022).

Alèxia Rué’s research explores state structures for the governance of migration. As part of her PhD research, she studies the experiences of frontline NGO reception workers in implementing asylum and welfare policies and their relationships with state actors and asylum seekers in Catalonia and Madrid, in Spain. She has conducted 35 interviews with NGO employees in public asylum reception programmes (most of them women, holding different positions, such as social workers, psychologists, and lawyers) and 6 interviews with other stakeholders and experts (as policymakers, researchers, civil servants, trade union representatives and non-frontline NGO representatives). The research also included participant observation in asylum reception sites, comprising NGO offices and an asylum reception accommodation facility in Catalonia (part-time for three months). Finally, her methodology has also involved documentary research of asylum policy documents, reception bureaucratic and procedural documents, and state budgets.

This brief description of our respective methodologies shows that our primary interest when interrogating the state approach to asylum is the resulting experiences of the individuals subjected to it. In our reading of the asylum bureaucracy, the focus is on something other than state officials’ daily interactions with laws and regulations: deeper scrutiny of their perspectives would require further research, and this is beyond the scope of the present work. Instead, we deliberately assume a bottom-up viewpoint, concentrating on asylum seekers and nonstate actors navigating asylum bureaucracy. Indeed, ethnography is an unbeatable lens through which to grasp how this system is apprehended, experienced, used, and navigated by asylum seekers and the nonstate actors involved: it accounts for daily practices that escape the gaze of the administrator (Malkki 1995, p. 499) but simultaneously shape asylum bureaucracy. We are particularly interested in the perspective of employees of subcontracted NGOs since, as Mountz (2010, p. xvii) argues, it only looks at those that make the state through their daily work that we can gain deeper understanding of the actual workings of the state beyond its institutional façade and, in this case, of the mechanisms of migration governance.

This paper consists of five parts. After a brief introduction and these methodological notes, section three presents the theoretical framework we applied to analyse the bureaucracy and welfare access in the asylum regime. In section four, drawing on the analysis of ethnographic results from the field, we look more closely at street-level bureaucrats from their absence, exploring asylum seekers and social workers’ encounters with an omnipresent but unreachable state through closed offices, digital bureaucracy, and third-party actors. The article ends with a concluding remark about the concept of “ghost bureaucracy” and future theorisations around current implications of the colonial legacy of asylum.

3. The Bordering Nature of Asylum “Ghost Bureaucracy”

Despite the romanticised lens (Jubany 2020) that presents asylum as the ultimate act of generosity states offer to humanity in distress, governments in the Global North are today more concerned with security issues and migration control imperatives than with the provision of protection and refuge (Gammeltoft-Hansen and Tan 2017). Even when some shifts along the spectrum of openness/closedness of borders could be appreciated, the offer of protection has always responded to the economic, political, and ideological interests of those who hold power (Schuster 2002; Akoka 2020). Far from challenging the state’s monopoly on migration control, asylum works instead as a reaffirmation that nation-states are territorial and membership organisations (Malkki 1995; Torpey 2000). Consequently, the asylum regime pivots on an arbitrary process of definition of deservingness that excludes from state membership those who do not fit into narrow legal definitions (Crawley and Skleparis 2018). It filters those who are eligible and deserving of protection and those who are not: to do so, it classifies and places noncitizens in a social hierarchy of stratified rights, obligations, and opportunities with unequal social and institutional support (Thomaz 2018). Asylum is, therefore, a continuous practice exerted over time in power relations of spatial compartmentalisation and social differentiation of mobility and belonging management (Yuval-Davis et al. 2019).

Ethnographies of the asylum regime unmask its inherently bordering nature in the sense of the constant process of production of otherness, which, when it cannot be rejected, is relegated at the fringes of society, where it is tamed, controlled, and contained both spatially and temporarily (Griffiths 2014; McNevin and Missbach 2018). In a word, it is a practice of building material and symbolic borders around certain types of people, as if those unwanted who dared to cross the frontiers of the Global North could never really arrive, never really land somewhere to settle. It is in these bordering processes that we can detect the coloniality of asylum (Picozza 2021), where coloniality stands for those “long-standing patterns of power that emerged as a result of colonialism, but that define culture, labor, intersubjective relations, and knowledge production well beyond the strict limits of colonial administration” (Maldonado-Torres 2007, p. 243). As Ann Stoler puts it, the “distribution of compassion” in the present world maintains a deep bond to the imperial endeavours of the past and their moralising mission:

“Detention centers for unwanted immigrants in France, refugee camps for Palestinians displaced by the Israeli state and fenced reserves for Native Americans and Australia’s indigenous population have been reconnected to imperial practices and their spatial logic. In using ‘natural’ landscapes and sophisticated architectural design to limit access to resources, to intensify vulnerabilities, and to cordon off subject populations in the name of protection and welfare, they rehearse as they elaborate upon practices long in imperial use.”

(Stoler [2002] 2010, pp. xii–xiii)

However, detention, encampment and other glaring formulas of segregation are not the only ways to keep subject populations at bay. Indeed, the production of the “other” is also present in the obscured colonial legacy of asylum, where asylum is marketed as a universal right while denying the exclusion of former colonies and their citizens from it (Akoka 2020; Mayblin 2017; Jubany 2020). These exclusions have been explicit, through the original territorial restriction in the Geneva Convention (Mayblin 2017), or implicit in “deracialised” (Masocha 2015) bureaucratic requirements (Pallister-Wilkins 2022). Such neglected history is also closely tied to the expansion of capitalism and imperialism (Chimni 1998) and undoubtedly critical in today’s international protection: from this perspective, such policies appear driven by the colonial logic and racialised discourses that maintain a hierarchical classification of humans (Mayblin 2017), contributing to what Van Houtum (2010) calls “the global apartheid”.

A shrinking protection regime, the product of restrictive transformations that, since the 1990s, have been reducing the spaces of asylum (Akoka 2011) within the national territory, in border landscapes and beyond national borders, into countries of origin and transit (Mountz 2010), seems to also be part of this categorisation of human beings according to the “colour bar” (Balibar 2002, p. 82). On national territory, multiple practices of deterrence and exclusion have been implemented within the asylum infrastructure to hinder social integration (Gibney and Hansen 2003) in what Hinger and Schweitzer call “disintegration policies” (Hinger and Schweitzer 2020). As Darling (2011) argues, reception conditions are often designed to deter the arrivals and tame those who enter by producing discomfort. Similarly, the notion of “welfare bordering” has been used to capture how the exclusionary politics to keep the unwanted others out are implemented on national territory, deploying new bordering devices around social rights and public welfare provisions (Misje 2020). Preference hierarchies, in any case, are not expressed only in processes of exclusion, but rather in differential inclusion (Mezzadra and Neilson 2010).

Given that states filter out who enters and who leaves and regulate the conditions under which those considered as being “always at the border” are allowed to stay, most of the asylum literature exploring how the exclusion/inclusion of asylum seekers is produced (Taylor 2013) assigns the leading role to state power (Gill 2010). Although recent studies nuance its omnipotence, they continue to place the state at the very centre of the day-to-day practices of street-level bureaucrats, asylum seekers, and citizens in general (Gabrielli et al. 2022). However, at the level of analysis, downscaling state centrality is crucial to make sense of its governance strategies and the experiences of asylum seekers. This becomes evident as policies of outsourcing are examined (López-Sala and Godenau 2022). In Spain and Italy, private actors have taken on the provision of welfare services to noncitizens either by direct subcontracting from the state or stepping in when the public welfare is made inaccessible, in what has been known as externalisation “from below” (Kritzman-Amir 2011). Our ethnographies reveal how these actors, despite not being state agents, become almost the sole visible face of the state for many asylum seekers and play a central role in constructing asylum bureaucracy.

With this article, we want to contribute to the debate on the governance of asylum seekers’ reception in Spain and Italy, and more broadly in the EU, by paying attention to that retraction of the state. We show how, in the specific case of asylum seekers, the state has made itself increasingly unreachable by retreating asylum bureaucrats from public contact. There is a contradiction in this process. A person seeking asylum in the Global North may encounter the authorities of these countries before departing. At this stage, state power is very present, directly or by proxy, due to the externalisation of border control (Gammeltoft-Hansen 2011). Nevertheless, it starts to fade on national territory, its corporeal presence becoming evasive. The outsourcing of public services, closed offices protected by private security guards, and complex digital bureaucracy are making direct contact with the state more difficult to establish. This phenomenon does not exclusively affect noncitizens: the digitalisation and virtualisation of bureaucracy, accelerated by the global pandemic, is a widespread tendency that we have observed in our respective terrains. While this process has certainly made the lives of the more tech-savvy population considerably more effortless, it has also widened the digital divide, increasingly excluding the digital illiterates and those that do not have access to digital technologies. We could appreciate this daily in our interactions with asylum seekers in their attempts to keep up with the requirements of a progressively digitalised administration.

Observing how asylum seekers struggle to find their way around administrative mazes rendered more impenetrable by digital bureaucracy reveals processes of abandonment and neglect by the state that unquestionably hamper access to welfare rights. This aligns with a more general tendency towards the governance of migration based on laissez-faire, self-governance, and neglect (Gabrielli et al. 2022). In this context, we propose the notion of “ghost bureaucracy” to refer to the evanescence of the state on national territory, the selective retraction of its officials and, therefore, of its corporeal presence.

With it, we account for the specific type of state-asylum seekers relationship we have observed in our fieldwork. This relationship seems to work one way: although the state is very much active through a complex apparatus of discipline and control of asylum seekers’ lives, it progressively becomes unavailable to them. In this article, we show, using ethnographic vignettes, the mechanisms by which the state continues to withdraw and how such unreachability is experienced by asylum seekers and their associates.

4. Experiencing State Absence

While the state is very present at its margins due to the externalisation of border control, once asylum seekers have managed to cross into national territory, it frequently vanishes; insufficient information, opaque proceedings, or difficulties in reaching state agencies, which have increased with the COVID-19 pandemic, often translate into a denial of asylum seekers’ rights and their exclusion from welfare programs. Using ethnographic vignettes from Italy and Spain, this section illustrates asylum governance through what we have called “ghost bureaucracy”. The first part deals with different forms of state withdrawal through closed offices and digital bureaucracy. In the second, we tackle the role of third-party actors in this retreat of the state, particularly those working in asylum reception acting as street-level bureaucrats. Finally, the section concludes by looking at how the state becomes a nebulous entity, also for these workers, and how this contributes to the governance of asylum through neglect.

4.1. The Withdrawal of the State: Closed Doors and Digital Bureaucracy

Historically, the Spanish asylum and reception systems have suffered from an infrastructural vacuum, including a systemic shortage of personnel and administrative offices (López-Sala and Moreno-Amador 2020). Although a typical story is that of a landslide of asylum applications leading to a collapse of the system, another version of it explains its oversaturation with the chronic lack of resources to manage it. This scenario of bureaucratic scarcity has created a bottleneck situation at many points along the asylum procedure that translates into a denial of asylum seekers’ rights by political neglect (Jubany and Rué 2020). The institutional abandonment of asylum applicants has materialised in extreme difficulty in reaching street-level bureaucrats due to long queues, delays of months for the appointments and the arduousness in obtaining them, most of which now must be requested online. In this context, the COVID-19 pandemic has not only meant closed offices: it has also accelerated the digitisation of asylum bureaucracy.

Within the Spanish territory3, the current procedure to apply for asylum is divided into two steps: the “manifestation of will” and the “formalisation” of the application. Applicants must first secure a registration appointment to certify their intention to apply for asylum. Upon completing this first step, the applicant receives another appointment for the interview and official filing of the asylum claim. Only then can the case be studied (see Accem 2019, p. 46).

As we described elsewhere (Poy et al. 2023), appointments to manifest the intention to claim asylum are no longer available through face-to-face interaction4; they can only be booked online. While this process is supposed to speed up and facilitate administrative procedures, the introduction of digital tools has reproduced and reinforced the existing barriers to access while eliminating the human factor of frontline bureaucracy, as the following quote shows:

“Before, people who wanted to apply for asylum queued up in front of the police station, even at night. The lines were long, but at least the problem was visible. Now it is not. We do not know how many people are trying to apply for asylum because now they first have to request an online appointment. They must request an appointment online even to express their willingness to apply for asylum. However, they cannot get an appointment on the police website because no appointments are ever available. So, access to asylum is blocked by this digital bureaucracy. Now the queue is done virtually.”

(Lawyer, NGO, Spain.)5

Interviews carried out in Barcelona show that asylum seekers spend up to 8 months to obtain an appointment to manifest their intention to apply for protection: it is important here to note that, during this time, they are not entitled to any kind of reception measure because, officially, they are not yet asylum seekers to the state. Against this situation, official complaints to the Oficina de Asilo y Refugio (OAR, Asylum and Refuge Office) and the ombudsman filed by NGOs and activists had so far gone unheard. In the meantime, the extreme difficulties in accessing appointments through the government platform have led to the flourishing of an underground market. Desperate applicants, unable to book appointments for months—either because of the persistent crashes of the platform or simply because they would get a repeated message informing them that “there are no appointments available”—use this market daily. Despite often finding themselves in situations of poverty, they feel compelled to pay (with prices up to 150–200 euros) to book an appointment for their asylum application or the renewal of their permit. An asylum seeker living in a squatted house in Barcelona explains:

“I was running out of money, but to get the appointment, I went to a call centre that a friend had recommended. Because getting the appointment alone was impossible. The web was always crashing, and I didn’t know how to handle the internet well. I paid 30 euros, and after a while, they told me I had an appointment.”

(Colombian female asylum seeker, 60 years old, Barcelona.)

In contrast with shady law firms and intermediaries profiting from the setbacks of the digitalisation of the system, NGOS and activists deploy alternative strategies—such as using free online platforms where volunteers spend hours trying to find appointments or negotiating with police officers on behalf of the most vulnerable ones—to overcome such obstacles. However, digital barriers do not end at the first step of the procedure. Waiting times to formalise the asylum application, for example, have increased significantly since 2015 due to the lack of state personnel: in 2019, the average wait to formalise was 6 months (Accem 2019). This way, we see how in Spain, the progressive digitalisation of asylum bureaucracy has created a scenario in which applicants have difficulties in accessing street-level bureaucrats both physically and virtually. In other words, the digital divide has added to the obstacles that asylum seekers already encountered along their administrative journey.

Regarding documentation renewals, we can find a similarly thorny situation in the Italian scenario, where an already muddled procedure got further complicated by the COVID-19 pandemic’s emergency measures. However, it is essential to first explain how an asylum seeker’s residence permit (permesso di residenza) works in Italy. Unlike residence permits for regular migrants6, this document’s duration in the specific case of asylum seekers is flexible and may range from one month to one year or more. The decision on the expiration date of the asylum seekers’ permit is made at the moment of its renewal at the Questura (police headquarters), according to factors that are never explained to applicants nor the social workers accompanying them. Here is another oddity: asylum seekers, differently from other immigrants, cannot make an appointment to renew their documents by themselves since the process is managed through the organisation hosting them7. However, at least in the city where Caterina conducted fieldwork, it is not directly social workers who can make appointments: these are communicated by the Questura that, with varying frequency (sometimes weekly, sometimes once every several months), opens a time slot for a certain number of asylum seekers hosted by a specific organisation. At the reception centre where she worked, no matter how many people had expired permessi, it would still be necessary to wait passively for an email from the Questura reading: “Send us ten people next Monday”. Trying to make a pre-emptive move and reach the immigration office before the documents’ expiration date would provide no results: emails systematically go unanswered, and the phone line only offers registered messages.

When the established day arrives, a social worker from the managing organisation must come along, pick up everybody’s folders containing all collected documents and pass them to the police officer at the door. This operation happens around 8 am, when the Questura opens. Asylum seekers are the last on the waiting list, their names being called after all the other clients (noncitizens with different types of permesso di residenza) have finished their procedures. The waiting usually takes 4 to 5 h. Since the COVID-19 pandemic, access to the building is not open to the public: applicants and social workers must wait outdoors. After hours in the street, an officer exits the building and shouts the name of each applicant out loud: that is when one finds out whether they have had their permesso renewed (one missing paper and the whole process fails) and for how long, with no possibility of reply. The physical impossibility of entering the facilities where the decision over the renewal is taken entails that no interaction with the person taking it will ever occur. It all looks like a roulette game, and everyone involved feels the Questura works in mysterious ways.

These two ethnographic cases expose the practical consequences of the increasing digitalisation of asylum procedures that, combined with inaccessible physical offices, we consider one of the key features of “ghost bureaucracy”. Digitalisation, far from alleviating the backlog of cases experienced by the system, seems to have reproduced, if not enhanced, barriers to access. At the same time, it has significantly reduced the margins of discretion that the human factor would grant, which, at times, could have helped mitigate the roughness of the system. Paired with a tendency towards externalisation, it has also meant that asylum seekers’ access to the right of asylum and the welfare entitlements attached to it is increasingly mediated by a vast range of nonstate actors. These actors end up playing a crucial role in the construction of asylum bureaucracy. As the next section will show, some residual room for manoeuvre remains in the interstices of overlapping laws and regulations.

4.2. Who Is the State?

The fact that, in both countries under scrutiny here, most encounters between asylum seekers and the state occur through third-party intermediaries reinforces the notion of “ghost bureaucracy”. Indeed, while the state directly manages most procedures linked to international protection claims, all those concerning reception have been outsourced to non-profit organisations. This is both a consolidated course of action and a growing tendency, made evident by reducing the percentage of services managed directly by the state, in reception and other areas.

In Italy, the management of reception programmes has been conceived as entirely relying on third-sector organisations from the beginning: the Ministry of Interior would finance those programmes but never be directly involved in their practical implementation. On the contrary, up to 2014, the Spanish state directly managed 50% of reception slots. However, as the programme grew between 2014 and 2020, the increase in the number of places has fallen entirely on the non-profit organisations: by 2020, reception slots directly managed by the state were only 5% of the total. First reception screening, which determines if an asylum applicant complies with the minimum requirements to access the reception programme has also been progressively vested upon the non-profits in the system. The process was completed in 2022 when the Spanish Red Cross took on managing Madrid’s office, the last directly managed by the OAR. The overrepresentation of these organisations in the management of welfare programmes directed to foreign citizens leads to the identification of their frontline workers with state agents, as they become almost the sole visible face of the state for asylum applicants:

“We do not receive clear instructions, and we cannot give accurate information. We say to all asylum seekers that the information we give is provisional and that there may be changes that do not depend on us. And that we are very sorry, but… We receive all the complaints from the asylum seekers, we implement the program, but we don’t make any decisions. They don’t always distinguish between us and the Ministry. For them, the state is an abstract entity.”

(Social worker, NGO that manages the entrance to state reception program, Spain.)

Thus, the progressive retraction of the state has meant that for asylum seekers, social workers from externalised organisations often represent their first—and almost the only—contact with the asylum system. It is these workers who, most of the time, provide emergency support upon arrival, manage public welfare services, and are an entry point to the asylum system where those seeking international protection may find legal and social support. Thus, social workers often act as a first filter to read and understand the receiving society, its codes, and mechanisms. They also can function as brokers and advocates on behalf of their beneficiaries by “giving applicants the opportunity to find their way around an otherwise impenetrable bureaucratic maze” (Tuckett 2018, p. 59). This multitude of functions places them in an odd position in between, a role that the increasing opacity produced by the virtualisation of public administrations is reinforcing. When access to the physical spaces of administration is denied, and the virtual ones are impenetrable to those with poor linguistic or technological skills, the mediation of social workers becomes more necessary than ever. This unreachability of public apparatuses results in social workers often being perceived by their clients as the state’s only visible face. Even if they are not public officials but employees of third-sector organisations, direct observation has shown us that this distinction may smudge until one can hardly discern the boundary between what is the state and what it is not.

However, sometimes such blurriness appears in full view; hence, it becomes more easily observable from an ethnographic perspective. Those are the moments when, from their contradictory position, social workers must directly implement state policies perceived by most as discriminatory and unjust while trying to maintain the humanitarian and rights-oriented character of their job. We can find a clear example in Italy when the Prefettura8 sends an official communication to a beneficiary via the reception centre managing organisation. In most cases, it demands that proof of their economic situation be provided or a direct notification that they have lost their right to reception (they receive the revoca, the revocation of the reception contract) and must leave the centre. Upon receiving the official communication by a certified email, within 14 days, social workers must forward the notification to the interested party, have them sign an acknowledgement of service and send it back to the Prefettura9. While fulfilling this specific duty, social workers are automatically invested in a role they did not sign up for: that of a public official. Here is a particularly critical point: at that moment, social workers are acting on behalf of the state; therefore, they should be serving its interests exclusively, even if—let us emphasise this one more time—their employer is not the state but the NGO that undertook a contract for the managing of the reception centre. That contract represents a double bind for the organisation and its employees nonetheless: if, on the one hand, it invests them with the control and care of the beneficiaries of reception measures, whose interests and needs should supposedly be their only focus of attention and intervention, on the other hand, it requires these organisations to respond for their actions to the state, their ultimate source of funding and legitimacy. Most times, this acting on behalf of the state means being responsible for kicking someone out of the programme. Refusing to do so, delaying or hampering the procedure in any way, could result in a disciplinary sanction for the employee and a pecuniary penalty for the organisation. Here lies the moral dilemma: the person who has received the revoca is someone the social worker may have known for months or years, with whom a relationship has probably been established, and of whose complicated situation the worker is undoubtedly well aware. They know that, in some cases, exclusion from reception may lead the former beneficiary to homelessness; however, they are still responsible for delivering the bad news and ensuring the expulsion occurs.

The provided example illustrates how the state delegating the policing of reception facilities to social workers practically turns them into street-level bureaucrats (see also Giacomelli 2021), which may seriously jeopardise their role as sociocultural mediators, and undoubtedly puts them in an uncomfortable situation. However, when the state retracts, it also leaves some spots uncovered, thus creating margins of discretion for NGO employees and possibilities of social navigation, meaning “the action of moving tactically in relation to social forces that confine or seek to move you” (Vigh 2009, p. 97), for asylum seekers. To show evidence of this, we shall quickly go back to the abovementioned example. Exit from reception usually happens quietly: when, for instance, the Prefettura asks for an income check, and the documentation produced by the organisation on behalf of the interested party confirms that the person is no longer entitled to reception measures10, then a second official notice is expected within some weeks, informing the beneficiary that they have 14 days to leave the centre premises. This deadline is compulsory on paper, but organisations apply some clemency in practice. The time that passes between the receiving of the first communication, the sending of the requested documentation and the final notice (which can easily sum up to a couple of months in total) is used to prepare the outbound beneficiary for life outside reception.

Actors can also use the margins of time left by the non-direct involvement of the state in the revocation of reception measures to fix administrative matters. As Abarca and Coutin (2018) observed, the documentation regime in immigration policies implies control over noncitizens’ lives but also opens up some space for resistance. What occurred when John, a 25 year-old man from Nigeria, received the revoca, is an excellent example of the human factor exploiting the void left by “ghost bureaucracy”. John had received a permesso with a very short validity from his previous renovation, and it expired before being assigned a new appointment; hence, he found himself in an irregular situation for a while. On top of that, the Cassazione (the court of third—and last—instance) rejected his claim, and eventually, the much-dreaded revoca arrived. Caterina and a co-worker decided they wanted to do one last thing for him: they had just received a new convocation by the Questura, and even if the revoca had already been noticed to John—so technically, he was no longer an asylum seeker entitled to reception measures—the 14 days for his expulsion had not passed yet, and he was still under the watch of the managing organisation. The two social workers also knew that state authorities do not often exchange data, or at least not promptly: the Questura was undoubtedly not yet aware of the Cassazione’s rejection and the consequent Prefettura’s revocation. They decided to take advantage of that loophole: they squeezed him into the next batch of appointments, finally getting him at least a new permesso before he left the reception centre11.

These are just two examples of the micro-negotiations between asylum seekers’ expectations and needs on one side and the functioning of the asylum regime and its practicalities on the other that for social workers of a reception centre are everyday routine. In our terrains of observation, incessant decisions are necessary to tend to the needs and defend the rights of beneficiaries while respecting the requirements of the bureaucratic and juridical apparatus. Such apparatus rests on laws and regulations that have piled up over the years and are often contradictory. Legal confusion, lengthiness, and the impossibility of putting everything on paper into practice create a grey zone of discretion that what we call “ghost bureaucracy”, meaning the non-direct involvement of state officials, further enhance. This in turn allows social workers some margins to calibrate their actions according to what they consider is “right” or “convenient”. Each time, the definition of this rightfulness or convenience may shift depending on various factors and personal preferences. In the case of John, a quiet and respectful man who never made any trouble in the centre, helping him have his permit renewed even if he was not entitled to it anymore seemed the right thing to do. In other cases, the revoca had offered social workers the opportunity to remove the unruliest ones. Officially, they did so in compliance with the state mandate. In practice, their main concern was preserving a peaceful cohabitation in the centre. The Prefettura’s notification gave them the perfect excuse to do so without being directly blamed for the expulsion.

In this section, we illustrated another critical dimension of the concept of “ghost bureaucracy”: the role third-party actors play in the retreat of the state, particularly those working in externalised public welfare services acting as street-level bureaucrats in asylum reception. In their relations with asylum seekers, the strategic identification and disidentification of the NGO employees with the state become another tool for the governance of migration made possible by the progressive withdrawal of public officials. While the state gains a reduction of welfare costs and a detachment from potential violations of human rights in dealing with asylum seekers (Kritzman-Amir 2011), organisations leverage this confusion both to gain compliance from asylum seekers and to detach themselves from the negative consequences of the implementation of restrictive reception policies.

4.3. Looking for the Human Factor

While the state extends its presence through the work of reception workers, who navigate between the reproduction of and resistance to state projects, from their perspective too, the state is quite a nebulous entity. It is where laws and directives that regulate the job come from, but guidelines are often obscure and contradictory in practice. Both in Italy and Spain, as Giacomelli (2021) and Pasetti (2020) also noted, the reception of asylum seekers continues to be managed, despite migration being a structural phenomenon, as an emergency matter, which implies that the whole sector lacks long-term vision.

For instance, Italy has yet to conceive a unified handbook that clearly explains duties and procedures: workers must learn on the ground by doing. During the first phase that followed the Arab Spring, when it was urgent to build a massive reception system overnight, social workers’ recruitment was rushed without paying much attention to employees’ backgrounds and previous experience (Giacomelli 2021). The longer-serving workers of the organisation where Caterina conducted fieldwork had been mainly recruited among cultural mediators who, before the “refugee crisis”, used to work predominantly as translators in courts; two of them were former beneficiaries of the centre. They undoubtedly had gained a good deal of practical knowledge of the sector; however, none possessed any formal education in social work or similar areas of specialisation, nor had they been trained for the job after being hired. On the other hand, the new batch of younger social workers who stepped in after the organisation had managed to survive the 2018–2019 cutbacks may have had more solid social sciences, international law, and social psychology qualifications. However, in practice, they still had to acquire the specific skills and competencies required by their position on the ground by doing. Given that professional profiles are not consistent and working procedures are not unified and clear-cut, the result is that, even within the same reception centre, beneficiaries receive a different treatment and consequently may receive different solutions to their problems, depending on which social worker assists them.

Spain has experienced a very similar situation. The rapid expansion of the state’s reception programme in 2014–2021 required NGOs to build teams almost from scratch. Hence, organisations drew from their pools of interns, long-term volunteers, and recent graduates: for most of them, the asylum reception programme was their first job in that field of expertise up to that point. Most of those with some experience in social work had never worked with asylum seekers before: for many, the right to asylum was something completely new, which they had barely heard of before. While organisations tried to put long-term workers as heads of teams to guide and train their colleagues, the programme expanded so rapidly that this was only sometimes possible, so numerous teams were made up exclusively of people without any experience in asylum reception. In the Spanish case, there is a unified handbook12 for the management of the state reception programme; nonetheless, this is subjected to repeated changes and amendments, making it difficult for frontline workers to keep up, especially for those who work in small organisations without prior experience in the field. These constant changes attest to the state’s approach to asylum from an emergency perspective and create a sensation of confusion and arbitrariness among reception workers:

“It does not make sense. We have been without a regulation of the asylum law since 2009, and, if there are no contradictions, the previous regulation continues to be applied. But the previous regulation dates back to 1995! Moreover, it is no longer useful because it does not address our current problems. In addition, they have been changing the Management Handbook every few months […]. It is tough for us to keep up because the norm changes rapidly. I do not know anymore how many management handbooks we have had. That causes us a lot of trouble, and it is upsetting because we go from telling a person, ‘you have the right to certain support’ and then, with the sudden changes, which are sometimes applied retroactively, telling them that they were no longer going to have it.”

(Coordinator, NGO of the state reception program, Spain.)

In the previous section, we have seen how the increasing withdrawal of the state’s frontline certainly leaves these workers room for manoeuvre: the other side of the coin is that their contact with state bureaucrats is becoming increasingly difficult. In Spain, reception workers’ communication with state officers is almost exclusively through the SIRIA database13. This database was created in 2008 to unify all the information on the state’s asylum reception programme. Frontline social workers input reports, requests, and any information demanded or deemed relevant on asylum seekers and their financial records regarding reception benefits and expenses. On the other hand, the Ministry of Inclusion, Social Security, and Migrations uses the database to request information and communicate the distribution of accommodation slots or any resolutions regarding asylum seekers and the organisations. In many cases, only one or two team members will have access to this database as “it is not user friendly, and it is easy to make a mistake that will have consequences for the users” (social worker, NGO). This software undoubtedly helps keep a close record of all the documentation and deadlines. Nonetheless, the automatization of the bureaucratic communication between the ministry and the organisations has played to the detriment of more unfiltered personal relations between the different agents of the reception system. A Spanish lawyer who has been working in the reception programme since before 2008 has the feeling that the lack of communication with the state has worsened with the expansion of the reception programme after 2014: previously, they “knew who each case officer was and [we could] give them a call, but not anymore, it is harder and harder to contact the OAR”.

The lack of direct contact with state officers makes it more complex to file complaints, propose improvements or find accountability. In charge of implementing the reception programmes in an ever-changing system, NGOs are almost the only visible face of the state in the eyes of asylum applicants and certainly the most accessible. Thus, complaints over the asylum system tend to address frontline NGO social workers who have minimal resources to communicate these to the state authority in charge: “the only way to get in touch with (state officials at the ministry) is to file an official request at the central state registry, but we do not know if we will ever get a response or if anyone will ever even read it” (social worker, NGO).

Communication channels between Italian social workers and state agents are possibly scarcer than in Spain. A similar database for the management of reception programmes is available for the ordinary reception programmes for refugees, but not for the extraordinary reception hosting the vast majority of asylum seekers in the country14. Here, the approach adopted by the state is undoubtedly a vertical one: we have already explained how decisions over new arrivals and revocations, or appointments for paper renewal, arrive, respectively, from the Prefettura or the Questura via certified emails to which NGO employees can only answer back, with little room for negotiations. In Caterina’s experience, the possibility of starting a conversation with such state bodies does not exist. Only on very numbered occasions was direct contact with the Head of Prefettura’s office established over the phone regarding the most severe and urgent matters and exclusively by the president, not by the coordinator or social workers.

Our respective ethnographies combined render an image of the asylum and reception systems as a foggy field where frontline workers grope around to deliver fair services to their recipients while trying to avoid stumbling upon some new directive that they were unaware of. When doubts or conflicts arise, employees have little-to-no way to transmit them to state officials. From their end, asylum seekers struggle to understand the functioning and all the requirements of these obscure and ever-shifting systems. Thus, they often end up referring uniquely to social workers who, despite not being state agents and having minimal contact with the state themselves, become the most visible face of an unreachable yet powerful apparatus:

“Many things are not clear to us either. When we have doubts about the program, which is frequent, we ask the central office to email the ministry. Then we wait. […] In the end, you have the feeling that everything depends on you, on your perception, and it shouldn’t be like that because it should be the same program everywhere in Spain. Despite our best efforts, we cannot provide clear information to asylum seekers, everything is provisional. Based on what I tell them during the first interview, they plan a course of action. Then, the program changes and you find yourself in a moral dilemma: what can you tell them? What do we do? Consequently, it is the social workers and beneficiaries of the program who must deal with the problems.”

(Social Worker, NGO of the state reception program, Spain.)

5. Conclusions

As noted in the introduction, the study of governance through administrative practices usually focuses on the encounters between street-level bureaucrats implementing public policy and their clients, particularly on their ability to make decisions within the “implementation gap” (Lipsky 2010). In this article, based on an ethnographic analysis of the asylum and reception systems working on the ground in Spain and Italy, we emphasised the need to contextualise these encounters. As argued by Gabrielli et al. (2022), the ordinary governance of asylum and reception conditions is conducted not only by implementing exclusionary policies but with multiple forms of laissez-faire, self-governance, and neglect. We have noted how, due to outsourcing and the digitalisation of asylum bureaucracy, asylum seekers’ access to welfare has become increasingly mediated by a vast range of nonstate actors and technological devices. Because of these processes, street-level bureaucrats are only one element, increasingly less visible and accessible, shaping the asylum landscape.

Taking this into account, in this article, we observed the occasions where asylum seekers’ encounters with street-level bureaucrats do not occur or are mediated by third-party actors and digital bureaucracy. This allows us to show how access to welfare is denied in practice by the retraction and abandonment of the state. This process, combined with the erection of other barriers that hamper direct contact with public officials, defines what we call “ghost bureaucracy”. With this concept, we propose an analytical tool that helps us reconceptualise street-level bureaucracy—in its acts of absence as much as in those of presence—as a part of an “architecture of repulsion” (FitzGerald 2019).

Focusing on the numerous hindrances that block direct access to state services and on the role of third-party actors in this configuration provides us with a more accurate understanding of state governance using selective retraction. State boundaries have always been elusive, teeming as they are with “porous edges where official practice mixes with semiofficial and the semiofficial with the unofficial’’ (Mitchell 1991, p. 82). In some places and times, and for some more than others, the state seems always there and present, the actors suffering its policing and exclusionary effects. However, just as it can be hyper-visible and far-reaching, it can make itself invisible, nebulous, and unreachable at discretion. To illustrate these dynamic processes, we employed the concept “ghost bureaucracy”.

Using ethnographic vignettes from Italy and Spain, we examined different forms of state withdrawal and the role of third-party actors in this retreat, particularly those working in externalised public welfare services acting as street-level bureaucrats. We also showed how the state also becomes a nebulous entity for these workers. An approach that interrogates the system at work has proved to be effective in unmasking the bordering nature of the asylum bureaucracy beyond the rhetoric of its neutrality. It also sheds light on the micro-battles that arise on the ground, thus revealing how the state, despite undoubtedly benefitting from outsourcing, simultaneously loses some control over its policies, leaving a broader room for manoeuvre to agents who wish to subvert the logic of state policies.

Indeed, these simultaneous presences and absences of the state that the term “ghost bureaucracy” captures are essential to understanding the Kafkian functioning of a system that, from the standpoint of social workers and their clients, seems to work in mysterious ways and at different speeds for some. We have recently witnessed how the approach to the reception of Ukrainians escaping Russian aggression operates in tandem with the far more restrictive regime reserved for other asylum seekers. It is difficult not to see here certain hierarchies of preference at work: the bias that makes us feel more sympathetic to those we consider closer to us, both geographically and culturally, while those deemed “too different” are subject to more meticulous processes of filtering, selection, and exclusion15. For the former, borders have been open with no state interference on movements (if all, assistance), and administrative procedures reduced to a minimum (Gammeltoft-Hansen and Hoffmann 2022; Rasche 2022).

The readiness with which the European Directive 2001/55 was activated for the first time since its adoption16, humanitarian corridors opened, and thousands of reception spots made available has left social workers of the sector amazed and puzzled in equal parts. For asylum seekers coming from the Global South on the contrary, the state is very present when it comes to restrictive border controls and bureaucratic requirements but makes itself unreachable to any reclamation of their rights. The ostensibly double standard in immigration and asylum policies that the European response to the Ukrainian war reveals is yet another proof that, as Mezzadra and Neilson argue, “borders are equally devices of inclusion that select and filter people and different forms of circulation in ways no less violent than those deployed in exclusionary measures” (Mezzadra and Neilson 2013, p. 7). Closure and repulsion are only one function of state borders, which are way more complex, flexible, and permeable institutions than that the classic image of the wall conveys. These authors focus instead on the capacity of borders to hierarchise and stratify, a subject that we tackled elsewhere (Poy et al. 2023), and that recurs here in our take on the “ghost bureaucracy” of asylum, even if perhaps more implicitly.

Through the workings of asylum procedures, their lengthiness, and the opacity and distance of public administration, othered subjects are kept at bay for years, uncertain about their future. They constitute a reservoir of cheap labour ready to be differentially included, deported if considered not essential or, in any case, always deportable. This is how, as Nicholas De Genova has repeatedly stressed, the deportation power carries out its “crucial work of subordination on the ‘inside’ of the space of the state” (De Genova 2018, p. 24). We understand “ghost bureaucracy” as a function of this mechanism underlying the asylum system: first comes the literal border separating the outside from the inside, then the filtering procedures to select who is provisionally admitted, and finally state bureaucracy managing the lives of asylum seekers by proxy.

The fact that the administrative apparatus is becoming increasingly impenetrable for noncitizens represents yet another obstacle along the migratory routes connecting the Global South with the Global North. It also opens up further questions about the colonial legacy of asylum that must go unanswered for now but will guide future developments in our research. Is the asylum system, with its obscure bureaucracy, another manifestation (maybe secondary to more blatant forms of domination but no less discriminatory) of what Derek Gregory calls “the colonial present” (Gregory 2004)? The author reminds us that at the foundation of colonialism was the practice of depicting others as others as much as the exploration and physical occupation of foreign lands. Quoting Edward Said quoting Karl Marx, he points out that othered subjects are those that “cannot represent themselves; they must be represented” (2004, p. 8). Does not asylum function on the same basis, by filtering, categorizing, and labelling others to turn them into objects of compassionate intervention (Zetter 1991, p. 44)? As argued by Rosalynd Eyben, “[t]he modern forms of aid bureaucracies both shaped and were shaped by the relations between the colonial powers and the countries they were ruling” (Eyben 2007, p. 33). Considering this, it is imperative to question how current bureaucratic configurations of “othering” systems such as asylum reception relate to historical mechanisms of colonial dominance through bureaucracy.