**IMISCOE Research Series**

# Claudia Finotelli Irene Ponzo *Editors*

# Migration Control Logics and Strategies in Europe

A North-South Comparison

**IMISCOE Research Series**

This series is the offcial book series of IMISCOE, the largest network of excellence on migration and diversity in the world. It comprises publications which present empirical and theoretical research on different aspects of international migration. The authors are all specialists, and the publications a rich source of information for researchers and others involved in international migration studies. The series is published under the editorial supervision of the IMISCOE Editorial Committee which includes leading scholars from all over Europe. The series, which contains more than eighty titles already, is internationally peer reviewed which ensures that the book published in this series continue to present excellent academic standards and scholarly quality. Most of the books are available open access.

Claudia Finotelli • Irene Ponzo Editors

# Migration Control Logics and Strategies in Europe

A North-South Comparison

*Editors* Claudia Finotelli Complutense University Madrid, Spain

Irene Ponzo FIERI (International and European Forum of Research on Immigration) Torino, Italy

ISSN 2364-4087 ISSN 2364-4095 (electronic) ISBN 978-3-031-26001-8 ISBN 978-3-031-26002-5 (eBook) https://doi.org/10.1007/978-3-031-26002-5 IMISCOE Research Series

This work was supported by IMISCOE

© The Editor(s) (if applicable) and The Author(s) 2023 **Open Access** This book is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made. . This book is an open access publication.

The images or other third party material in this book are included in the book's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the book's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specifc statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

### **Foreword**

*Migration Control Logics and Strategies in Europe: A North-South Comparison* is a highly successful book. The lion's part of the credit corresponds to the editors, Claudia Finotelli and Irene Ponzo, who have designed and guided it, assembled the group of scholars responsible for the excellent collection of articles, decided upon the conceptual framework, selected the focus shared by all the participants, and contributed the introductory chapter and the conclusions.

The relevance of migration control in nowadays Europe does not need to be pondered. Paralleling the growing concern for irregular migration, in recent decades it has become the overriding policy priority, relegating to a less prominent place other areas of migration policy. Its centrality has greatly grown in an increasingly politicised and securitised context.

The priority generally accorded by governments to migration control has grown with the increasing realisation that the efforts deployed to secure the borders yield less than satisfactory outcomes, despite the growing fnancial resources and manpower devoted to it. This is especially true of mass irregular arrivals by sea, after dangerous maritime crossings. The alarm created by these situations in recent years has reached the point of overshadowing other avenues for unauthorised residence, including overstaying, reputedly the largest in volume. It has fed feelings of sovereign incapacity to secure the borders, in the absence of cooperation with sending and transit countries. This inability is increasingly recognised by the European Union and by individual Member States, which have resorted to strategies of externalisation of border control, prospectively extensive to the processing of asylum demands. The ensuing need of securing the collaboration of origin countries is enhancing the relevance of a formerly secondary and hardly effective policy area – international cooperation with origin countries to foster development in order to remove the root causes of migration – reorienting its contents towards the externalisation of border control in exchange for compensations. This is, in broad strokes and impressionistic terms, the convulse context in which this timely and rich volume comes to light.

The book uses a comprehensive lens to look at migration control, shedding fresh light over a vast policy territory that encompasses the effectiveness of control

policies; visa policy and external controls; cooperation with sending and transit countries and externalisation strategies; the functioning of the asylum system and its increasingly complex relations with border control; the enforcement of internal controls; the management of irregular migrants; deportations; and regularisations, old and new. To these it adds other proximate topics that have to do with other similarities and differences to be found between Northern and Southern countries in Europe, such as labour migration policies, and especially seasonal labour demand in agriculture; particularistic selection practises; and the welfare-migration nexus of both young labour migrants, in one direction, and non-labour elderly migrants in the opposite one, both generating diverse degrees of welfare chauvinism.

The detailed analysis of these policy domains yields a staggering wealth of information and thought that no single scholar could possibly provide. It brings together a frst-rate group of researchers specialising in the different domains of migration control and connected felds in Europe. But the book's collective authorship does not detract in the least from the degree of consistency and thematic integration that could be expected from a single author. This is largely due to the use of a common focus and a conceptual and theoretical framework, which owes much to the design of the editors, shared by all the participants in the endeavour. The ensuing consistency is one of the assets of the book, together with a collective penchant for confronting inherited ideas.

The axis of the adopted conceptual framework is the notion of migration regime, in this case of migration control regime. It is a pillar of the book, and a wise choice, for the double reason that it confers consistency to the volume, as it is shared by all the participants, and that it works smoothly, almost naturally, across the different chapters. Admittedly, migration regime is still an under-debated and insuffciently thematised concept, and no scholarly consensus about it has been reached yet. But the vigorous popularity of the notion points to its very necessity. The analysis and understanding of the complexity of migration control in our days requires a framework that goes beyond the state, recognising the agency of a plethora of actors and implementers, public and private, at all levels of government, including street-level bureaucrats, and civil society organisations. It overpasses normative contours, pays due attention to policies and to practises, wanted and unwanted, and recognises the existence of diverse interests, be they geopolitical or economic in nature. The notion of migration regime decentres the government and helps to understand that the states' degrees of freedom are not unlimited, and that policies do not happen in a vacuum. The development of the notion adopted here is the one proposed by Cvajner, Echeverría, and Sciortino, the last two participants in the volume. And the fact that it works perfectly in the complex and multifaceted scenario of migration control attests to the operationality of the concept.

The unifying focus of the different chapters of the volume is the alleged North-South divide in matters of migration control in Europe. It is surely the most vigorous effort to rethink the divide, aiming to come to terms with a debate that has been intermittently in the fore of attention in the last quarter of a century. The dichotomic view of migration in Europe had witnessed two different, albeit connected, applications, thematically broad and of purely academic interest the frst, and more practical and focused on the realm of migration control the second. The former, subjacent to the one dealt with in the volume, stemmed from the interest aroused by the telling novelty constituted by the migration transition experienced by countries of Southern Europe in the last quarter of the twentieth century, followed by the dramatic explosion of numbers registered in the frst decade of the twenty-frst century. Unsurprisingly, the novelty attracted scholarly attention. The addition of a handful of countries to the relatively short list of immigration-receiving countries led researchers to put the double question of whether Italy, Spain, Portugal, and Greece could be seen as a unitary group in terms of migration, and whether they were signifcantly different from their Northern neighbours, representing if this were the case a new model of migration. The focus was the identifcation of similarities and differences among the four Southern countries, primarily in terms of socio-economic structures and policy stances, against the background of an often undefned and taken-for-granted Northern model. Such interest led to a scholarly debate, animated by contrasting positions.

With the passage of time, the interest for this version of the divide tended to fade, despite the fact that the existence of signifcant structural differences between North and South were corroborated by the much bigger impacts of the Great Recession on the Southern countries, in terms of unemployment, associated with higher proportions of young, less educated, and recently arrived immigrant workers.

Yet, the version of the alleged divide that constitutes the unifying thread of this book, although not unrelated to the former, is the one that has to do with real or supposed North-South differences in the realm of migration control in Europe. The dichotomic representation of such differences stemmed from the distrust of Northern countries towards the capabilities and stances of their Southern partners in the matter, and more precisely from the fear that the accession of three of them to the EU, and the four of them to the world of Schengen, seen as transit countries, would provide easy avenues for the arrival of irregular migrants to the EU. Such suspicions and misgivings were tributary of a stereotyped vision of an EU divided in two blocks which opposed strict to lax countries, strong controllers to weak ones, or good to bad partners, a representation not devoid at times of Manichean overtones and hardly veiled moral judgements and expressed with colourful metaphors.

Affected and nuanced in different forms by the Great Recession, the European refugee crisis, and the pandemic, this view has persisted until our days, even though the recurrent exigences of Europeanisation from the North were unreluctantly accepted by the South, as proven *inter alia* by the early adoption of reassuring laws by Spain and Italy. This vigorous commitment towards Europeanisation by Southern countries would be made compatible with the development of their own immigration agendas, in an exercise of top-down convergence which would later give way to relationships of a different sort.

Subjecting to critical scrutiny the supposed divide in matters of migration control, to ascertain if it had ever existed or if it is still with us, required a far from easy comparison. To make it feasible and agile, Finotelli and Ponzo devised a smart analytical strategy. They selected two countries to represent the North – Germany and the Netherlands, and two to stand for the South – Italy and Spain, complemented by chapters on the United Kingdom, Portugal, and Greece illustrating other policy domains.

After analysing the different components of migration control and the articles overseeing germane policy areas, the volume questions the supposed North-South cleavage, proposing instead a more nuanced, varied, and complex view. The divide is replaced by a mosaic that crosses the geographical boundaries. The book discusses a variety of cases in which the alleged dichotomy is not validated by reality. In some areas, a dominant trend towards policy convergence is found; in other, differences prevail. Infuences do not only run from North to South, but also in the opposite direction, signalling cases of Southernisation, in addition to transversal or bidirectional trends. Any hint of asymmetrical relations in the matter has disappeared. The variety of similarities and differences between countries is no longer captured by the idea of a divide. If it ever existed, the dichotomy was not as clear-cut as it has been often presented. And the state of things it evoked has evolved, as the book shows, amid a changing context characterised by new migration dynamics, Europeanisation, changing national migration realities – including variations in the age and family composition of the migrant population, changing policy orientations, and mutations in the international sphere. In this context, control practises have not been immune to the increasingly restrictive drive which presides over migration policy at large. In its balanced and wise style, the book concludes that migration policy boundaries in Europe have become blurred.

More than a century ago, in a famous dictum, Alfred North Whitehead advised to seek simplicity and distrust it. In this case, the ones who distrust the view of two markedly different blocks of countries, each internally homogeneous, separated by the Alps and the Pyrenees, stand in the opposite position to the ones who sought simplicity. The critical orientation adopted by the authors of the book has been validated by the surgical dissection of the major tenets of the divide, paying due tribute to the complexity inherent in the realm of migration control in Europe, in stark contrast to the simplicity of a dichotomic divide.

Complutense University Joaquin Arango arango@cps.ucm.es Madrid, Spain

# **Acknowledgements**

We would like to thank Ferruccio Pastore and the two anonymous referees for their insightful comments to the frst version of this book. We are also grateful to Eleonora La Rosa, Sofa Centroni, and Daniel González Benavente for their help in the manuscript editing. The editing of this book has been supported by the Programme HorizonEurope of the European Commission [Grant Nr. 101061314 – MIrreM] and the Spanish Ministry of Science and Innovation [Grant Nr. PID2019-104706GB-I00]. A special thank goes also to FIERI (International and European Forum of Research on Immigration) and Compagnia di San Paolo Foundation for their fnancial support.

# **Contents**




xii

# **Chapter 1 Introduction: Understanding Migration Controls in Europe**

**Claudia Finotelli and Irene Ponzo**

#### **1.1 From Models of Migration Control to Migration Control Regimes**

Migration control has largely dominated the public and political debate on immigration since the beginning of the 1990s. In particular, the mismatch between the restrictionist goals and the expansionist outcomes of migration control policies has captured the attention of the public as well as academia (Cornelius et al., 1994; Joppke, 1999; Joppke & Guiraudon, 2001). The presence of a large number of politically unwanted migrants (in the form of asylum seekers and irregular migrants) in Western European states despite increasing barriers and controls certainly represented the most evident example of such a contradiction.

The research interest in the mechanisms of migration controls and their outcomes has led to two distinct types of literature. On the theoretical level, researchers have focused on the limits of what migration controls can achieve, addressing the power of liberal constraints (e.g. through the action of domestic institutions), international norms, organised interests and, more generally, the role played by different types of actors and venues in the feld of migration policy (Hollifeld, 1992; Soysal, 1994; Freeman, 1994; Jacobson, 1996; Joppke, 1998; Guiraudon, 2002; Lahav & Guiraudon, 2006; Castles, 2004a, b; Boswell, 2007). On the empirical level, the question of migration controls led to a large number of comparative studies in Europe and overseas, where researchers' attention focused on the similarities and differences of Western European countries' policy performance, and, in particular,

C. Finotelli (\*)

Complutense University, Madrid, Spain e-mail: cfnotel@cps.ucm.es

I. Ponzo

© The Author(s) 2023

FIERI (International and European Forum of Research on Immigration), Torino, Italy e-mail: ponzo@feri.it

C. Finotelli, I. Ponzo (eds.), *Migration Control Logics and Strategies in Europe*, IMISCOE Research Series, https://doi.org/10.1007/978-3-031-26002-5\_1

on the question of whether some countries were more effective than others in controlling migration. Although a relevant strand of literature has focused on convergence trends between the two sides of the Atlantic (Hollifeld et al., 2014; Widgren, 1994; Papademetriou & Sumption, 2011; Finotelli & Kolb, 2017), the large majority of studies have identifed different models of immigration control; these models were considered the result of different types of migration histories, different political contexts and different levels of policy effcacy (Miles & Thränhardt, 1995; Freeman, 1995; Brochmann & Hammar, 1999; Martin & Miller, 2000; Boswell, 2003). In general, the empirical comparison of migration control in Western countries has been framed in terms of divergence, "effcacy gaps" (Czaika & de Haas 2013) and policy models.

Especially in Europe, these classifcations have been defned by more or less developed asylum traditions, more or less exposed borders, or more or less experience in immigration when sorting countries of immigration in terms of state regulation capacity. In such a context, attention turned quickly to Southern European immigration countries, which had become new immigration countries and the new "guardians" of the European border in the space of a very short time (Baldwin-Edwards & Arango, 1999; King et al., 2000; González Enríquez & Triandafyllidou, 2009; Ambrosini & Triandafyllidou, 2011; Peixoto et al., 2012). They were considered latecomers that had to manage unexpected fows on the fy without a clear immigration model. They appeared to be both more exposed to and less able to deal with the challenge of unwanted migration fows than other European countries. In this context, the idea of a "Southern European model" of migration started to gain ground in the mid-1990s, driven by the conviction that the capacity to control migration was defned by a European divide on immigration between "strong" Northern and "weak" Southern countries in Europe (Freeman, 1995; Baldwin-Edwards, 1997). Since then, the North-South divide in immigration control policies has been a persisting feature of the migration debate, which has contributed to forge a "negative exceptionality" of the Southern European countries in comparison with the rest of Europe, and which still underpins the political and academic debate over immigration control.

The persisting relevance of the European North-South divide in migration studies points to a widespread understanding of migration control outcomes in Europe as simply the result of more or less effective state policies. Yet, are national models of immigration based on perceived national divides an adequate heuristic tool to grasp the complexity of migration control policies and their outcomes in Europe? And from a more empirical perspective, can we really understand migration control outcomes of European countries as the result of more or less effective policies? Even when so, by what rationales should we assess state effectiveness? The number of unplanned entries and irregular migrants as well the prevention of secondary movements or the ability to meet economic and social demands? More generally, can European countries, having different economies, institutional cultures and geopolitical positions, reasonably pursue shared policy goals in terms of migration controls – and adopt similar means to achieve them?

The goal of this book is to contribute to the current literature by showing that the understanding of migration controls and their outcomes requires going beyond the juxtaposition of more or less effective state policies and clear-cut national models. We intend to overcome the tendency to develop country-based typologies (Boucher & Gest, 2015) that over the last few decades have turned Europe's traditional destination countries into benchmarks against which other countries are compared, with the risk of grouping the outliers into a single cluster and subsequently framing dissimilarities as pitfalls, as has occurred in the case of Southern Europe (Ponzo, 2021). This does not imply denying differences among European countries. If anything, it implies the very opposite: we argue that structural differences in institutional and welfare cultures, economic dynamics and geopolitical positions play crucial roles in the way countries regulate and conceptualise immigration (Bommes & Geddes, 2000; Bommes & Thränhardt, 2012). From this perspective, it becomes clear that the policy goals and tools developed by European countries in the feld of migration control cannot be understood and compared outside of this context. Migration controls are not regarded as the mere outcome of rational planning and, even less of state effcacy but rather as an imperative deeply embedded in a dynamic interplay of internal structural constraints, different geopolitical and economic interests, and ever-changing external context.1

Starting from this assumption, we employ the concept of regimes, which we regard as better able to grasp the complex reality of migration policy. Nation-states, according to the regime concept, are conceived as political organisations with different welfare cultures, institutional traditions and regulation frameworks that respond to a number of functional imperatives (e.g. security, equity, institutional legitimacy etc...) (Bommes, 1999; Boswell, 2007). They include different types of organisations, with different logics; these organisations such as state ministries and agencies interact with other social systems, such as the economy, but also other states. That is why the notion of a migration regime allows scholars to understand migration control policies not as the consequence of "bad" or "good" political wills but rather as embedded, negotiated outcomes of multiple actors and organisations with different interests and different functioning logics. Against this backdrop, the concept of regime

brings to attention the effects of norms in contexts, rather than operating a simple review of juridical rules.…[A] country's migration regime is usually not the outcome of consistent planning. It is rather a mix of implicit conceptual frames, generations of turf wars among bureaucracies and waves after waves of "quick fx" to emergencies, triggered by changing

<sup>1</sup>A similar effort could be done in the feld of migrant integration where the adoption of the concept of regime could produce equally fruitful results. Still, in this volume we have prioritised migration controls given that the contestation of the North-South divide and "national models" (eg. the French assimilationist, the British and Dutch multiculturalism, the German ethnocentrism) appears more advanced with regard to migrant integration where several scholars have looked at different types of actors, strategies and structural constraints conceiving of national models as mere discursive frameworks (Baldwin-Edwards, 2012; Bertossi, 2011; Cebolla-Boado & Finotelli, 2015; Favell, 2004; Fellini, 2018; Garcés-Mascareñas & Penninx, 2016; Giugni et al., 2005; Ponzo, 2021; Finotelli & Ponzo, 2018; Schain, 2009).

constellations of actors. The notion of a migration regime allows room for gaps, ambiguities and outright strains: the life of a regime is the result of continuous repair work through practices. (Sciortino, 2004: 32–33)

The concept of regime makes it possible to understand migration policies not only in terms of goals and outcomes, as often conceived in the migration control literature, but also as a process "through which public and private bodies, as well as decision-makers and administrative agencies, can coordinate (or at least try to coordinate) their expectations and produce and carry out governing decisions" (Cvajner et al., 2018: 13). In such a process, national models lose their explanation function and are seen "as loose discursive frameworks actors may use to make sense of the problems at hand and to locate themselves in relation to others" (*ibid*.).

Taking this approach to study migration controls allows us to unravel the policy practices and organisational strategies of different components of migration regimes, thus enabling us to understand their actual functioning while going beyond nationalbased typologies. Specifcally, the authors in this volume conceive of control outcomes as the intended (and unintended) results of strategies pursued by a wide range of public and private actors. Against this backdrop, states are just a few actors among many, and they have to deal with different interests, such as geopolitical or economic priorities, as well as with established routines, compelling public expectations, and a tangle of formal and informal rules.

Given that interests, public expectations and rules are not geographically blind, the differences between the North and South of Europe cannot be disregarded. Leaving aside the diverse history, we cannot consider geographical position as simply a minor inconvenience with no bearing on the issue at hand. Territorial location shapes migration regimes both directly, by producing different levels of exposure to different migration fows at different points in time, and indirectly, by shaping international relations, trade agreements, security priorities, etc. that in turn impact immigration policies. Hence, our aim is not to deny the existence of any difference between Northern and Southern European countries. Instead, we attempt to reframe those differences making sense of them instead of reducing them to Southern Europe's pitfalls or a lack of Europeanisation.

We intend that states' control imperatives are embedded in a complex interplay of varying geopolitical and economic interests, and internal and external constraints. More precisely, we focus on the imperative of border controls, internal checks, and residence regulations that we assume is present in all European countries. On the one hand, this common imperative might push states towards policy convergence; on the other hand, different geopolitical and economic priorities as well as different internal structures (e.g. welfare structures or institutional settings) might lead to the persistence or emergence of policy divergence among European countries. Consequently, it becomes impossible to draw clear-cut dividing lines across the continent.

This theoretical perspective brings about practical implications as well. Avoiding oversimplifcation and accurately framing those differences should prevent defning common policy goals and tools where there is no ground for them. At the same time, this reframing allows us to highlight the fact that convergence among Western European countries might be more advanced than suggested by the ongoing European-level political disputes around migration and asylum: this book points out the blurring of boundaries of "national" migration regimes and shows how similarities in specifc domains of immigration policy (e.g. labour migration, external controls, internal controls, asylum, etc.) among European states might prevail over internal consistency of individual countries' overall immigration regime.

Finally, this volume speaks to the Europeanisation debate. First, we take a different stance upon Europeanisation that, as the book chapters make clear, appears as a hybridisation of strategies, logics of action, and practices rather than either the topdown adoption of common regulations issued by EU-level entities or horizontal convergence among clear-cut national approaches. Second, we highlight how Southern European countries are taking on a new role in this process, turning from students to teachers. Policies of traditional immigration countries constituted the initial blueprint for EU legislation in this feld (Boswell & Geddes, 2011; Post & Niemann, 2007; Zaun, 2017), whereas Southern European countries were generally portrayed as passive receivers and dysfunctional implementers of EU norms and standards. However, on the ground, the differences are less cut and dry, and we can observe a sort of "Southernisation" of models, where Southern European countries have to some extent inspired EU's more recent approaches, especially with regard to external controls and asylum.

The last fnding is a picture of how Europe really works in the feld of migration, exposing and countering a double myth: not only that of a North-South divide, but that of "national models" of migration control. In this vein, the book responds to the need to reassess and give nuance to the (mis)conception of a neatly divided and clustered Europe, thus contributing to a proper understanding of the migration panorama, and to adopting appropriate policy strategies.

#### **1.2 The North-South Divide as the Undying European Cleavage**

Europe's North-South divide in migration policies has appeared as the analytical opposition *par excellence* in the European migration panorama. The rhetoric of the North-South divide has been present in a good deal of the recent European migration history. Until the 1980s, it referred to the existence of two different migration realities, i.e. emigration countries in Southern Europe, and traditional destination countries for labour migrants in Central and Northern Europe (Castles & Kosack, 1973; Miller, 1981).

In the 1990s, the suppression of internal borders after the enforcement of the Schengen Agreement and the creation of a common external border marked a turning point in the defnition of the North-South divide. Since then, countries in Southern Europe have been the target of endless pressure to reinforce their borders (which had become European borders) as well as of repeated accusations of not being up to the task.

With the frst European migration crisis, triggered by the breakdown of the Soviet Union, the idea of a North-South divide also began to be applied to the control of intra-EU movements, especially of asylum seekers, and Southern European countries started to be accused of promoting the transit of unwanted secondary fows to Northern migration countries. The argument was that asylum seekers who had arrived in Southern Europe decided to move towards Northern European Member States, where they expected more generous welfare benefts and fairer asylum procedures (Efonayi-Mäder et al., 2001; Thränhardt, 2003).

The Southern European countries' reputation as shrewd transit countries went along with the one of incubators of irregular migration, with a tolerant civic culture towards law breaking, plenty of opportunities offered in the informal economy, and frequent regularisation processes. Such a contrast with the Northern European "asylum magnets" triggered the perception of asymmetric migration regimes in Europe. While Northern European countries were seen as traditional destinations for asylum seekers, Southern European countries were considered the main destinations for economic migrants, whose irregular employment was facilitated by a large informal sector of the economy and weak internal controls (Santel, 1995; Finotelli, 2009; Echeverría, 2020).

With the 2008 Great Recession, the idea of the North-South divide was extended to intra-EU mobility. It revived the perception of Southern European states as sources of emigrants, often regarded as welfare scroungers. Concerned by the substantial increase of infows from the disrupted economies of Southern Europe, some Northern European countries restricted access to social rights of EU citizens in an effort to promote the return of those EU foreign citizens without employment who were dependent on welfare (Lafeur & Stanek, 2017).

The refugee crisis of 2015 reinvigorated the image of Southern European countries as ports of entry to Europe. Media images of obsolete and overcrowded reception structures in Italy and Greece suggested that Southern European countries were still ill-prepared to face the new migration challenge. On the other hand, the crisis refuelled the political resentment of Southern European Member States for having been turned into the "guardians" of the common European border, receiving only some economic and technical support from the European Union and their Northern European counterparts without any real responsibility-sharing.

The crisis of 2015 showed that the perception of the North-South divide in immigration was still alive and kicking in Europe, where Southern European countries appeared as Europe's soft underbelly in contrast not only to Northern Europe, but even to Eastern Europe, an area often described as merciless. Indeed, the tough approach of new Member States has further fostered the idea that migration control does not depend on experience in the feld, but rather on political will – which is lacking in Southern Europe. Therefore, if the tough migration policies of the Eastern countries had prevented an East-West divide to rise to prominence in the scientifc and public debate on immigration control for third-country nationals,2 they might have exacerbated the North-South divide and further increased the blame on the Southern countries.

Finally, in the face of the coronavirus pandemic, the urgency to ensure a migrant labour supply entered the public debate at the very frst stage of the health crisis. The European Commission tried to ensure common guidelines through the adoption of a common list of "critical workers" that had to be granted freedom of movement across EU internal borders and the opening of "green lanes" for agricultural workers within the European Union. In general, Member States' responses somewhat converged. However, there were exceptions that, even if not framed as such in the public debate, essentially corresponded to the North-South divide. For example, three countries in Europe dealt with the pandemic by adopting large-scale regularisation or by clearing out huge backlogs of individual regularisation applications through mass positive resolutions: Italy, Portugal and Greece (OECD, 2021).

Similarly, the refugees feeing from Ukraine seem to have stimulated a similar reaction in terms of migration controls among Member States, also thanks to the application of the Temporary Protection Directive. Actually, a deeper look suggests that a North-South divide might soon re-emerge, although not in the form of Southern toleration and Northern rigour in border controls. Indeed, Northern European countries, such as Germany, have taken the opportunity to actively attract and recruit Ukrainians who are regarded as a much-needed skilled labour force – as happened with Syrians during the 2015 refugee crisis. Instead, Southern European countries like Italy, although hosting a larger Ukraine diaspora, are not actively taking advantage of this asset and the current contingency to fll the relevant labour shortages.3

This brief excursus suffces to show that rethinking the North-South divide is necessary not only for the theoretical reasons explained in the previous section, but also because successive major crises – the Great Recession, the European refugee crisis, the pandemic, and the Ukraine war – have impacted European immigration policies in the new millennium and the diverse responses to these crises require a correction to the usual frames of interpretation.

<sup>2</sup> In fact, though Eastern European countries are new Member States and recent immigration countries, the East-West divide and the North-South divide represented two different "axes of contention" (Hampshire, 2016). While the North-South controversy mostly concerned unwanted migrants from third countries (Freeman, 1995; Baldwin-Edwards, 1997), the East-West dispute essentially addressed the question of free movement, particularly the challenge of Eastern European "free movers" for Western European labour markets (Hampshire, 2016; Favell, 2008).

<sup>3</sup>According to Eurostat, there were around 83,000 Ukrainians with a valid residence permit in Germany and over 230,000 in Italy at the end of 2021 (https://ec.europa.eu/eurostat/statisticsexplained/index.php?title=Ukrainian\_citizens\_in\_the\_EU). Almost one year later, there was an apparent reversal of the distribution of recorded Ukrainian refugees between these two countries: according to the UNHCR, there were over one million in Germany and around 170,000 in Italy at the end of November 2022 (see https://data.unhcr.org/en/situations/ukraine).

#### **1.3** *E Pluribus Unum***: Bringing Like-Minded Scholars Together**

This volume builds on the work of a wide range of scholars who have investigated the logics and routines of action in the feld of immigration control in recent decades with the aim of bringing their valuable, yet still unconnected, work within a single and innovative theoretical framework on migration controls in Europe. Specifcally, the authors use the framework of a migration regime to focus on organisational strategies, structural features, logics and practices, rather than on legal frameworks, to reframe some persisting differences between European migration regimes, and show where the boundaries of these regimes have started to blur.

In addressing the North-South divide, this book takes Italy and Spain as the reference countries of the South, and Germany and the Netherlands as those of the North. While the frst two countries have been regarded as latecomers in developing effective immigration policies, and as the main sources of uncontrolled secondary movements towards other Member States because of their relaxed immigration policies and frequent amnesties, the second pair of countries have been viewed as key shapers of EU migration and asylum legislation and "champions" of migration controls. Moreover, three other countries have been added in individual sections because of their relevance in the specifc policy domains addressed there. One chapter on Portugal examines internal controls, since the country has adopted a wide range of strategies to regularise undocumented migrants, thus displaying the entire variety of options implemented by Southern European countries: general amnesties, amnesties based on conditionality (economic conditions and bilateral agreements), caseby-case regularisation and even "emergency regularisation" to cope with the pandemic. The analysis of the welfare-migration nexus, which highlights the connection between welfare chauvinism and intra-EU freedom-of-movement policies, includes a chapter on the UK since it is the most evident case of welfare chauvinism against mobile EU citizens (including Southern Europeans) and its impact on immigration policies; this ultimately contributed to the drastic decision to leave the EU and rid itself of the EU's internal free movement rules. Finally, the case study of Greece is scrutinised with regard to asylum policies, given that the country was indeed at the forefront during the so-called European refugee crisis together with Italy and Germany; moreover, it has served as an inspirational model provider for the New Pact on Migration and Asylum proposed by the European Commission in September 2020, especially with regard to the option of asylum applications processing at the border.

The book is divided in six thematic sections, where two or three chapters contrast Northern and Southern European countries in a given policy domain. These are Visa Policy and External Controls, the Externalisation of Control, Internal Controls, Labour Migration Policy, the Welfare-Migration Nexus, and Asylum Policy. The chapters presented in these six sections use a variety of methods ranging from the analysis of statistical data and offcial documents to interviews with stakeholders and migrants and, in one case ethnographic research.

The section on External Controls focuses on visa policy, which appears as a key tool of common external controls and is widely acknowledged as a rather successful example of legislative harmonisation in the European Union. Nonetheless, the book's chapters go beyond the legal framework to investigate visa implementation, suggesting a slow blurring of the divides between Southern and Northern European control logics.

In the frst chapter, Irene Landini and Giuseppe Sciortino suggest that migration control policies in Western European states should be considered as an interdependent, yet politically segmented, system. They test this view by analysing two migration policy felds widely different in terms of history and development, i.e., visa and return policy, and comparing the relative fgures across Member States. With regard to visa policies, their results show that, over time, the original Northern model of visa controls has become the widely accepted normative model across all European states today, while policy harmonization and cooperation in return policies and practices have always remained low. The authors do not observe, however, evidence of a North-South cleavage in either of the policy felds. Instead, with regard to return policies they observe a process of the facto convergence, since all EU states have shown to be largely ineffective in removing unauthorised TCNs from their territories.

The second chapter by Federica Infantino uses the case of Italy, one of the countries that has issued the highest numbers of Schengen visas, to shed light on how and why day-to-day implementation practices challenge "national models" as well as the assertion of a divide between Northern and Southern European countries. Her analysis focuses on the entanglements of logics on paper, policy narratives and organisational practices in a context of continuities and innovations. Infantino notes that the logics and practices governing Italy's visa policy are historically distant from the EU model since the "Schengen model" refected the interests and needs of the Northern countries that initiated and designed it. Nevertheless, that distance dissipates at the stage of the implementation, so that national boundaries of organisational action are blurred on the ground, while dynamics of policy change are triggered from below.

The section on the Externalisation of Controls unravels the delegation of control responsibilities to third countries, which has become a key instrument to control unwanted fows to the European Union. Such tasks were often considered to be a priority of Southern European Member States, which tried to "pass the buck" to third countries after having been turned into the guards at the external border of the EU. The two chapters show how Southern European countries were indeed pioneers in this feld and have since been enthusiastically followed by Northern Europe and the EU as a whole.

The chapter by Lorena Gazzotti, Mercedes Jiménez Álvarez and Keina Espiñeira challenges the assertion of the very existence of a structured European externalisation front, demonstrating that the implementation of a specifc border externalisation programme is reactive and inconsistent in nature, driven by dynamics of *ad hoc* reaction to sudden punctual crises. The authors, focusing on Morocco, show how third countries' "migration diplomacy" changed the long-lasting European

North-South divide with respect to externalisation strategies by increasingly involving Northern European countries. In such a context, they stress the tendency of Member States' border agencies to behave like NGOs as they become the actors implementing EU development funding. Against this backdrop, the boundaries between the strategies of Northern and Southern European states blur into the technicalities of delegated cooperation, whereby the agency of a given Member State seems to submit to the functioning of the EU, driven by a clear, yet contradictory, objective to advance securitisation policies.

The chapter by Lorenzo Gabrielli examines the genealogy of practices, logics and organisational strategies within the multilevel policy framework that fostered the development of the Spanish migration regime's external dimension. His analysis focuses on the changing relations between Spain and the EU associated with the policymaking on border externalisation, showing that Spain shifted from a passive receiver of European norms and standards to an active player in European policymaking, fostering changes and new developments in the EU immigration regime. Ultimately, Spain became a model and inspiration for migration policies implemented at the EU level in the 2010s.

Internal Controls represent another major group of control instruments aimed at preventing unwanted settlement, especially in countries where external controls are traditionally weak, such as in Southern Europe. The goal of this section is to analyse different types of internal control logics in the European context, with special attention to regularisations that, more than other migration regulation tools, scholars have used to highlight the "effectiveness gap" existing between Northern and Southern European control regimes, and which have been the object of several confrontations among EU Member States. The chapters of this section show how different European states have to deal with both economic pressure coming from different labour markets as well as with social demands and expectations towards the state. Hence, different strategies, practices and outcomes of internal controls cannot be understood in terms of *laissez-faire* versus strong public policies, and instead have to be explained by mobilising the diversity of social and economic demands.

The chapter by João Peixoto and Jorge Malheiros highlights the fact that the most frequent strategy towards irregular immigration in Southern European migration regimes until the mid-2000s had been the enactment of extraordinary regularisation processes. Afterwards, some countries adopted an ongoing, case-by-case regularisation model. They use Portugal as a reference to develop a comparative analysis and position it at the European level to evaluate the convergence or divergence hypothesis and the blurring of migration regimes' boundaries. The authors conclude that the Southern European migration regime is less homogeneous and exceptional than it is generally presented, and that irregular migration levels depend on economic cycles and the type of economic demand, rather than on the implementation of policy mechanisms facilitating regularisation.

The chapter by Claudia Finotelli addresses the use of regularisations and *ex-post* regulation measures as instruments to produce knowledge on social problems. Using Italy and Germany as comparative examples, the author argues that the function of regularisations should be assessed beyond the dichotomic distinction between "weak" and "strong" migration control regimes in Southern and Northern Europe. Instead, regularisations and *ex-post* regulation strategies should be seen as an instrument to overcome weak internal controls as well as to gain knowledge about the presence of irregular migrants and stabilise the precarious immigrant population.

Gabriel Echeverría investigates two other countries that have often been portrayed as opposite examples when it comes to internal controls: the Netherlands and Spain, with the former as "top of the class" in strict control enforcement and effective migration deterrence, and the latter as an example of weak control measures and inconsistent results. The author challenges the Manichean hypothesis of "good guys" and "bad guys" in migration controls, by showing that countries display very dynamic conduct that may at times converge or diverge with others, depending on the confguration of societies as a whole and the relationships between their subsystems (culture, politics, economics, welfare, etc.). This analysis shows an ambiguous reality – with similarities and differences, degrees of convergence and the persistence of variance – that is complex enough to elude a clear-cut description of diametric opposites.

The fourth section deals with Labour Migration, a feld that has remained the least harmonised in the EU. Labour migration policies have remained relatively distinct within the EU, shaped by the diverse needs of Member States' national labour markets. Insofar as Northern European labour markets are traditionally associated with demand for high-skilled workers, it is generally assumed that in Southern Europe, the demand for low-skilled workers predominates. This section highlights how the logics driving their labour migration regimes in Northern and Southern European countries have begun to overlap, both by balancing restrictions and openings, and by mixing skills-based and low-threshold pathways, albeit to a different extent.

In their chapter, Jan Schneider and Holger Kolb address Germany's slow but steady return to ethnic selectivity and particularistic features in the area of labour migration policy after a decade in which German labour migration policy had moved towards a universalistic regime that applied similar conditions to most thirdcountry nationals applying for admission to the labour market. The authors argue that although human capital remains at the heart of the regulations and institutional settings that govern the process of selecting labour migrants, the factor of the respective country of origin of applicants, which had been regarded as a peculiar trait of Southern European countries' policies, has regained importance in Germany in recent years. These observations run counter to the proverbial North-South divide, suggesting instead unexpected convergence in this area.

The chapter by Camilla Devitt turns to labour migration policies in Southern European countries. Her contribution challenges the common perspective of the North-South divide, typifed by Southern European countries exhibiting a distinct approach to the admission of migrants. By exploring Italian labour immigration policy, the author fnds that the similarities with Northern European regimes have increased since the Great Recession of 2008, with, for example, a more restrictive approach to infows of non-seasonal workers from third countries and a stronger reliance on the free movement of workers from Eastern Europe and non-economic forms of migration for low-medium skilled labour needs. This convergence among allegedly different European migration regimes is explained by the stage of migration, European integration and the impact of the economic crisis.

Finally, the chapter by Jeroen Doomernik, Blanca Garcés-Mascareñas and Berta Güell revisits the debate on the South-North divide on migration regimes by comparing the cases of Spain and the Netherlands with regard to migrant seasonal workers in agriculture, paying particular attention to their situation before and during the COVID-19 pandemic. The authors conclude that differences between the two countries are not that relevant. In both cases, seasonal labour demands were initially covered by newly arrived immigrants, followed by immigrants already in the country; recently, this has been complemented by Central and Eastern European workers who can go back and forth without the constraints imposed by international borders. Moreover, in both countries, the authors see convergence towards major deregulation of the sector, particularly due to the increasing use of temporary agencies.

The section on intra-EU mobility deals with the nexus between immigration and the sustainability of welfare programmes. This has been an object of increasing research interest since the end of the 1970s and has mainly examined the movement of irregular migrants and asylum seekers from Southern European countries to the more generous welfare states of Northern Europe. Yet, this section shows that welfare chauvinism and welfare restrictions are nonetheless not limited to Northern European countries and to traditionally "unwelcome" migrant categories such as asylum seekers or irregular migrants.

The chapter by Alessio D'Angelo critically examines the last few decades of policy and political debates around intra-European migration in the United Kingdom, the key trends that led to the (not so) unpredictable Brexit referendum, and the scenarios that have since then been set in motion with the UK-EU Agreement of 2020. In spite of the strong sense of British exceptionalism that informed debate in the UK, D'Angelo shows that some of the fundamentals underpinning this process have a great deal in common with what we are witnessing elsewhere in Europe, with the stratifcation of (welfare) rights for different categories of migrants being used as a pragmatic – if not cynical – mechanism to regulate entry and settlement. In fact, what at a political and institutional level currently appears as a major rupture within the European framework may end up revealing itself to be part of a wider trend among both Northern and Southern European regimes: the restrictionist reconfguration of the welfare-migration nexus.

Claudia Finotelli reverses the North-South perspective on intra-EU mobility and welfare by shifting attention from the intra-EU mobility of young Eastern and Southern Europeans in Northern Europe to the non-labour-motivated mobility of Northern European citizens in Southern Europe. Using Spain as a reference case, her chapter explores to what extent the generally welcome presence of intra-EU retirees from Northern European countries in Southern Europe had an impact on the welfare provisions to inactive EU citizens. As she argues, the increasing demand for healthcare services by elderly European migrants has triggered unexpected forms of welfare chauvinism in Spain and raised the issue of the healthcare costs related to the presence of intra-EU retirees. This has lead to restricted access to public health care for EU citizens without full residency in Spain, and confrms that the restrictionist turn in the regulation of welfare access to EU citizens has not been limited to Southern European labour migrants in Northern Europe, but can be easily extended to the intra-EU non-labour-motivated mobility in Southern Europe.

The sixth and last section of the book deals with Asylum Policy. Despite the establishment of the Common European Asylum System after the European Council of Tampere in 1999, the North-South divide has been on full display in the feld of asylum over the last two decades, with Southern countries accused of failing to fngerprint people crossing the border irregularly, fostering secondary movements of asylum seekers to other Member States, and providing inadequate reception facilities. Against this backdrop, the section contrasts three countries that have been at the forefront of the European refugee crisis – Germany, Italy and Greece – highlighting, on the one hand, signifcant differences in the degree of institutionalisation of their national asylum regimes, and, on the other, some convergence in the logics of action they have adopted to respond to the increasing arrivals.

The frst chapter by Dietrich Thränhardt challenges the widespread idea of Germany as having an "effcient asylum machinery" by contrasting the "culture of welcomeness [*Willkommenskultur*]" in Germany's complex asylum regime with its bureaucratic ambiguity. Thränhardt shows that during the asylum crisis in 2014–17, the Federal Offce for Migration and Refugees (BAMF) was unable to process all the asylum requests, leading to a large backlog of applications. This notwithstanding, the author argues that slow asylum processing does not seem to have jeopardised Germany's inclusive potential. The courts correct a high percentage of BAMF decisions, with lawyers and volunteers assisting rejected asylum seekers, while the government provides integration courses supported by employers, churches and many volunteers. In the end, most refugees fnd work, learn the language, and become part of the social fabric, demonstrating that inclusive elements can prevail despite an incoherent, complex and somewhat dysfunctional bureaucratic apparatus.

Using a biological metaphor of infancy to adulthood, the chapter by Irene Ponzo illustrates how Italy's frames, strategies and practices concerning asylum have changed over the last three decades. For a long time, Italy perceived itself as a transit country and, as a consequence, allowed and even fostered secondary movements of asylum seekers towards other countries and kept its asylum system underdeveloped. Since 2011, the sharp increase of unplanned infows and the modifcations in the institutional settings where negotiations among Member States occur (the full inclusion of Italy into the Schengen Area and the CEAS) led to the failure of those solutions. The result was that the Italian asylum regime came of age: the country of Italy adopted a new policy frame by acknowledging itself as a destination country for asylum seekers, overcame ad hoc emergency solutions, and joined the Northern European countries' call for more responsibility-sharing. In contrast, the country's weak political-institutional capacity has slowed down the consolidation of the new practices.

The third and fnal chapter of this section, by Angeliki Dimitriadi, deals with the Greek asylum system. She argues that since 2015, Greece has undergone a gradual transformation that has reinforced its role as an external border guard of the European Union, but which has also provided a fertile ground rich in data regarding the policies in place on asylum processing and the reception of irregular arrivals. On the one hand, the European refugee crisis of 2015 resulted in a newly formed reception system, with non-state actors taking on an unprecedented role in offering reception services. Here, Greece has continued in its role as a "student", seeking to provide reception conditions similar to most of its European partners. On the other hand, a differentiated asylum system emerged between land and sea borders, making Greece the only country with two parallel asylum processes. The chapter shows that some of the practices on border controls found their way into the New Pact on Migration and Asylum. This suggests that in this case, Greece has functioned as an inspirational model provider for Europe.

In the concluding chapter we will illustrate the main fndings drawn from the comparative reading of the book chapters. Connecting the book contributions, we will show to what extent the combination of practices and organisation logics in different national contexts blurs the North-South divide into a far more complex and overlapping migration reality. On the one hand, we will explain how the book chapters reveal patterns of policy convergence highlighting the key drivers of similarities between Southern and Northern European countries. On the other hand, we will argue that analyses presented in the chapters confrm the persistence of divergence driven by different types of imperatives and internal constraints.

#### **References**


Zaun, N. (2017). *EU Asylum Policies. The Power of Strong Regulating States*. Palgrave.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Part I Visa Policy and External Controls**

# **Chapter 2 External Controls: Policing Entries, Enforcing Exits**

**Irene Landini and Giuseppe Sciortino**

#### **2.1 Introduction: Do External Control Policies "Converge"?**

The capacity to control geographical mobility across political borders is a key aspect of sovereignty among modern states. Unauthorised movements across borders are consequently seen as a challenge to their very *raison d' être*. Migration has always raised a variety of existential fears: the control of the intra-European labour supply had triggered bellicose concerns already in the period leading up to World War I (Olsson, 1996). The fear of unmanageable "surplus populations" has accompanied the extensive redrawing of Europe's maps and all the unmixing of its empires (Gatrell, 2019). More recently, the (allegedly) inadequate control of European borders has been described as a clear and present danger, a threat to the survival of the European project and even European civilisation itself.

These fears may appear far-fetched, but public opinion reveals a different story. Opinion polls across Western European states have consistently shown the existence of a sizeable, and increasingly easy to mobilise, bloc of voters opposed to further immigration. The deep restructuring of European party systems in the aftermath of the 2015–16 asylum wave provides further evidence for the intrinsic appeal of playing the "immigration card" for a variety of populist challengers.

Because the effectiveness of border controls is seen as an important attribute of sovereignty, the claim that some states are better than others at securing their borders takes on a strongly normative connotation. Global migration control is depicted by some scholars as very uneven, a world in which some states are highly effective in policing their borders, while others are unable to control unauthorised migration (if not even colluding with it). The spectre of a radical difference between Northern and Southern states (with the former "good" at managing migration, and the latter unable to provide effective control) has haunted Western Europe since the very

I. Landini (\*) · G. Sciortino

Department of Sociology and social research, University of Trento, Trento, Italy e-mail: irene.landini@unitn.it; giuseppe.sciortino@unitn.it

<sup>©</sup> The Author(s) 2023

C. Finotelli, I. Ponzo (eds.), *Migration Control Logics and Strategies in Europe*, IMISCOE Research Series, https://doi.org/10.1007/978-3-031-26002-5\_2

beginnings of a distinctly "European" migration policy. Migration control in Western European countries is in fact highly interlinked, and the enduring presence of unwanted migration is often considered the consequence of ineffective action – or even inaction – by the (allegedly) "inexperienced" Southern partners.

In the aftermath of the migration "crisis" of 1989–91, Southern European countries were openly described as both ineffcient and lacking the necessary experience in terms of mechanisms of immigration control and humanitarian protection (Baldwin-Edwards, 1999, 2001; Thränhardt, 2003). They were associated with lax migration and border controls, weak or non-existing asylum programs, and a high degree of public ambiguity towards irregular migration. Northern European countries, in contrast, were considered to have reliable and effcient mechanisms of migration control and refugee protection, as well as little tolerance for irregular immigration (*ibid*). The popularity of such a vision not only derived from widespread stereotypes, but was also politically important, giving priority to the interests of Northern European countries. It cast the countries trying to enter into the Schengen Agreement in the role of unruly pupils that needed to be disciplined (Baldwin-Edwards, 1999, 2001).

In 2001, Baldwin-Edwards described the Mediterranean countries as the "weak underbelly of the EU control system" (2001, p. 23). Southern countries were basically transit countries used by the masses of unwanted migrants to gain a foothold in Europe. Passing through these countries, migrants could prepare to move towards the "real" migration destinations in the North (Baldwin-Edwards, 2001). Southern European migration systems, which were gaining momentum precisely in those years, were thus interpreted as "trouble" for Northern European countries, rather than as independent migration systems in their own right (Sciortino, 2005).

The idea of a North-South divide in the effectiveness of migration control has been, however, contested on several grounds. By comparing Italy and Germany (typically considered the two showpieces of "soft" and "hard" lines in immigration control policies), Finotelli (2009) concluded it was a myth. She documented how the existing differences both in refugee reception and irregular migration fows could be better explained as the outcomes of different inclusion and reception mechanisms, rather than by ineffcient and lax border controls (Finotelli, 2009). Furthermore, as Finotelli and Sciortino (2009) have argued, the functioning of mechanisms for immigration control must be assessed within the historical development of a given country's migration regimes. From a different angle, it has been shown that Mediterranean control systems have been reasonably effective (Colombo, 2012).

The idea that EU Member States may be distinguished according to their control effectiveness is not limited to the idea of a cleavage between Northern and Southern states. It also provides the background for the ever-popular debate concerning the existence of convergence (or lack thereof) among "core" and "outer" members (Meyers, 2002; Toshkov & de Haan, 2013; Hollifeld et al., 2014). In several migration policy felds – labour migration, integration policies and asylum, in particular – many scholars have sought to identify a trend pointing (or failing to point) towards increasingly similar policies and outcomes. Scholars have highlighted how, beyond

country-specifc variations, there exists a set of mechanisms – operating across Western European states – that account for increasingly similar policy outcomes (e.g., Meyers, 2002). EU Member States, (indeed, all industrially advanced states) converge on a very similar approach: "Courting the Top, Fending-off the Bottom" (Joppke, 2021, p. 68).

With regard to national asylum policies, and, specifcally, overall asylum recognition rates by European states, a study by Toshkov and de Haan (2013) supports the convergence hypothesis. In their view, such increasing similarity is linked to the creation and consolidation of the Common European Asylum System (CEAS).1 Even as regards hyped and heated integration debates, some scholars have pointed to a slow convergence towards similar patterns of "civic" integration inspired by repressive liberalism (Joppke, 1998).

A main stumbling block for debates on convergence, however, has been the lack of evidence on increased systematic similarities among the EU Member States. The above-mentioned claims, in fact, have been quickly challenged by contrasting studies. Analysing fve European countries from 1990 to 2016, Consterdine and Hampshire (2020) fnd scarce evidence of a general change in the direction of restrictive (or liberal) labour migration policies. Some scholars have identifed different degrees of integration policy change at the national level, making the dividing line among national regimes more blurred than it appears at frst glance (Finotelli & Michalowski, 2012; Caponio et al., 2016).

In this chapter, we analyse the similarities and differences among EU Member States in two migration policy felds that have not received much analytical attention: visa policy and deportation/return policy. We have chosen to focus on policy felds widely different in terms of history and policy development. Visa policy is likely the oldest and most stringent area of coordination among EU Member States. Deportation/return policy, in contrast, is an area in which, besides ritualised statements, supra-national interventions have been fimsy, if not utterly contradictory. As both visa and deportation/return policies play an important role in the European migration control system, the institutional differences among them are compelling.

#### **2.2 A Critique of the Implicit Conceptual Framing of Debates on Convergence**

This chapter provides an empirical critique of the implicit conceptual frame of increasingly polarised debates on convergence among EU states. This lack of agreement and reasonable dialogue leads to a scenario in which important differences among EU Member States are overlooked. Equally important, the *belief* in the existence of deep differences among Member States plays an important role in European

<sup>1</sup>They also observe that some important national differences in the recognition of applicants from the same country of origin persist.

migration policy making. It supports a very simplifed, and unrealistic, interpretation of the social dynamics of unwanted migration, making more likely the chances of confict and lowering the chances of adopting adequate solutions.

The debilitating weakness of these debates is the (implicit) assumption that the migration control strategies of EU Member States *could*, at least in principle, be fully independent. The development of an 'EU migration policy' is consequently measured out against an (unspoken) ideal in which states renounce such independence to adopt a standardised and uniform "EU" policy. As a category of practice (Bourdieu, 1997), this assumption is clearly signifcant. National politicians are always ready to claim the existence of "national" control goals thwarted by "European" rigidities. "European" politicians and bureaucrats, similarly, are always happy to point to the "egoism" of states as impeding the rational development of an adequate migration control system.

Categories of practice, however, hardly ever work satisfactorily as category of analysis (Bourdieu, 1997). In fact, there are several reasons for defning the European migration regime *not* as a *set* of Member States with a super-imposed authority, but rather as a highly interdependent and yet politically highly segmented *system* (Bommes, 2012). The interdependence among European powers with regard to migration policy is much older than any EU attempt to "regulate" the phenomenon. It is much older than the Union itself. Already during the Huguenot crisis in the 1680s, European kings and princes had tailored their admission choices through careful anticipation and monitoring of what other powers would do (Orchard, 2014). More recently, the so-called Tamil "crisis" of the early 1980s – when asylum seekers were arriving from Sri Lanka to the German Federal Republic and France (and ultimately to Canada) through East Berlin – represents an especially important lesson for the design of any subsequent migration policy (Sciortino, 2017). No European state has ever been able to control its borders autarchically.

In the end, it is simply impossible to analyse migration control policies and their outcomes for each European state individually. Because they are part of a system, the differences and similarities among them can be understood only by looking at the role each state plays in it. If the numbers at the border between Belgium and the Netherlands are low, this is not because Dutch migration controls are somewhat more "effective". It is simply because many other Member States are willing – enthusiastically or not – to apply similar visa requirements to the citizens of some sending countries, preventing transit migration. If some Mediterranean countries experience strong pressures over their maritime borders, this is not because their border controls are "ineffcient". In fact, boats are identifed long before their landing. Boats arrive on Mediterranean shores because the entire EU control system is designed to make some Mediterranean corridors the only available option for those entering Europe to claim asylum. Moreover, international legal protection for refugees makes it impossible for Mediterranean states to push them back once they have arrived on national territory. This scenario stems from the fateful decision, at the Tampere meeting, of binding European migration policy to "full and inclusive application of the Geneva Convention thus ensuring that nobody is sent back to persecution, i.e., maintaining the principle of *non-refoulement* the legal protection guaranteed by the *non-refoulement* principle of international refugee law" (Presidency Conclusions, 1999).

To analyse the development of European migration policy – and the strains it reveals – it is necessary to abandon the rhetoric of "effective" vs. "weak" states, to focus instead on functional and segmentary variations. The importance of such a distinction becomes clear when we acknowledge and confront a feld in which European states have achieved a large degree of supranational integration, i.e., in terms of visa policy (Finotelli & Sciortino, 2013; Nicolosi, 2020), with a feld in which formal harmonisation and cooperation among European states remain very low (return and removal policies).

In both cases, however, careful analysis reveals a similar picture: the differences between Northern and Southern countries are not particularly strong or clear-cut. Most of these differences may be accounted for by functional variations, by the different roles states play within the European system of migration control. As far as visas are concerned, we fnd a slow process of inclusion for Southern (and Eastern) countries in a control mechanism long shaped by the overall control objectives of Northern European countries. In contrast, return and removal policies have been – and largely are – intentionally kept outside of European coordination (De Bruycker et al., 2016). Even in this case, however (and even if the data available are rather spotty), there is no systematic evidence of a North-South dichotomy. If there is a similarity to be detected, is the generalised low level of effectiveness in removing unauthorised third-country nationals (TCNs) across all European states.

#### **2.3 A Tale of Policy Convergence: Short-Term Visas as a Generalised System of Migration Control**

Many of the tools used today by states for controlling mobility are little more than a century old. Consider the case of the travel visa, which had come to be used as a generalised system of migration control only at the beginning of World War I (Czaika et al., 2018). Its salience in European migration history has shifted considerably along the different periods of European migration history.

In the two decades after World War II, Western European migration policies were confronted by the presence of a "surplus population" comprised of approximately 11 million internally displaced persons (IDPs), refugees and asylum seekers. This was further augmented by a "refux" of settlers from Eastern Europe, the Balkans and newly decolonised countries (Peach, 1997). Many European citizens (including many national leaders) had direct experience with exile and forced displacement. The pressures of the Cold War made refugees living proof of the superiority of the "free world".

All these elements contributed to the introduction of a highly liberal regime, anchored in explicit provisions in new national constitutions and in the adoption, in July 1951, of the Convention Relating to the Status of Refugees (Gatrell, 2000; Rinauro, 2009). The Convention provided, although initially only for European refugees after World War II, a clear defnition of *refugee*, sharply differentiating them from migrant workers. It further established the principle of *individual* protection and the binding obligation of *non-refoulement*. The main priority for European migration policy at that time was securing visas to allow as many refugees as possible to leave Europe. Categorised as a "surplus population", their extra-European mobility was seen as essential for the stabilisation of the European continent. Very little attention was paid to regulate new arrivals on the continent, as they were considered rare. With the consolidation of the communist bloc, fows from Eastern Europe were severely curtailed, making the fear of new arrivals in Western Europe quite limited.

The situation changed with the European "economic miracle" of the 1950s and 1960s. In a context of extraordinary economic growth, securing an adequate supply of foreign labour in the form of a low-skilled workforce become a pressing concern (Judt, 2006; Bernard, 2019). This "infux" phase – to use Peach's (1997) periodisation – was characterised, primarily, but not exclusively in Northern European countries, by the very selective use of visa requirements – largely absent, or informally ignored in the case of citizens from certain origin countries, rigorously enforced, even beyond diplomatic agreements, against the citizens of others (Schönwälder, 2001).

This phase ended with the oil shock of the early 1970s, accompanied in Western Europe by the interruption of all active programs for low-skilled labour recruitment (Bernard, 2019). With the adoption of an increasingly restrictive approach, several countries experimented with the use of visa requirements as a tool to prevent unwanted migration. In the 1980s, visas become a central element in migration control, targeting fows of potential asylum-seekers. Two factors converged to make Western European public opinion increasingly adverse to refugees. The frst was the sharp increase in the number of asylum seekers. From 1970 to 1999, the number of asylum applications in Western Europe increased dramatically, from 15,000 to 300,000 a year (Hatton, 2004; Van Mol & de Valk, 2016). The second was the diversifcation among places of origin, with a growing number of non-European asylum applicants. They were often blamed of not being "real" refugees but rather migrant workers in disguise, trying to compete unfairly with natives in the national labour markets or abusing national welfare systems (Van Mol & de Valk, 2016; Sciortino, 2017).

Given the protection granted by the international refugee regime (especially by the *non-refoulment* clause), the prevention of unwanted fows of asylum-seekers requires barring them from arriving on the territory of a state in which they could claim protection. The introduction of visa requirement was an especially convenient control tool. In the Western European context, Germany was the frst country to experiment systematically with visa requirements for the prevention of unwanted migration fows of Bangladeshis, Indians, and Sri Lankans in 1980 and Ethiopians in 1982. A few years later, France and the United Kingdom followed suit. Such decisions produced, in the short term, the required effect – the number of asylumseekers decreased (UNHCR, 2011).

Very quickly, however, the limits of country-based visa policy became evident. The emblematic case occurred in Germany between 1980 and 1985. When the Federal Republic of Germany (FRG) introduced visa requirements for Bangladeshis, Indians, and Sri Lankans, the DDR began to encourage potential asylum seekers to fy visa-free to East Berlin, from where the potential applicants could easily reach the FRG (since the city was still considered a single administrative unit by West German laws). The FRG was able to contain the arrivals from those areas only when, in exchange for sizeable amount in loans, the DDR agreed to introduce similar restrictive actions. In short, since its very beginning, the effectiveness of visa policy for preventing the arrival of asylum-seekers was strictly contingent upon the willingness of neighbors to participate in the action.

Such actions were at the centre of intergovernmental cooperation in the feld of Justice and Home Affairs, starting with the Schengen Agreement in 1985 and the following Convention in 1990. The original signatory states – only Belgium, France, Germany, Luxembourg and the Netherlands – were, unsurprisingly, the largest receivers of asylum-seekers in Europe (Hatton, 2004; Van Mol & de Valk, 2016). The Schengen Agreement, considered nowadays a key milestone in establishing an internal market with the free movement of persons, was an objective consolidated by the Treaty of Maastricht in 1992. These actions have made possible the abolition of internal border controls for all nationals of Europe's Schengen Area, today, all EU nationals. For our purposes, however, the Schengen Agreement represents, above all, the establishment of a system of external border control, common to all states adhering (and desiring to adhere) to the Schengen Area. Members and prospective applicants have been required to introduce visa requirements for the citizens of several non-European states (TCNs) and to accept a collective procedure for selecting those that could enjoy visa-free travel. In addition, states who wished to adhere to the Schengen Agreement were required to introduce sanctions for all the carriers transporting irregular migrants, adopting *ad hoc* asylum policies, strengthening border controls, developing more severe measures against irregular migrants in the national territory, and contributing to a common dataset of all detected irregular migrants (Sciortino, 2017).

This set of measures was a reaction to an important bifurcation existing at the time among Western European states. Because visas were strongly associated with the prevention of asylum-seekers, Northern European states, at the time, the nearly exclusively targets of asylum applications, were imposing many visa requirements. Conversely, Southern European countries, where the number of asylum-seekers was negligible, maintained many visa-free agreements (Finotelli & Sciortino, 2013). Most of the migratory fows to these countries consisted of seasonal migrant workers who played an important role in supporting national economies, especially with agricultural labour, and care work in households and elderly people (Cvajner, 2012; Sciortino, 2017).

From the point of view of the original Northern signatories, the highly discretional inclusion of Southern states in the Schengen system was to be balanced by their willingness to participate in the control strategies of Northern European countries. One of the most feared side effects of the pact was the possibility of potential asylum seekers fying or landing in Southern countries and subsequently moving more freely across all Schengen states, especially those in the North. The German delegation was particularly explicit about these concerns, asking for an additional annex to the Agreement, in which the abolition of internal border controls was conditioned to the introduction of visa requirements for TCNs and other compensatory measures. These measures were presented as necessary for safeguarding internal security and the strengthening of European cooperation against unauthorised migration and asylum fows (Finotelli & Sciortino, 2013; Paoli, 2018).

The objective of integrating neighbouring countries into the prevention of the arrival of asylum-seekers was complemented by the Dublin Convention (1990, then modifed in 2003 and 2013). The Dublin Convention, meant to prevent what was dubbed "asylum-shopping", was designed to make countries with external borders responsible for the management of asylum-seekers who, unable to receive a visa, would try to reach the territory of the EU.

The Schengen Agreement and the Dublin Convention provide clear evidence that the development of the European visa regime may be considered an outward diffusion process governed by the migratory interests and control goals of core Northern European countries. Southern European states – and much later, Eastern states participating in the Eastern Enlargement – have progressively shaped their control policies and practices in accordance with these goals.

This process did not occur without resistance. Indeed, Southern countries have been often reluctant to accept the Northern model, especially because the fows of irregular migrant workers to these countries have played an important role in supporting national economies and welfare (Sciortino, 2017). Nevertheless, being an inter-governmental initiative, insider Northern states were able to exert pressure on Southern candidates, pushing them to introduce stricter visa restrictions as a precondition for participation in the Schengen process (Sciortino & Finotelli, 2013; FitzGerald, 2019). The Southern expansion of the Schengen system started with the participation of Italy in 1990, Portugal and Spain in 1991, and Greece in 1992 (Paoli, 2018). Joining the Schengen "club" required introducing visa requirements for countries that had a long history of unencumbered travel, such as the countries of the Southern rim of the Mediterranean (for instance, the case of Italy for citizens from Tunisia and Turkey) and those with whom they had historical and colonial ties (the case of Spain and many Southern American states).

In 1992, the Treaty of Maastricht reaffrmed the key role of intragovernmental cooperation among European states on migration policy, locating it within the third pillar (Justice and Home Affairs) of the Treaty on European Union (TEU). In 1995, when seven member countries established effective border-free travel among them, Regulation No. 2317/1995 of the European Council introduced a frst common list of 101 countries whose nationals were required to obtain a visa to enter the EU. The Amsterdam Treaty, in 1995, provided, in one of its protocols, for the transfer of the Schengen *acquis* into the legal and institutional framework of the EU (Peers, 2000). In simple terms, the process of visa harmonisation was further strengthened by taking on a supranational shape. Namely, states and perspective members are now formally bound to implement Schengen rules as part of the pre-existing body of EU law that any applicant is obliged to accept.

A further fundamental step took place in 2009, with the adoption of the Visa Code by the Regulation of the European Council No. 810/2009. The Code systematised the visa application procedure, by setting out the detailed procedures and conditions for issuing short-stay visas for visits to the Schengen Area and airport transit visas. In February 2020, the New Visa Code entered into force. *Inter alia*, the new code defnes visa requirements as a potential bargaining chip in gaining collaboration from origin and transit countries for the readmission of third-country nationals illegally present on the territory of the EU (i.e., the "paradigm of conditionality", cf. Nicolosi, 2020, p. 471). In short, if the origin and transit countries do not collaborate in in re-admitting their citizens illegally present on the territory of the EU, third countries could be "punished" with restrictive measures, *in primis* concerning the issuing of visas (Nicolosi, 2020).

#### *2.3.1 Patterns of Short-Term Visas Issued by European States*

The trends in granting short-term visas (STVs) by the various Schengen states does not provide any evidence concerning the existence of a North-South divide in control practices. If such a divide existed, we should expect the older Schengen members to have a strict line on granting STVs, providing fewer STVs and turning down more applications, and the Southern countries would be expected to adopt a more relaxed line, refusing fewer applications and granting more STVs. None of these assumptions has been confrmed (Finotelli & Sciortino, 2013).

First, the majority of STVs granted by EU countries are released by a mix of older and newer members, notably France (39,124,476), Germany (30,776,452), Italy (23,736,365), Spain (19,963,026) and Poland (13,419,190). Visa practices refect a variety of geopolitical, economic (tourism and trade), and historical considerations, rather than different attitudes towards migration controls (see Table 2.1).

Furthermore, the available data show the existence of signifcant differences among Schengen states in terms of the percentage of applications for STVs that are turned down (see Table 2.2). Nevertheless, the differences are not between North and South, but rather East and West. High rates of rejection defne the visa practices of Western European states, both Northern and Southern. As Finotelli and Sciortino have shown for France and Belgium, the high rate of rejection must be understood in the context of these countries' special role in Africa, a continent for which rejection rates are systematically higher (Finotelli & Sciortino, 2013). More recently, both Malta and Portugal have signifcantly increased their visa rejection rates, likely in the context of the post-2015 migration "crisis".





Percentage of short-term visas not issued by round of Schengen membership, 2004–2020

#### **2.4 Return and Removal Policies: Failed Convergence and the Inconsistency of the North-South Divide Argument**

The "return" (voluntary or forced) of unauthorised foreign residents to their country of residence (or transit) is an important tool for preventing the settlement of an unwanted fow of asylum seekers and migrants (Coutin, 2015; Lindberg & Khosravi, 2021). Sovereign powers have always used political power to modify the composition of populations, often resorting to deportation and expulsion (Lindberg & Khosravi, 2021). Across the developed world, a striking fact is that, despite ostensibly great institutional and infrastructural efforts to remove irregular migrants and "failed" asylum-seekers, the actual returns are rather limited (Lindberg & Khosravi, 2021).

The term "removal" is used at the EU level to refer to what is commonly defned (in national policies) as "deportation" (De Genova, 2002). The expression "forced return" is also understood as synonymous with "removal", especially in the EU political and legal context. Regardless of the term used to defne it, removal is a specifc form of return policies and practices. At the EU level, the 2018 Return Directive (henceforth, RD or the Directive) defnes return as " […] the process of a third-country national going back – whether in voluntary compliance with an obligation to return or enforced – to his or her country of origin, or a country of transit in accordance with Community or bilateral readmission agreements or other arrangements, or another third country, to which the third-country national concerned voluntarily decides to return and in which he or she will be accepted". Namely, the Directive distinguishes between voluntary departure and removal.2 According to the same articles, EU Member States shall take all necessary measures to enforce the return decision when no period for voluntary departure has been granted or when the obligation to return has not been complied with within the period granted for voluntary departure.

Return policy is one of the most controversial components of the Common European Asylum System (De Bruycker et al., 2016; Carrera, 2016). The European Council has repeatedly stressed, with particular intensity after the 2015 refugee "crisis", the need for increased supranational harmonisation (EuroMed Rights, 2021). However, despite several attempts to increase policy harmonisation and/or intergovernmental operational cooperation among Member States over time, these have remained low. Removal is still largely a preserve of each state (Giuffré, 2015; Carrera, 2016; De Bruycker et al., 2016; Lindberg & Khosravi, 2021).

The very frst attempt to move toward greater harmonisation of state return and removal policies and practices at a supranational level occurred in 1995, with the

<sup>2</sup>Voluntary departure refers to the compliance with the obligation to return within the time-limit fxed for that purpose in the return decision (Article 3(8) of the RD). Removal (or enforced/forced return) is the enforcement of the obligation to return, namely the physical transportation of unauthorised residents out of the Member State (art. 3(5) and 8)*.*

Treaty of Amsterdam (TOA). The Treaty conferred express power to the European Community (EC, today the EU) to address the issue of "illegal immigration and illegal residence, including repatriation of illegal residents" (Article 63(3) (b) of the EC Treaty). Since the very beginning, any attempt to use such power has turned out to be an "enduring *punctum dolens*" (Giuffré, 2015, p. 284). In May 1999, the Justice and Home Affairs Council attempted to manage the controversy between the Commission and the Member States, by accepting the provision that the EU and its Member States share the responsibility of closing readmission agreements with third countries (Giuffré, 2015; Carrera, 2016).

Further efforts towards this policy harmonisation were undertaken at the Justice and Home Affairs Council on 28–29 November 2002 with the adoption of the Return Action Programme. The Programme seeks to enhance operational cooperation among the readmission practices of the Member States, for example, introducing the systematic exchange of information among Member States and common training programs for return offcials. The goal of greater policy harmonisation has been rather left in the background (Cassarino, 2010; Giuffré, 2015; EuroMed Rights, 2021). The most substantial effort to implement the Return Action Program is the 2008 Return Directive. It clearly distinguishes voluntary from forced return, and it succeeds in developing common minimum standards and guidelines on return. These include: the principle of voluntary departure (i.e., a general rule that a "period for voluntary departure" should normally be granted to irregular TCNs); a minimum set of basic rights for irregularly staying migrants pending their removal, including access to basic health care and education for children, etc., and a limit on the use of coercive measures in connection with the removal, based on the principle of proportionality. Nevertheless, most of the other purposes of the Return Action Programme, which should have led to greater operational coordination among Member State practices, have remained unaddressed (Coleman, 2009; Cassarino, 2010; Giuffré, 2015).

In response to the growth of refugee infows to Europe following the 2015 refugee "crisis", the European Commission has put forward several new measures in the feld of return, such as a Recast of the Return Directive (EuroMed Rights, 2021). This brings a series of changes to the 2008 Directive, amongst which the most important is a new connection between return and asylum policies. Namely, under proposed Article 8(6), states shall issue a return decision (voluntary departure or removal) immediately after adopting a decision ending a legal stay, including a decision refusing refugee or subsidiary protection status. Additional novelties are the introduction of some limits to the applicability of voluntary departure (draft Article 9(1)) and of a short time period (5 days maximum) for refused asylum seekers to lodge an appeal against a return decision (the RD does not regulate the timelimit for appealing return decisions).3

<sup>3</sup>Blamed for breaching of Article 13 of the European Court of Human Rights (ECHR or ECtHR) since a fve-day period is usually too short for preparing an appeal, so it would render the remedy inaccessible in practice.

The topic of return is also addressed by the New Pact on Migration and Asylum ("the Pact"), released in September 2020, together with fve accompanying legislative proposals (Jakulevičienė, 2020; Moraru, 2020; Vedsted-Hansen, 2020; EuroMed Rights, 2021). One of its goals is increasing the returns of irregularly staying TCNs from the EU, by means of four main instruments. The frst is the appointment of a Return Coordinator within the Commission DG HOME, supported by a Deputy Executive Director for Return within Frontex and a network of high-level representatives (Moraru, 2020). These new positions should contribute to enhancing coordination, cooperation and consistency among domestic return practices as well as providing for clear monitoring tasks – e.g., accessible appeals mechanisms, special protection for vulnerable groups and independent monitoring mechanisms during the return procedures (*ibid*).

The second instrument is the connection between asylum and return procedures, following on from the 2018 Recast Return Directive. According to articles 53 and 54 of the Asylum Procedure Regulation,4 asylum application rejection should be issued within the same administrative act with a return decision, in both border and ordinary return procedures. In addition to that, articles 40(i) and 41a(5) of the Asylum Procedure Regulation link the detention of asylum seekers to pre-removal detention during border procedures (Moraru, 2020).

The third set of instruments envisaged is novel screening procedure and a mandatory return border procedure, to prevent unauthorised entry into the EU and accelerate returns (Jakulevičienė, 2020; Vedsted-Hansen, 2020). These are applied to both asylum seekers – requesting international protection at border crossing points without fulflling entry conditions – and irregularly entering third-country nationals – i.e., apprehended in connection with unauthorised crossing of external borders, disembarked following Search and Rescue (SAR) operations (*ibid*).

Finally, the Pact introduces a novel instrument, the "return sponsorship", as a form of solidarity cooperation among the Member States (Moraru, 2020; Milazzo, 2021). Under this new scheme, a Member State commits to support returns from another one (Art. 45(1) (b) of the Regulation on Asylum and Migration Management).5 The scheme also implies logistical, fnancial, and counselling help (Art. 55) provided by the supporting Member State (*ibid*).

Both the 2018 Recast Directive and the 2020 Pact have received much criticism by, as they raise several concerns regarding: human rights violations, violation of the right to asylum and principle of *non-refoulement* and the measurement of "effectiveness" of returns (Moraru, 2020; EuroMed Rights, 2021). At the time of writing, the negotiations concerning the approval of the Recast Directive are still ongoing. The Council reached a partial agreement, but the European Parliament is working within its Committee on Civil Liberties, Justice and Home Affairs Committee on negotiating its position (*ibid*).

<sup>4</sup> one of the fve main legislative proposals accompanying the Pact.

<sup>5</sup>Another accompanying legislative proposal.

#### *2.4.1 EU Return Policies: A North/South Divide?*

Given the highly fragmented approaches to returns in the EU (De Bruycker et al., 2016), it is diffcult to see a process of diffusion or policy convergence in the area of return policy. It is likewise diffcult to ascertain the possible existence of a North-South divide. A main problem is the lack of reliable and suffciently detailed comparative data. Although Eurostat has recently been publishing some data on returns, they are very uneven, making estimates diffcult. Nevertheless, through analysing these data, we can draw some useful conclusions.

Relying on the scant available data, we do *not* observe any systematic evidence of a North-South divide. Starting from the number of TCNs ordered to leave, we can see in Table 2.3 that the situation has been relatively stable between 2011 and 2019. More specifcally, until 2015, it followed an alternating pattern of increases and decreases, yet without any remarkable changes. In 2015, with the refugee "crisis", many European states announced the intention to strengthen cooperation with third countries on the identifcation and readmission of returnees in order to increase their increase return rates (EuroMed Rights, 2021). Notwithstanding, the general trend is still far from impressive. The total orders to leave increased only slightly between 2014 and 2016 (from 472,555 to 486,150).

A similar scenario emerges when comparing 2011 with 2019 – the two temporal extremes of the period under observation. The number of TCNs ordered to leave is higher in 2019 but only to a limited extent (5%). Unsurprisingly, in 2020, the number of removals decreased sharply. In short, the available data do not support the North-South divide hypothesis. The countries who issued the highest numbers of orders to leave in such a period were from both Northern and Southern Europe: France (916,310), Germany (455,225), Spain (445,275), Belgium (333,275) and Italy (280,760).

Similar considerations apply to the volumes of removals carried out by each EU state.6 The overall number of removals for the EU 28 area does not reveal any impressive increase in the aftermath of the refugee "crisis", growing only by 9% between 2014 and 2019 (from 69,712 to 76,259). In 2020, because of the COVID-19 pandemic, removal operations plummeted remarkably, decreasing by 45% compared to the previous year. However, the number of removals carried out in the EU 28 area had diminished already in 2019, by 10% (see Table 2.3). Again, the data on actual removals do not support a North-South cleavage. Among the top fve countries in the number of enforced returns implemented between 2014 and 2020 (looking at the totals in Table 2.3), we see a mix of Northern and Southern countries: Germany (115,737), France (75,010), the United Kingdom (69,862), Spain (65,215), and Italy (31,230).

An overall measure of effectiveness would certainly be the ratio between the total number of TCNs expelled and the orders to leave issued. The availability of data allows only for measuring it very roughly: many orders to leave are issued

<sup>6</sup> In this case, data constraints again limit our analysis to the period between 2014 and 2020.


37




(continued)

**Table 2.3** (continued)


(continued)


For data about the UK: Immigration Statistics Yearbooks (UK government's public website). For data about Germany: German Ministry of Interior and Federal Police Reports

For Germany, only data about the *Abschiebung* (deportations) have been considered

All other missing data were not available in Eurostat years before the actual return. We have considered the time interval between 2014 and 2020, since we have data for both the total number of TCNs expelled and the orders to leave issued (see Table 2.4). Our estimates are consequently indicative. Some of the results, however, are worth discussing.

First, none of the countries considered is particularly effective in carrying out the enforced return operations. Even Germany, which stands out for the absolute number of persons forcibly returned, actually manages to remove less than a quarter of the TCNs ordered to leave. Second, we do not fnd any evidence of a North-South dichotomy. The most "effective" states are Germany and the United Kingdom (both remove 24% of the TCNs ordered to leave), Spain (23%), Italy (16%) and France (11%). The existing differences among them, moreover, seem to refect the composition of undocumented populations – some nationalities being much more diffcult to return than others – rather than the effectiveness of control procedures and infrastructures. Across Europe, the same factors – the implementation challenges of many readmission agreements, the low levels of collaboration among sending and transit countries, states' administrative and budgetary constraints, and the enduring strength of embedded liberalism – operate in favour of low structural effectiveness.

A fnal consideration is needed. We are aware that, within this broader trend of low effectiveness, substantial differences exist, at both the national level and the EU 28 area, in the percentages of third-country nationals effectively returned depending on nationalities. A relevant report issued by the European Parliament in 2020 and the results of a research carried out by the European Migration Network in the same year clearly show that, between 2009 and 2019, the top 5 nationalities considered as "easier" to be returned are (alternately yet constantly) Morocco, Ukraine, Albania, Afghanistan and Algeria (Crego & Clarós, 2020; Vogel, 2020). This trend suggests a more collaborative approach to return procedures on the part of these fve countries, as well as the reality of the situations in some other countries preventing the enforcement of return decisions issued against their citizens (e.g., Syria). In spite of that, as we have shown along this section, even when considering the most successful cases like the ones mentioned above, the overall return rates and ratio TCNs expelled/orders to leave issued remain overall remarkably low.


**Table 2.4** Ratio between the total number of TCNs forcibly returned and the orders to leave issued, for the top 5 countries in the number of enforced returns, in the period 2014–2020

Source: Own elaboration from above data

#### **2.5 Conclusions**

In this chapter, we have provided an empirical critique of the implicit conceptual frame of debates on convergence and divergence among EU states regarding migration policies. The key assumption underlying such debates is that EU Member States are essentially *independent* units, with their own admission and control challenges. This widespread belief undoubtedly plays an important role in European migration policy making. It is, in fact, a main "category of practice" (Bourdieu, 1997). The problem is that, very often, such a category of practice is turned into a category of analysis, becoming a stumbling block to the interpretation of reality.

We suggest, on the contrary, that migration control policies in Western European states have always been highly interdependent. From the Huguenot crisis of the 1680s to the so-called Tamil "crisis" of the early 1980s to the Belarus-EU border crisis of 2021–22, it is simply impossible to understand the control dynamics based on the idea of a set of essentially independent units, subject to the super-imposed authority of the EU. We should consider European migration controls as an interdependent, yet politically highly segmented, system (Bommes, 2012). Hence, it is diffcult to meaningfully compare the policies and their outcomes of each European state individually. Because they are part of a system, the differences and similarities among them can be understood only when looking at the role they play within it.

This systematic confguration emerges clearly in the case of visa policies. Initially using inter-governmental powers, and, subsequently, the supranational mechanisms at their disposal, a group of core, mainly Northern European states, has managed to progressively impose their migration control goals upon reluctant Southern European states. As a result, the original Northern model has become the widely accepted normative model across all European states today, formalised in the New Common Visa Code. Turning to policies to actual practices, we have seen in Sect. 2.3 that no signifcant North-South differences exist in terms of STVs granted. The same is true as far as the numbers of not issued STVs are concerned.

Existing differences between North and South Europe have to be read in light of this generally homogenous background. As a fact, the duty of monitoring Europe's "external" borders is thrust upon only a few states, i.e., the Southern Mediterranean – and, increasingly, Eastern – ones. As the requirement to obtain a visa prior to arrival in Europe makes reaching Northern European countries by fight far more complex and costly for many TCNs, they often opt to enter Europe irregularly using Southern (or Eastern) states as entry gateways. Such a role implies high costs for Southern and Eastern states, both fnancially and organisationally (Giordano, 2015; Italian Ministry of Defence, 2022).

At the same time, Southern countries often tend to "turn a blind eye" (Giordano, 2015, p.25) toward irregular migrants who refuse to register and apply for asylum in the frst country of arrival (Following the Dublin Regulation), as they prefer to ask for asylum in Northern Europe. Looking at the offcial data about asylum requests, we can see that between 2014 and 2016 Sweden hosted the highest number of refugees per capita (2359 per one hundred thousand inhabitants against 254 per one hundred thousand inhabitants in Italy, in 2016). Moreover, in absolute terms, Germany received one in three asylum requests of the total in Europe, in the same years (European Parliament, 2022; UNHCR, 2022).7

As regards return policy, we have shown how the scenario is highly fragmented and atomised. Strong coordination could actually be very useful, for example, in exerting unifed pressure on sending and transit countries. The actual interdependence is, however, quite low and the barriers to the development of joint efforts quite substantial. While the amount of available data is overall small, these seem to point to two important considerations. The frst is that there is no evidence of a North-South cleavage (as seen in Sect. 2.5). The second is that a process that we could defne as "converge" is at play. Namely, all EU states (also including the United Kingdom) have shown to be largely ineffective in removing unauthorised TCNs from national and European territory.

#### **Appendix A Note on Data**

For visa policy, we have used data about short-term visas issued by the embassies and consulates of EU states between 2003 and 2020. These data are available on the websites of the European Council and the European Commission, DG Home Affairs.

Data on forced returns (used by Eurostat as a synonymous with "removal") are much more diffcult to access and of lesser quality. Therefore, a few clarifcations are in order.

A main problem is that Eurostat does not provide data on enforced returns for Germany and the United Kingdom. We have been forced consequently to use alternative sources: the Annual Reports issued by the German Federal Police, the Annual Migration reports published (since 2014) by the German Ministry of Interior, and the Immigration Statistics Yearbooks published by the UK government.

The collection of German data has been particularly diffcult. The German Residence Act distinguishes two different categories of "forced removals": *Abschiebung* (typically translated as "deportations") and *Zurückschiebung* ("forced removals"). We have chosen to focus on the data about deportations only (*Abschiebung*), since it corresponds to the concept of removal as spelt out in Art. 3(5) of the 2008 RD, i.e., physical transportation out of the Member State (when no period for voluntary departure has been granted or when the obligation to return has not been complied with within the period for voluntary departure granted, usually 6 months in the German legislation) (Sect. 58 of the German Residence Act). By contrast, the *Zurückschiebung* is primarily linked to the act of illegally crossing the

<sup>7</sup>To be sure, this was also linked to the former chancellor Merkel's decision in August 2015 to take in an unlimited number of migrants (especially from Syria, Afghanistan, and Iraq), as the death of hundreds of refugees making their way to Europe in 2015, sparked outrage in Germany and among the international community.

national border by TCNs rather than the illegal permanence in the country after an order to leave.

A further major problem with Germany is that the number of deportations listed in the Reports analysed include the so-called "Dubliners", the transfers to other EU or Schengen Member States under the Dublin procedure. German reports print the total numbers of deportations *and* the relevant percentage of Dubliners for each year since 2011. We have consequently calculated the number of deportations for TCNs (i.e., removals according to the 2008 RD) ourselves.

#### **References**


#### *Documents Cited*


Communication to the European Parliament, the European Council and the Council, Progress report on the Implementation of the European Agenda on Migration, COM (2019) 481 fnal, 16 October 2019.

International Committee of the Red Cross. (1983). *Basic rules of the Geneva conventions and their additional protocols*. International Committee of the Red Cross.

Regulation (EU) 2019/1155, Art 25a.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 3 Practices of External Control: Is There a North-South Divide?**

**Federica Infantino**

In the making of the European Union, external migration and border control is a key dimension. The Schengen visa is a policy instrument of external control that has constituted a *conditio sine qua non* for the achievement of free movement (Groenendijk, 2020). Because it aims at controlling migration at a distance, the Schengen visa policy can be characterized as a Europeanized instrument of the old "remote control" strategy (Guiraudon, 2003; Fitzgerald, 2019). The introduction of visa requirements to a common list of countries was the frst step towards the Europeanization of visa policy. Uniform lists pushed the European border in countries of departure while allowing for the removal of inter-state border checks. However, visa policy implementation remains a national issue since national consulates are those responsible to issue the Schengen visa. In doing so, they carry out the fltering work of borders. Just like borders, visas categorize, identify and flter between undesirable and desirable candidates to mobility. Understanding the implementation of visa policy inevitably gains analytical salience. The policy and legal frameworks are uniform: The list of countries whose nationals are submitted to visa requirements as well as the conditions and procedures to issue Schengen visas. However, national consulates put visa policy into practice. Given that, a series of questions might be raised. How do national practices translate EU policy on paper? What are the determinants of cross-national differences in policy practice? Do practices of external control support the thesis of a North-South divide within the EU?

To reply to those questions, this chapter builds on the case of Italy, a southern European country, and puts forward that despite the striking differences in the logics and practices governing Italy's pre-Schengen visa policy, a series of adjustments have been emerging, since Italy's frst steps towards its participation to the Schengen

F. Infantino (\*)

Migration Policy Centre, European University Institute, Fiesole, Italy e-mail: federica.infantino@eui.eu

<sup>©</sup> The Author(s) 2023

C. Finotelli, I. Ponzo (eds.), *Migration Control Logics and Strategies in Europe*, IMISCOE Research Series, https://doi.org/10.1007/978-3-031-26002-5\_3

Agreement. The logic of external migration control has penetrated Italian visa policy. Also, practices put that logic into action. I use an anthropological perspective on public policy (Wedel et al., 2005) that de-masks the framing of policy issues to show that, in regard to migration control and security aspects, the Schengen visa policy "Europeanizes" the original Schengen countries model, one that sees visa policy as an "external mechanism of immigration control" (Brochmann & Hammar, 1999) and a means at managing the risks for the security of contracting parties. One European model exists, and it derives from some (Northern) European countries. However, the analysis of the entanglements between logics in texts and actual practices allows for putting forward the national (North European) infuence on the European regime of external control while arguing that national boundaries of organizational action are blurred on the ground.

The street-level view brings insights into the migration regime perspective adopted in this volume, for it focuses on "the continuous repair work through practices" and "the effects of norms in contexts" (Sciortino, 2004, p. 32–33). I adopt a comparative perspective to show that in the feld of the EU external control, day-to-day responses to understandings and narratives about migration, visas and "risks" craft a specifc practice regime which triggers policy change from below and blurs North/South boundaries. This analysis of implementation practices builds on the scholarly literature which considers policymaking to be ubiquitous (Palumbo & Calista, 1990), recognizes bureaucratic discretion as inherent to law application (Brodkin, 1987; Dubois, 2016; Lipsky, 1980) rather than seeing a dichotomy between law and discretion according to which more law entails less discretion (Emerson & Paley, 1992; Pratt, 1999). Therefore, this analysis focuses on the social limits to the uses of discretion, the socially constructed perception of appropriateness in organizational settings (March & Olsen, 1989), the sense-making that underlines the understanding of the purposes to which organizational action is bent (Hawkins, 1992), and the ways in which knowledge and skills required to put policies into practice are acquired (Feldman, 1992; Yanow, 2004) and shared among trans-national community of practitioners (Wenger, 1998).

The chapter proceeds as follows: frst, by building on historical perspectives about visas introduction, I show how the logic of external control penetrates Italian visa policy, a process that is induced by the Schengen Agreement. Then, I continue by pointing out the roots of the current understandings of the issues at stake in issuing Schengen visas, because those understandings underline the EU legal framework. By building on the comparative analysis of implementation practice, I show that, due to the interactions of implementers on the ground, novel Italian work routines translate logics that are well-established for those states that have contributed to the drafting and crafting of the original Schengen framework, on which the EU model builds. Finally, I discuss the contribution of this case study to the wider literature that is interested on policy convergence and policy change within the frame of the European integration.

#### **3.1 The Logics of External Control: From Schengen to Italy**

Several large-N analyses of visa introduction and visa lifting at a global scale have put forward the lack of asymmetry between low-income and high-income countries (Neumayer, 2006), although signifcant regional variations (Czaika et al., 2018) that have led to the defnition of a global mobility divide (Mau et al., 2015). In the case of Schengen visa policy, Meloni (2006) identifes restrictiveness as its main characteristic. Bigo and Guild (2005) put forward that the lifting of inter-state frontiers to create the Schengen Area has been coupled with the strengthening of external borders and the displacement of control in countries of departure and before the actual arrival on the territory. The making of the Schengen Area beginning in the 1980s has been characterized by the idea that the lifting of inter-state frontiers caused a security defcit to be compensated with more external control (Bigo, 2016). The restrictiveness of visa policy and its understanding as an external mechanism of migration control represent a policy change only for some Southern countries among which Italy. Pre-Schengen Italian visa policy can be defned as relaxed. Until 1990, Italy kept 78 countries visa free.1 Interests in tourism and an unobtrusive Mediterranean foreign policy led to the avoidance of strong controls over temporary entries (Meloni, 2006; Sciortino, 1999). In 1990, the new Act *Legge Martelli*<sup>2</sup> marked a shift towards external control by introducing a series of measures such as the obligation to stamp the passports of non-EEC nationals entering the territory of Italy, the collection and storage of data of the individuals crossing the borders, yearly quotas for new entries, and new criteria for visa introduction that resulted in a restrictive visa policy. The new criteria were based on two specifc logics namely migratory and security risks, two kind of concerns that came to terms with the formerly dominant foreign affairs and economic interests. Migration control and the association of migration with crime started to occupy the scene of visa policy. According to those novel views, visa restrictions needed to be introduced towards the population of migrant-sending countries and towards the country of origin of those immigrants who were sentenced for drug dealing in the preceding 3 years. The Italian immigration policy was too liberal, not focused on external control, therefore un-European. Conformity to external control was the essential condition to join the European club (Sciortino, 1999). Such a policy change exemplifes the audience-directed nature of border control (Andreas, 2011).

In the case of the so-called old immigration countries, mainly Northern European, such as the countries that have originated the Schengen process (Belgium, France, Germany, the Netherlands and Luxembourg), the restrictive visa policy is not induced by the making of the Schengen Area. Let us take the case of France. In

<sup>1</sup>Circolare del Ministero dell'Interno, 19 August 1985, n. 559/443/225388/2/4/6 reproduced in Nascimbene (1988), 221.

<sup>2</sup>Legge 28 febbraio 1990, n. 39 (Legge Martelli) Conversione in legge, con modifcazioni, del decreto-legge 30 dicembre 1989, n. 416, recante norme urgenti in materia di asilo politico, di ingresso e soggiorno dei cittadini extracomunitari e di regolarizzazione dei cittadini extracomunitari ed. apolidi già presenti nel territorio dello Stato. Disposizioni in materia di asilo (Gu 28 February 2090).

1986, France introduced visa requirements to every national from a foreign country in the wake of a series of terrorist attacks.3 Although justifed on security grounds by French authorities, such a measure has been described by an insider to the process, former head of the cabinet of the Secretary of State for Immigration, has a classical use of a "policy window" (Kingdon & Thurber, 1984) – the terrorist attacks – to enact a decision that the French authorities have been advocating for more than 20 years (Weil, 1991). The plan of introducing visa restrictions, most notably to Maghreb countries (Morocco, Algeria, Tunisia), was dictated by the problem of irregular stays. The reluctance at enacting such a decision was due to its political costs, namely the negative impact of visa introduction to bilateral relations with former colonies. The policy window allowed for introducing visas to the entirety of nationals of foreign countries, except for nationals of the European Community and Swiss citizens, therefore making the decision acceptable, and then to progressively remove the restrictions for the nationals of OECD countries. Visas are maintained only for those countries that were the original target of visa policy, among which African countries, north and south of the Sahara.

From the 1980s, the logic of visa introduction relates to migratory concerns not just in the case of France but also in the case of the Benelux, another border-free region that established common visa requirements to lift the inter-state frontiers. In the 1960s, when the Benelux region was established, the common list of countries submitted to visa requirements resulted from the lifting of restrictions when differences existed among the members of the Benelux Agreement (Infantino, 2020). Only at the beginning of the 1980s, Benelux countries decided to terminate several of the agreements on visas removal as a strategy to control migration, following debates in both the political arena and the public opinion that were focusing on the 'problem' of immigration. The logic governing the emphasis on the more restrictive visa policy follows the lines of the post-1970s transformation towards the so-called new migration world, characterized by the objective of stemming rather than soliciting migration (Guiraudon & Joppke, 2001).

The logic of external control that characterizes the Schengen visa policy is novel for a country like Italy whereas it pertained to other countries that have shaped the making of the Schengen Area, as demonstrated by a series of provisions included in the Convention Implementing the Schengen Agreement. Signed in 1990, the Convention Implementing the Schengen Agreement is the outcome of 5 years of secret negotiations among the original Schengen countries. It signals the Schengen process' shift of focus towards the so-called compensatory measures to the achievement of the lifting of internal border checks, including the conditions that nationals of third countries have to fulfll to enter the Schengen territory, the harmonization of visa policies, the introduction of carrier sanctions, and the creation of the Schengen Information System (SIS), a joint database containing information on objects and persons used for the maintenance of public order and security. Those policy

<sup>3</sup>Terrorist attack on 23 February 1985 (on a Marks & Spencer shop), 9 March (the Rivoli Beaubourg cinema), and a double attack at the Galeries Lafayette and Printemps Haussmann department stores, 7 December (Bigo 1991).

instruments are the foundation of the external dimension of border and migration control in the European Union (Guiraudon, 2003).

To assess how the logic of external control connects to the practice, the next sessions address frst the EU texts and then, the ways in which they are put into action.

#### **3.2 The Practices of External Control: A View from EU Texts**

The European legal framework that provides for the criteria to introduce visa obligations as well as the conditions and requirements to cross Schengen external borders and issue Schengen visas are useful to analyze what kind of "frame", understood as the structures of belief, perception, and appreciation which underline policy positions (Schön & Rein, 1994), informs the issue of external control. The Acts under scrutiny are hard law, meaning legally binding and self-executing EU regulations, namely the Schengen Borders Code4 and the Visa Code5 as well as soft law, nonlegally binding guidelines like the Handbook for the processing of visa applications and the modifcation of issued visas.6 In these Acts, all the aspects related to migration control and the security of Schengen signatory states are not created from scratch. A series of provisions stem from the original Schengen process most notably the Convention Implementing the Schengen Agreement and the Common Consular Instructions.7

The Convention Implementing the Schengen Agreement provides for common instructions for the Contracting Parties' diplomatic and consular posts and consular cooperation at a local level formally known as the Common Consular Instructions. Common Consular Instructions are aimed at ensuring uniform implementation, given the condition of interdependency that characterizes the national issuance of a visa that authorizes entry to multiple states. A number of provisions included in the Schengen Borders Code and the Visa Code stem from the Common Consular Instructions. These include:

the verifcation of entry conditions at border crossing points;

the obligation for consular authorities to consult the Schengen Information System before a uniform visa can be issued;

specifc criteria in relation to the examination of visa applications;

<sup>4</sup>Regulation (EC) No. 562/2006 of the European Parliament and of the Council of 15 March 2006 establishing a Community Code on the rules governing the movement of persons across borders (Schengen Borders Code, OJ L 105 of 13 April 2006).

<sup>5</sup>Regulation (EC) No. 810/2009 of the European Parliament and of the Council of 13 July 2009 establishing a Community Code on Visas (Visa Code). OJ L 243, 15 September 2009.

<sup>6</sup>Handbook for the processing of visa applications and the modifcation of issued visas. Commission Decision, 28.01.2020, C(2020) 395 fnal.

<sup>7</sup>Common Consular Instructions on visas for the diplomatic missions and consular posts (2005/C 326/01) (OJ 2002 C 313/1).

the consular cooperation at local level that focuses mainly on the exchange of information regarding false documents, illegal immigration routes, *bona fde* applicants, and exchange of statistical information on visas issued and refused;

the obligation to stamp the passport of visa applicants to prevent and monitor contemporary visa applications to multiple consulates.

Article 21 of the Visa Code provides for the criteria to make decisions on a Schengen visa application, beyond supporting documents. It establishes that the examination of applications consists in:

the verifcation of the entry conditions provided for in Article 5(1)(a), (c), (d) and (e) of the Schengen Borders Code (verifcation that the travel document presented is not false, counterfeit or forged; verifcation of the applicant's justifcation for the purpose and conditions of the intended stay, and that s/he has suffcient means of subsistence; verifcation of the authenticity and reliability of the documents submitted and on the veracity and reliability of the statements made by the applicant);

the assessment of security risk, risk of illegal immigration, and the applicant's intention to leave the territory of the Member State before the expiry of the visa applied for.

A series of continuities and novelties can be observed. The correlation between the decisions on visa issuing and the assessment of the future probability of immigration is stated also in the Common Consular Instructions, as Box 3.1 shows:

#### **Box 3.1 Criteria for Examining Applications According to the Common Consular Instructions**

The diplomatic mission or consular post shall assume full responsibility in assessing whether there is an immigration risk. The purpose of examining applications is to detect those applicants who are seeking to immigrate to the member states and set themselves up there, using grounds such as tourism, business, study, work or family visits as a pretext. Therefore, it is necessary to be particularly vigilant when dealing with "risk categories", unemployed persons, those with no regular income, etc. to the same end, fundamental importance attaches to the interview held with the applicant to determine the purpose of the journey. Additional supporting documentation, agreed through local consular cooperation, if possible, may also be required. The diplomatic mission or consular post must also draw on local consular cooperation to enhance its capacity to detect false or falsifed documents submitted in support of some visa applications. If there is any doubt as to the authenticity of the papers and supporting documents submitted, including doubt as to the veracity of their contents, or over the reliability of statements collected during interview, the diplomatic mission or consular post shall refrain from issuing the visa.Source: Common Consular Instructions, part V, third paragraph

Pursuant to this paragraph of the Common Consular Instructions, "the purpose of examining applications is to detect those applicants who are seeking to immigrate to the Member States and set themselves up there, using grounds such as tourism, business, study, work or family visits as a pretext." Such a formulation comes very close to the one included in the Visa Code that focuses on the assessment of the applicant's intention to come back to his country of origin. In both cases, the "risk" is understood as immigration as such, not just undocumented immigration. However, an important difference exists between the Common Consular Instructions and the Visa Code namely the level of precision and details provided. The Common Consular Instructions exemplifes the "risk categories" (unemployed persons, those with no regular income) whereas the Visa Code remains vague. The Common Consular Instructions were a sort of guidelines with no legal value. The Visa Code is a legally binding and self-executing Council Regulation. It is adopted following Community decision-making rules that involve all EU institutions, starting with the Commission proposal, then the Parliament amendments and fnally, the adoption by the Council. The process of adoption of the Visa Code has shown that "No job, no visa" is not a sustainable position within the frame of the EU legislative policy making (Infantino, 2019).

Even in the European process of the crafting of legislations dealing with the visa policy, more precisions can be found in the Handbooks, most notably the Handbook for the processing of visa applications and the modifcation of issued visas, a piece of regulation that is not legally binding. The adoption of Handbooks does not follow the Community decision-making rules that involve the European Parliament. The Commission is assisted by the Visa Committee to draw up the operational instructions on the practical application of the Visa Code (Art. 51 and 52 of the Visa Code). Committees allow for the exercise of the implementing power of the Commission: They are composed of the representatives of the Member States and chaired by the representative of the Commission. The Handbook builds on the knowledge provided by Member States. The practical meaning of the slippery notion of migratory "risk" is specifed and detailed (point 6.13). It is split into two parts. The frst meaning clearly refers to undocumented migration and reproduces the phrasing of the Common Consular Instructions. It reads as follows: "The risk of illegal immigration by the applicant to the territory of Member States (i.e., the applicant using travel purposes such as tourism, business, study, or family visits as a pretext for permanent illegal settlements in the territory of the Member States)". The second part is about the future intention to leave the territory of the Member State ("Whether the applicant intends to leave the territory of the Member States before the expiry of the visa applied for"). The migratory risk is not only the risk of illegal settlements but also a more general risk of non-return. The Handbook also provides detailed criteria to establish "profles" of risk, a term that is completely absent in the Visa Code. Features that fall into the category of migratory risk are clearly stated. Features pertain mainly to the applicant's level of "stability" that can be ascertained by assessing her socioeconomic situation (family ties in both the countries of origin and destination, employment, marital status, level and regularity of income, social status). These operational guidelines convey the understanding that legitimate travelers must have

a series of socio-economic characteristics, a view that can be easily attacked on grounds of discrimination. These provisions also enter the realm of national sovereignty. The Handbook shows the specifc role of EU soft law, which is the making of regulations where no regulation would otherwise be possible (Cini, 2001).

Binding and not binding legal texts indicate the practical meanings of the migratory "risk", the features that fall into profles of "risk", the procedures and conditions to issue visas. Whether they provide more or less detail, these instructions fail to describe the ways in which they should be put into practice in the specifc cases that implementers face in the everyday. In the migration and border control policy domains, several studies have shown how implementation is about using discretion to adapt general rules to the specifc situations encountered on the job while solving dilemmas and responding to multiple yet competing objectives (Calavita, 1992; Eule et al., 2017; Tomkinson, 2018; Infantino, 2019). In the context of Europe, the vagueness of EU law is even more salient, as EU laws result from inter-state and inter-institutional compromise. The legislative framework regulating the Schengen visa policy is not exceptional in that regard, also due to visa policy's specifc characteristics, namely the fact that the same rules apply to a large variety of foreign countries and for it lies at the core of states sovereignty. The means of implementing control must be tailored to specifc contexts. Room for maneuvering ensure the exercise of the sovereign power to decide who gets in.

Inherited by the original Schengen process, the logic of external migration control and migratory concerns certainly permeates the texts regulating the implementation of EU visa policy. However, reading the instructions does not inform us on the actual practice. The next section draws analytical scrutiny on the extent to which these logics stay on paper and how they are entangled in the practices.

#### **3.3 The Blurring Boundaries of Control Practices**

The ways in which actors on the ground understand policy purposes matter since these understandings give sense to action. Putting policy into action is also about making sense of the objectives of policies and laws. A nexus exists between policy narratives, sense-making and organizational actions (Roe, 1994; Czarniawska, 1997; Banerjee, 1998). These kinds of narratives are essential knowledge to implement visa policy. To carry out their job, every newcomer to the organizations appointed to visa policy implementation wants to learn what are the issues at stake and what visa policy should do in the specifc local context, beyond regulations about procedures and conditions to issue visas. In the rare case that implementers read texts such as circulars or *vademecum*, they say texts are "theoretical" and that the theory is very different from the actual practice. Implementers also acknowledge that one should know the local context and that ways of doing things change according to the country in which visa policy is implemented. The same applies to trainings, deemed to be too broad. Sharing narratives with colleagues is a crucial means to make sense of what should be done.

In the case of "old" immigration countries also initiators of the Schengen process like Belgium and France, the narratives that associate visa policy to external migration control and the organizational understandings of the objective of visa policy as the pre-emption of immigration "risks" have historical depth. Senior and junior offcers, both at the consular frontline and in the offces of central ministries, develop narratives that convey specifc understandings of what are the issues at stake in issuing visas and the national priorities of visa policies. These narratives use a specifc terminology to explain the series of events that provoke organizational concerns. In the context of Belgium, to describe the issues at stake and priorities in assessing visa applications, offcers speak of "procedure circumventing" (*détournement de procédure*) namely the risk that somebody applies for a short-stay visa, such as a family visit or tourism, although the intention is settling in Belgium by applying for a residence permit (most notably in the case of relatives) or by getting married. Senior offcers know that the notion existed well before 2009, when the Visa Code entered into force. Adapted to current times, the concern of procedure circumventing involves the uses of the welfare state benefts like unemployment benefts. In the context of France, a very similar notion indicates the concern underlining visa policy implementation namely the risk of circumventing the purpose of the visa (*détournement de l'objet du visa*). The collection of pieces of information for the assessment of visa applications aims at establishing the probability of such a risk. Somebody applies for a short-stay visa for family, private visit or tourism while the intention is settling in France by getting married or applying for a residence permit. Decision making is also informed by the concern that some visa applications were aimed at receiving health treatments in France.

In the case of Italy, the actors that put visa policy into action, mainly foreign affairs offcials at different hierarchical levels, develop narratives that convey specifc understandings of visa policy. These understandings can be differentiated following a generational divide, also documented in other organizations of border control (Côté-Boucher, 2018), which sees a clear opposition between senior offcials and newcomers. Senior offcials nostalgically focus on the time when the core of consular matters were Italians abroad rather than migration control. The prestige of consular posts has always been measured by the presence of nationals in a foreign country. Issuing visas consisted in checking lists of banned foreigners mainly for public order and political reasons. Junior, newly appointed offcials tend to make a much clearer connection between visas and the "protection" of the nation state from undesirable migration, therefore joining the understandings of migration control as a mission that allows for appropriating repetitive administrative tasks, a way to make sense of one's job typically observed in some street-level bureaucracies, like prefectures in France (Spire, 2008) or Belgian and French consular sections (Infantino, 2019). As a result, the assessment of visa applications consists in preempting the risk of lawful settlement, embodied by parents who can apply for residence permits, young persons who can get married and protecting the welfare state from those who might use welfare benefts most notably public healthcare. In-depth ethnographic feldwork in Schengen consulates in a specifc local context like Morocco has been key in bringing insights on the ways in which day-to-day implementation practices become more similar although historically rooted crossnational differences. Participant observations and interviews with the offcers that make decisions on visa applications and the researcher's extensive presence in the policy and social worlds of consulates have substantiated the comparative analysis of how national boundaries are blurred on the ground. Rather than the reading of the common EU law or of the common guidelines, the informal exchange with peers from "old" immigration countries, triggered by the condition of interdependency that a common visa policy entails and the search for ways to solve the problem of putting visa policy into action, accounts for growing similarities in practices (Infantino, 2021). Consuls general meet informally and exchange views about the local contexts and issues at stake in implementing visa policy. Offcers make telephone calls to know the motivation for previous visa denial, therefore learning about other consulates' ways of making decisions. Offcers see each other at parties or even go out for dinner and talk about their job. Implementers share understandings of risk, issues at stake in implementing visa policy, objectives, and ways of doing things, as a result. Informal exchange among the local "community of practice" (Wenger, 1998) is the process whereby learning occurs and triggers policy change from below. In the case of Italy, that entails appropriating the logic of external control and translating the objective of visa policy into the assessment of the risk of non-return.

Two elements deserve mention. First, the objective of Schengen visa policy as suggested in legal texts – the assessment of the misuses of travel purposes to actually immigrate – builds on existing understanding and practice of some EU Member States like Belgium or France. The practical meaning of the migratory risk as the risk of non-return is not a novel EU notion. The logic of external control as the preemption of migration, most notably lawful migration, is included in the EU texts but builds on existing practices. Second, specifc logics are translated into practice even in the case of Italy, a country that shows an historically grounded understanding of visa policy that is very distant from the logics of external migration control.

Logics in EU texts and national practice are entangled in a twofold direction. Some (Northern) national practices have informed EU texts, and some logic in EU texts are translated into national (Southern) practice. However, that does not happen only by reading the texts but rather by sharing narratives, constructing perceptions of appropriateness, framing the problem of making decisions, developing practical knowledge in interaction with peers from other national contexts.

#### **3.4 A Model of "Europeanization" from Below?**

Policy convergence or policy divergence between (Northern and Southern) European Member States is a key issue for all those interested in the construction of the EU. There is a vast body of literature that goes under the label of Europeanization that focuses on the European integration. In that literature, the implementation tends to be understood as the transposition of EU Directives (Falkner et al., 2008; Sverdrup, 2007). However, cross-national comparisons about the practices that put EU laws and policies into action are an expanding feld of research that builds on a growing interest into the study of the implementation stage, also in the domains of migration and border control (Dörrenbächer & Mastenbroek, 2017; Eule et al., 2019; Jordan et al. 2003). In the visa policy area, some scholars have pointed to cross-national differences in implementation practices by analyzing rejection rates (Infantino, 2019), the regional variation in visa supply (Finotelli & Sciortino, 2013), and by taking the perspective of applicants' experiences in dealing with different consulates (Jileva, 2003).

This analysis of the practices of European external control that focuses on one Europeanized policy instrument of migration control – Schengen visa policy – and takes the case of Italy in comparative perspective allows for putting forward on the one hand, a series of factors that account for divergences and sustain the thesis of a North-South divide and, on the other hand, some processes and actors that account for convergences from below. Divergence cannot be explained using some hypotheses about the Europeanization most notably the "goodness of ft" (Börzel & Risse, 2000) or the "worlds of compliance" (Falkner & Treib, 2008). As Guiraudon (2007, p. 303) has noted in the context of anti-discrimination policy, the Europeanization literature that relies on the notion of "goodness of ft" overlooks cases where the ft is a priori "good". The original Schengen process has become the model of EU visa policy. Therefore, it is more familiar to some countries like Belgium and France. The logics on EU paper are based on existing practices of some EU Member States. It is expected to observe conformity in the modeling group of EU Member States whereas it is unexpected to observe the adoption of some logics and practices by a country that is historically far from the EU model. By taking the practice perspective, this analysis shows that Italy does not pertain to the "world of neglect" or "world of dead letters" – the typologies of compliance that Falkner and Treib (2008) defned. Falkner and Treib (2008) locate Italy in the world of dead letters because "what is written on the statute books simply does not become effective in practice". What is written on statute books cannot become effective in practice unless it is translated into practical meanings. That is valid for Italy (and the other countries of the world of dead letters) as for any other European Member State. As Italy is changing most, because it is adopting novel understandings, while revealing a concern about how to put EU visa policy into practice, which underline the gathering in communities of practice, one cannot argue that it is neglecting EU obligations.

The comparative research design in the analysis of the practices of external control via visa policy implementation shows processes of adjustment induced by the making of the EU that diminish cross-national differences between one Southern European country and some of the (Northern) European countries which have contributed to the designing of EU policy instruments of migration control. This analysis reveals the interactive nature of Europeanization understood as a process of institutional, strategic, normative (Palier & Surel, 2007) and cognitive adjustments (Hassenteufel, 2008). However, it takes the perspective of the worlds of practice, frontline organizations and policy implementation, therefore putting forward dynamics of change at a distinct level of the policy process. Dynamics of

change challenge the thesis of a North-South divide. The practice perspective to analyze external control in the European migration regime sheds light on the crafting of responses that might blur national boundaries. These dynamics consist in actors' interactions and their effects, which cannot be reduced only to convergence, but rather include translations (Hassenteufel, 2005), hybridization and synthesis (Rose, 1993). It is crucial to note that the actors under scrutiny are implementers which trigger processes of policy change from below. Ultimately, these actors translate logics that have been Europeanized (but stem from some EU Member States) into actual practice because of interactions on the ground rather than the reading of instructions. In sum, dynamics of policy change towards an EU model might be triggered from below.

#### **3.5 Conclusions**

This chapter has taken the practice and comparative perspective to contribute to the analysis of the European regime of external control. The thesis of a divide between Northern and Southern European Member States that hinder convergences has been put under scrutiny. By building on the case of visa policy, one Europeanized instrument of migration and border control, which lies at the very heart of the achievement of both free movement and the European border-free territory, we have seen that an EU model in this policy area exists and it derives from the original Schengen process. Such a process tracing shows the roots of contemporary understandings of the issues at stake in issuing Schengen visas and what are the logics that characterize the EU visa policy. These logics inform the crafting and drafting of the EU legal texts that day-to-day practice put into action. By focusing on a Southern European country and key player of Schengen visa policy, this chapter has argued that practices blur national boundaries and account for processes of change that hinder the policy legacies and other factors of divergence in implementation. A country like Italy, very far from the EU model when compared to original Schengen countries like Belgium and France, puts the logics of external migration control into action while translating the migratory risk into the assessment of the risk of lawful migration.

Such a fnding matters for both academic and society debates. First, it shows the importance of adopting the street-level perspective through the lenses of the migration regime concept. It encourages a reappraisal of certain hypothesis about the Europeanization by using the street-level implementation perspective in the analysis of the European integration. Implementers might be overlooked actors of "Europeanization" from below. This analysis also supports the perspectives on the making of the European Union that focus on processes of adjustment by including the implementation stage of the policy process. Second, at the level of political and media debates, the tendency at ranking, "naming and shaming" and classifying EU Member States according to best practices is widespread in several domains of governments' functions. In the context of migration and border control, the distinction

between "good" and "bad" pupils often overlaps with the North-South divide, putting Southern countries in the role of "weak" border controllers. It is safe to say that mistrust characterizes the lifting of interstate frontiers and the making of the Schengen Area just as the pooling of sovereignty. However, if one convergent tendency exists, it can be observed in the inclination towards restrictiveness and the adoption of understandings that see migration (whether undocumented or not) as a risk. That kind of tendency certainly represents an element of refection on the broader representations and attitudes about migration in both Northern and Southern EU Member States.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Part II The Externalisation of Control**

# **Chapter 4 A "European" Externalisation Strategy? A Transnational Perspective on Aid, Border Regimes, and the EU Trust Fund for Africa in Morocco**

**Lorena Gazzotti, Mercedes G. Jiménez Álvarez, and Keina Espiñeira**

#### **4.1 Introduction**

The summer of 2015 was baptised by some observers as the "Long Summer of Migration". The news of migrant arrivals in Greece and Southern Italy, together with the emergence of makeshift migrant camps along the Balkan route, were met by the unwillingness of EU Member States to work together on a strategy to receive and integrate the hundreds of thousands of people arriving on Southern European shores.

Unsurprisingly, the EU Commission met the "refugee crisis" with a renewed interest in the prevention of migration movements. In particular, the Commission emphasised the need to integrate migration in development cooperation policies. The 2015 Valletta Summit on Migration was an important step in this direction, as it resulted in the establishment of the EU "Emergency Trust Fund for stability and addressing root causes of irregular migration and displaced persons in Africa" (EUTF). With a budget of over 5 billion euro, the EUTF funds projects in 26 countries across three regions: the Sahel and Lake Chad, the Horn of Africa and North Africa (European Commission, 2020a). The Member States that are most engaged in funding the scheme are Germany (228.5 million euro), Italy (123 million euro) and Denmark (56 million euro) (European Commission, 2021).

The projects implemented as part of the EUTF tackle issues as different as the purchase of security material to improve border surveillance in Libya, to the funding of small business projects to reconvert the Nigerien smuggling economy into a

L. Gazzotti (\*)

University of Cambridge, Cambridge, UK

M. G. Jiménez Álvarez

Universidad Autónoma de Madrid, Madrid, Spain

K. Espiñeira Universidade da Coruña, Coruña, Spain licit market, to the improvement of the functioning of the justice system in Burkina Faso (European Commission, 2020b). With the EU taking charge of fundraising and disbursement, the EUTF thus created an overarching system of action in a feld – border externalisation – where Southern European countries had long been at the forefront of action, bearing contradictions with the protection of the international human rights regime. Since then, Northern European countries, which had a feebler presence in project delivery and execution, became more active in the funding of border control projects in the African context.

The EUTF tries to support the EU migration containment strategy by following three well established principles in border control cooperation: effective control over the arrival of irregular migrants can only be achieved by cooperating with "origin" and "transit" countries1 ; spurring development in contexts of "origin" and "transit" will reduce the incentive for irregular emigration; development cooperation, along with trade preferences, constitutes the main leverage to secure thirdcountry commitments to concrete outcomes in the feld of return and readmission (den Hertog, 2016). These three principles have guided EU relations with third countries in the feld of migration for the past 15 years: frst announced at the Tampere European Council in 1999, the EU formalised its external agenda in the Global Approach to Migration (GAM) 2005, renamed Global Approach to Migration and Mobility (GAMM) in 2011 – (Collyer et al., 2012; Hampshire, 2016). The GAMM set on paper the need to frame actions in the feld of the prevention of irregular migration within broader cooperation initiatives, tackling also the development of "sending" and "transit" countries and legal migration of the latter's nationals within broader migration debates (Collett, 2007). The GAMM gives a unique role to development fnancing as part of the strategy traced to establish "a balanced and comprehensive partnership with third countries" (EC, 2011), which in turn carries an inherent element of "conditionality" (Cortinovis & Conte, 2018).2

<sup>1</sup>Throughout the chapter, we use inverted commas to emphasise the political nature of the terms "country of origin" and "country of transit". These categories, in fact, have been vastly invoked by Global North policy-makers to justify their requests for shared responsibilities over border control to countries in the Global South. In particular, the term "country of origin" has been utilised to requests countries in the Middle East, Africa, Latin America and Asia to control the mobility of their own nationals. The term "country of transit", instead, has been utilised to request states to control the mobility of non-nationals suspected to "transit" towards destinations such as the European Union or the United States (Gazzotti et al., forthcoming).

<sup>2</sup>This manner of shared policing involves some forms of transactions or exchanges, often framed so that development aid is provided to third countries in return for their cooperation in matters related to migration and migration control. More precisely, one of the vital aspects of the migrationdevelopment nexus is that collaboration incentives are inherently embedded within "positive or negative conditionality" dynamics (Cortinovis & Conte, 2018, p. 4). This means that those countries that cooperate in managing migratory fows are, for instance, positively rewarded with "development aid, entrance to EU markets or diplomatic support" (Casas & Cobarrubias, 2018, p. 41). Alternatively, negative conditionality assumes "penalising", reduction or even termination of the support conditions if agreed expectations are not met (Cortinovis & Conte, 2018: 7). These different terms, however, are tightly interlinked and often utilised as a part of the same processes. Conditionality, therefore, forms a part of the externalisation process as the tool with which the EU can reward or coerce third countries into cooperating on the EU-driven interests (Gabrielli, 2016; Cortinovis & Conte, 2018).

Existing literature has tended to conceptualise the use of aid in border externalisation as part of a toolkit that countries in the Global North deploy to expand their reach beyond the geographical site of the border. Yet, thinking of a "European" border externalisation front risks to essentialise the reality of migration control in two ways. First, it fattens the engagement of European countries in border externalisation into a homogenous ensemble. However, Northern and Southern European countries have presented very different stakes in border externalisation in the past 30 years. Northern European countries, although always worried about the potential "transit" of unwanted migrants from the Southern European border to Northern Europe, have traditionally delegated externalisation strategies to Southern European countries. The signature and enforcement of the Schengen Agreement actually acted as a levelling measure for the immigration policies of countries such as Italy and Spain – considered during the 1980s and 1990s as the "soft underbelly" of the European Union due to their more open migration policies (Hollifeld, 1994). Consequently, countries as Spain have been amongst the frst to enact informal border cooperation strategies with third countries, such as Morocco (Gabrielli, this volume). Southern European countries have been at the forefront of border externalisation – to the point that their political leaders have time and again called for European leaders not "to leave them alone" (see il Fatto Quotidiano, 2017). Only recently, especially after the Arab Spring, Northern European countries like Germany have shown more interest to be directly involved in border control cooperation – especially in matters of return and prevention of irregular migration (den Hertog, 2017). Second, the idea of a "European" border externalisation front– like the idea of "migration governance" (Tazzioli, 2014) – somehow conveys the idea of "logical" process, taken by entities who collaborate – even though with ambiguities and setbacks – towards the achievement of a commonly acknowledged goal. Border practice literature, however, has shown that this imaginary could not be further away from the reality of development cooperation in the feld of border control – a feld where the security anxiety of the states often lead to totally contradictory outcomes, in situations where the very meaning of "controlling the border" is no longer readily identifable (Gazzotti, 2021).

Our paper advances a more nuanced and complex understanding of the political dynamics surrounding the externalisation of European borders. We ask: is externalisation a coherent and unifed process? Is it possible to identify a neat distinction between the migration control strategy pursued by countries in the North and in the South of Europe? How does aid affect border externalisation processes? Our fndings challenge the existence of a structured, coordinated European externalisation front. Rather, we show that the implementation of a specifc border externalisation programme is reactive and inconsistent in nature, driven by dynamics of temporary reaction to punctual crises and the NGO-isation of Member States. We show that differences between Northern and Southern European countries in this respect are less neat than one would expect, and that the use of aid in border control blurries the boundaries between the strategy of individual Member States. Furthermore, we demonstrate that Moroccan authorities – and in particular the Moroccan Ministry of Interior – substantially infuence the allocation of resources from the EUTF to Morocco as a country more broadly, and to hardcore security as a specifc sector of intervention.

The argument specifcally builds on the analysis of primary documents related to the implementation of EUTF projects in Morocco. The analysis is more broadly informed by research conducted discretely by the three authors in Spain and Morocco over the past two decades. The rest of the paper unfolds as follows. In the frst section, we discuss the theoretical framework underpinning our discussion. After that, we examine the use of development aid as an instrument of border control in the Western Mediterranean. We focus on how the EU has progressively emphasised the need to mainstream migration in development cooperation policies and how Morocco has been one of the leading destinations of that assistance. Following, the empirical section analyses the implementation of the EUTF projects in Morocco and uses fndings to challenge common understandings about border externalisation. In the concluding section, we challenge existing assumptions on border control, highlighting that EU Member States' strategy is not only affected by the EU's supranational policies, but also by the political manoeuvrings of the countries from the Global South.

#### **4.2 Theory: Transnational Governmentality, the Transnational Social Field, and Governing Through Aid**

Since the late 1990s, the study of migration has taken a transnational turn, insofar as scholars acknowledge that the experience of migration does not end with the arrival of the migrant into a new place. Migration creates durable connections between migrants' homelands and new places of residence, in the form of fnancial transfers, the emergence of circular mobility patterns, and the emergence of new religious or cultural centres in destination areas. The expansion of the migration feld occurs even when people do not actively live or circulate between two countries: in a seminal book, Abdelmalek Sayad (1999) highlighted that the migrant inhabits a double absence – simultaneously embodying both the country of origin and the country of destination, but never fully present in the frst, and often structurally marginalized in the latter.

In a context marked by increasing restrictions on cross-border movement, the transnational turn has also invested the study of migration governance. Globalisation has determined the emergence of actors who move, operate, and rule in ways that exceed the borders of the nation-state – such as transnational corporations, NGOs, and transnational criminal networks. The infuence that these actors exercise in the regulation of social, political and economic life blurs the boundaries between state and society, and challenges pre-established notions of local and global, state and non-state action (Ferguson & Gupta, 2002; see also Hansen & Stepputat, 2006, 2009). Eschewing the boundaries of the nation-state, these actors operate in a

"transnational social feld", which Liliana Suarez identifes as a "complex of dynamics that emanates from the impact of globalization of the labour market and in the governability of populations that are less and less linked to an only territory" (Suarez, 2007). In a seminal article, Ferguson and Gupta (2002) advanced the idea of "transnational governmentality" to conceptualise the indirect and composite patterns characterizing the establishment of global networks of power. Invoking the term "governmentality", in fact, alerts the reader to the slow, discreet workings of this kind of power – which does not necessarily manifest through straightforward foreign interference or overt coercion, but rather through more subtle techniques of indirect self-ruling (Tazzioli, 2014).

Aid agencies rank highly in the business of governing beyond the nation-state. The action of donors, NGOs and IOs transcends scales of action: aid agencies occupy an in-between position in the international arena, stretching their action between different sites in the globe and exercising a capacity to act locally according to transnational logics. Migration governance is an obvious arena to observe this phenomenon. Although states have often delegated private actors, in particular carriers and transport companies, to verify identities and surveil mobility (Torpey, 2000), the anxiety of the Western nation-state around the presence of foreigners on its territory has determined the proliferation of modes, sites and actors surveilling the mobility of those singled out as "undesirable". The frontier has "stretched" (Espiñeira, 2016) before and after its geographically fxed location, as controls on migrants' identity and administrative status take place "away from the border" (Lahav & Guiraudon, 2000, p. 55; Nieuwenhuys & Pécoud, 2007, p. 1676), both in "sending" and "transit" countries as well as inside destination countries (Ford & Lyons, 2013; Casella Colombeau, 2015; Infantino, 2016). Countries like Morocco have experienced a process of border externalization, which Cecilia Menjivar defnes as "a series of extraterritorial activities in sending and in transit countries at the request of the (more powerful) receiving states (e.g. the United States or the European Union) for the purpose of controlling the movement of potential migrants" (Menjívar, 2014, p. 357). Scholars are increasingly resorting to the term "migration regime" to encapsulate "the complex and multilayered political regulations of migration that escape realist defnitions of the state as an acting entity" (Cvajner et al., 2018, p. 7). Border control and its externalisation take place both inside and outside the nation-state, within and across the North and the South. Its enforcement is assured and contested by state and non-state actors alike, including the EU, its Member States, International Organisations such as the IOM and the UNHCR, grassroot organisations, and migrant people as well (Stock et al., 2019; Gross-Wyrtzen & Gazzotti, 2021). Such actors are in charge of different dimensions of migration control, as they operate at different scales – ranging from the national level of policy design to the local level of street-level implementation. Accounting for the different scales of migration control enforcement and contestation allows to include the informal sites of governance into the formal discourse – which, in turn, conveys a more truthful account of policy successes and failures (Cvajner et al., 2018).

Chief amongst the instruments adopted by Global Northern states to further their border externalization strategy is development aid. Since the late 1990s, in fact, the EU and its Member States have resorted to their aid budgets to fund the most disparate border-related projects, from the training of border guards to Voluntary Return, from advising on migration policy reforms to the delivery of equipment for security forces. This refects a well-established process of securitization of development aid, which emerged after the end of the Cold War and escalated after the outbreak of the War on Terror. Deemed to be the "root causes" of conficts, poverty and marginalisation in aid-recipient countries have become objects of development and humanitarian governance in virtue of their presumed dangerousness for donors' constituencies (Duffeld, 2001, 2013; see Gupta, 2015). In a world of potential – but not fully realised – threats, aid comes to target what Duffeld terms "surplus population", or "a population whose skills, status or even existence are in excess of prevailing conditions and requirements", a share of humanity that is made redundant and portrayed as "dangerous" by processes of local and international capitalist accumulation (Duffeld, 2013, p. 9).

Saying that development aid has been transformed into an instrument of border externalisation, however, does not mean that countries of migrant "origin" and "destination" are powerless spectators and subjects in that process. Much to the contrary, countries such as Morocco, Libya or Turkey have been historically able to use migration as a "geographical rent" to forge alliances or gain leverage power both in relation to their Northern and Southern neighbours. Tsourapas has coined the term "refugee rentier states" to conceptualise the ability of countries hosting large refugee populations to obtain economic advantages (in the form, for example, of preferential trade agreements or additional development budgets) from Northern donors in exchange for their engagement in border control (Tsourapas, 2019). As Nora El Qadim argues in her work on EU-Morocco border control cooperation, the capacity of resistance of actors in the South is often accompanied by the incapacity of actors from the North to operate in a unifed and coordinated way: the EU and its Member States hardly act as unitary, almighty, and omniscient entities deploying perfectly coherent strategies to secure their borders at home and abroad. In her book, for example, El Qadim shows how the different directorates of the EU were in competition with each other to access funding and negotiation spaces in areas related to forced return of Moroccan and third-country nationals to Morocco (El Qadim, 2015). As mentioned in the introduction, the development cooperation sector in Morocco is a set that challenges the presumed homogeneity of border externalisation: Northern European donors, in fact, have recently switched from a position of absence to one of scattered activity, where inter-state coordination and commonality of objectives are clearly lacking. The result is a border landscape that can sometimes be patchy and contradictory, and characterised by unexpected alliances and unexpected rivalries.

#### **4.3 Governing Migration Through Aid in the Western Mediterranean**

After entering the European Economic Community (EEC) in 1986, Spain became part of the Schengen Agreement in 1991. At the Spanish-Moroccan border, the range of legal modifcations associated with the "Schengenization" came together with the implementation of new securitization techniques, such as the establishment of visas and the reinforcements of the fences surrounding the Spanish enclaves of Ceuta and Melilla (Ferrer-Gallardo & Espiñeira, 2016).

The securitisation of the border reshaped mobility dynamics in the region, as it obliged migrants to take more dangerous routes to reach Europe. In the mid-1990s, North African countries like Libya, Algeria, Tunisia and Morocco started consolidating as key "transit" countries (Collyer, 2007; Bredeloup, 2012). The EU and its Member States started seeking the collaboration of non-EU countries to secure their external borders. Morocco provides a case in point: since the 1990s, security controls were reinforced all along the maritime and land border between Spain and Morocco with the fnancial assistance of EU institutions. Cooperation with third countries was also sought through external migration dialogues, conducted by the EU Commission and the European External Action Service (EEAS) on three levels: continental, in the EU-Africa Partnership on Migration, Mobility and Employment; regional, for example in the Rabat Process, a multilateral dialogue grouping European, North African, Central African, and Western African countries as well as IOs; and bilateral, through Mobility Partnerships (MP) and Common Agendas on Migration and Mobility. The MPs are conceived as the main long-term bilateral framework for facilitating policy dialogue and operational cooperation on migration management –based on the expectation that, in exchange for greater commitment by the signatory countries, the EU will offer access to visa facilitation regimes, support for student exchanges and migration and development initiatives (Reslow & Vink, 2015).

Initially, the emphasis of several EU-Africa declarations and partnerships were strongly marked by the fght against irregular migration. However, since 2005 EU institutions have said that securitisation alone cannot achieve this, hence the dimensions of cooperation, partnership and development have been promoted (Hansen & Jonsson, 2011). An example of this is the Rabat Declaration, which was created during the frst Euro-African Ministerial Conference on Migration and Development held in 2006 by the initiation of Spanish political action together with France and Morocco. Also called the Rabat Process, it promoted regional cooperation to prevent irregular migration (Frankenhaeuser et al., 2013), stating that: "[…]the management of migratory fows cannot be achieved through control measures only, but also require a concerted action on the root causes of migration, in particular through the implementation of development projects in Africa" (Rabat Declaration, 2006, p. 2). The inclusion of the "developmental" aspect on the premises of the EU migration agenda was strangely refected as the frst concrete action after the Rabat Process and resulted in the creation of joint border patrols between Africa and

Europe, such as the Senegal-Spain Joint Operations (Gabrielli, 2016, this volume). So essentially, migration management, although having adopted a developmental dimension, was still rooted within security and border control, but now through cooperation with the "transit" and "origin" countries.

Since then, there have been lasting changes to the relationship between migration and development in EU external funding. This relationship is increasingly shaped by the more-for-more principle, under which third-country cooperation with the EU's external agenda on migration, borders and asylum is becoming dominant (den Hertog, 2016, p. 14; Carrera et al., 2016, p. 12). Through the European border externalisation strategies, the security frame is also beginning to broaden geographically as the "European borders" towards Africa. Moreover, the outsourcing of EU "border-works" is not simply about the policing of migration, but is also "part of a broader attempt to 'secure the external'" (Bialasiewicz, 2012, p. 845).

The European Agenda on Migration (EC, 2015) emphasised the need to mainstream migration in development cooperation policies. Together with trade preferences, development cooperation is the main leverage to get commitments and outcomes in readmission and return, being these the top priorities on the Agenda. Under this framework, when the Valletta Summit on Migration took place envisioning the EUTF, the Commission then presented the New Partnership Framework (EC, 2016) to reinforce cooperation. It includes three common elements for a partnership to become effective: (1) conditionality based on the cooperation of the partner country on readmission and return; (2) effective incentives, in particular through EU trade and development policies; and (3) a tailored country package. The European Council aimed the conclusions of these migration compacts at a number of priority countries mainly from the Horn of Africa and the Sahel and, in North Africa, it focused on countries with a low return ratio and where negotiations were not advancing, such as Algeria, Tunisia and Morocco.

The EUTF is currently one of the leading fnancial instruments between the EU-Africa partnership in the feld of development linked to migration (den Hertog, 2016; Kervyn & Shilhav, 2017). The region of Sahel and Lake Chad receives the most signifcant amount of money and support compared to other areas, which are Northern Africa and the Horn of Africa (Kervyn & Shilhav, 2017). For most of Northern African countries, the majority of the budget is invested in migration management, as the countries in this region are seen as 'transit' countries. Whereas in the Sahel region and Lake Chad the adopted actions address multiple objectives. These include development fnancing for a root-causes approach based on the questionable assumption that this will limit the "push factors" for migration, and linked to this are actions around "migration and development", e.g. mobilising the diaspora of third countries for economic development, as well as migration management actions fnancing national migration policies, police capacities and data exchange (den Hertog, 2016, p. 13).

Morocco has always been a key partner for the EU and its Member States in border control cooperation, including in the area of development cooperation. Since the EU external cooperation in the feld of migration took shape in 2004, the country has been one of the main destinations of that assistance. Between 2004 and 2006,

Morocco was involved in 22 projects funded through the AENEAS programme, for a total amount of 18 million euro. This is a substantial fgure, especially considering that the programme funded 105 projects in total (EP, 2015: 127). In 2007, Morocco received a further 67.6 million euro to implement an integrated border management programme directly managed by the Ministry of Interior. The EU Commission never managed to audit the programme, and no offcial information is available regarding its implementation (EP, 2015: 130).

Morocco was the frst Arab Mediterranean partner country with which the EU signed a Mobility Partnership, in June 2013. The Joint Declaration between the EU, nine of its Member States (Belgium, France, Germany, Italy, the Netherlands, Portugal, Spain, Sweden and the United Kingdom) and Morocco was complemented by an Annex of cooperation projects by EU agencies and Member States supposed to contribute to the achievement of each of the objectives stated. These objectives are (i) a better management of the movement of persons for short stays, legal migration and labour migration, (ii) strengthening cooperation in the feld of migration and development, (iii) fghting irregular migration, networks of traffcking of human beings and the promotion of an effective return and readmission policy, and (iv) respecting international instruments on the protection of refugees (EP, 2015, p. 127). The fagship initiative funded by the EU in relation to the Mobility Partnership is the Sharaka Project, a fve million euro programme launched in 2014 to support the MP implementation through a series of capacity-building initiatives. Its main objective is to support national migration and development and mobility policies in a framework of reinforced cooperation between Moroccan and European administrations. The specifc objectives are to optimise the positive effects of migration, mobilising the expertise of the EU Member States to respond to the needs of Moroccan institutions –short and long term expertise, peer-to-peer exchanges, studies and benchmarking, pilot actions, networking and targeted communication (EP, 2015, p. 127).

The MP was followed, 3 months after its signature, by a change in the immigration policy in the Kingdom. The new policy prompted a quick reaction from the EU and its cooperation in this feld. In the framework of the Sharaka Project, a new ten million euro budget support programme to promote the integration of immigrants has been deployed, including two million euro each for the Ministries of Public Health and Education to support the extension of health care and school enrolment to all immigrants (EP, 2015, p. 129). In August 2015, there were in Morocco 25 different ongoing projects in the feld of migration funded by the EU, for a total of more than 20 million euro over their implementation period.3

<sup>3</sup>Of this amount, ten million euro corresponds to the new budget support programme to promote the integration of immigrants in Morocco, launched in 2015 after the adoption of the new national immigration policy. four million euro of it was added to existing budget support programmes in the health and education sectors, 1.6 million euro for technical assistance to the competent Ministry and 4.4 million euro for assistance to migrants, of which 1.4 million euro for integration assistance to immigrant women. Another pre-existing project (1.6 million euro) aims to improve the protection of human rights of Sub-Saharan immigrants in Morocco. Another fve million euro corre-

#### **4.4 What the Implementation of the EUTF Tells About "Externalisation"**

#### *4.4.1 De-centring and Re-centring Morocco in the European Border Control Strategy*

As mentioned earlier, the EUTF was established as part of the EU response plan to the "migrant crisis" in the Central and Eastern Mediterranean. Through the EUTF, the EU and its Member States aimed to address the increase in migrant arrivals by spurring development in countries of "origin" and "transit", so as to reduce the incentives for migrants to continue their journey onwards. When the EUTF was established, the majority of African migrants arriving in Europe came from Central and Eastern Africa and had transited through Libya. The funding allocation refected the geopolitical attention of the EU on the Central Mediterranean route: with the exception of Libya, the Maghreb was not central to the functioning of the EUTF, and Morocco in particular was quite marginal in the funding allocation. Such marginality was also evident in the discourse of policy-makers. During an interview conducted in October 2016, an offcer of the EU Delegation in Rabat dismissed a question on the EUTF as irrelevant to the Moroccan context. "The Trust Fund addresses the root causes of migration" the interviewee said. "Morocco is therefore not central, because it [the Trust Fund] really targets West African countries" (Gazzotti, 2019).

Morocco's geopolitical relevance within the EUTF changed in 2018, when a (modest) increase in migrant arrivals recorded at the Southern Spanish coast pushed Spanish authorities to sound the alarm about the emergence of a new migration "crisis" at Europe's borders. In 2018, the European Border and Coast Guard Agency (FRONTEX) recorded 57,034 irregular border crossings attempts on the Western Mediterranean route –compared to the 23,063 recorded in 2017. This increase occurred in a context where irregular arrivals at the EU external borders were "the lowest level in 5 years", as declared in a note circulated by the Romanian Presidency of the Council of the EU in February 2019 (Statewatch, 2019). However, it was enough to push Spain and the EU to escalate the attention on the Western Mediterranean border.

Since the EUTF was established in 2016, Morocco received a total amount of 17,894 million euro for national projects. Of these, 160,25 million euro were granted between 2018 and 2020 – and 145,75 million euro relate to projects focusing on hard-core border security. Between July and August 2018, the EU granted Spain 30 million euro to reinforce border control measures in the South of the country, and

sponds to the Sharaka Project to support the implementation of the MP, which also has a strong technical assistance and institutional capacity building component. one million euro funds a return programme to countries of origin from Morocco, plus 1.6 million euro in the framework of a multicountry project to improve cooperation between countries of origin, of transit and of destination on voluntary return to meet the needs of vulnerable migrants. The rest goes to small migration and development technical assistance programmes.

Frontex increased the human and infrastructural resources deployed at the Spanish-Moroccan border. The renewed impetus to securitise the Western Mediterranean border were accompanied by calls for Morocco to ramp up its migration control measures (GADEM, 2018). In late July 2018, Morocco started conducting internal dispersal campaigns to distance "potential irregular migrants" from border zones. According to data from the NGO GADEM, around 6500 people were arbitrarily arrested in the North of Morocco and bussed to areas in the centre and interior of the country, hundreds of kilometres away from their houses. The raids in the North of Morocco were conducted according to a logic of racial profling, as they exclusively targeted black people, regardless of whether they had regular residency papers or the right to international protection (Gazzotti & Hagan, 2021).

The funding allocation strategy pursued by the EU through the EUTF refects the renewed geopolitical salience of the Western Mediterranean border, and of Morocco as a partner of the EU. The factsheet of the project "Support to the actions of Moroccan authorities against the networks facilitating irregular migratory fows" (European Commission, 2020c) makes clear that the need to invest greater resources from the EUTF on Morocco is not only due to the renewed relevance of the Western Mediterranean route, but also to make sure that the decrease in the number of irregular border crossings is maintained with time.

The geographical situation of Morocco, in the immediate proximity of Spain (included on the African continent with the enclaves of Ceuta and Melilla) makes this country particularly sensitive to the irregular migratory fows directed to Europe. The so-called Western Mediterranean route via Morocco represented 12% of the total irregular arrivals in Europe in 2017 compared to 0,7% in 2015. The number of migrants that take this route has recorded a strong increase since May 2018 and the total number of arrivals increased by 131% between 2017 and 2018, to achieve a total of 64.298 people in 2018. Anyway, starting from February 2019, notably thanks to the intensifcation of the action of Moroccan authorities, the trend recorded a stark U-turn. In the period between January and October 2019, 26.082 irregular migrants arrived to Spain from the Western Mediterranean route, therefore 51% less than those arrived in the same period of the last year (53.268). (p. 4, translation by frst author)

The de-centring and re-centring of Morocco within the funding allocation strategy of the EUTF reveals the reactive nature of this particular border control instrument. Rather than responding to a long-term, structured strategy to address irregular migration, the EUTF is used by the EU to respond to short-term objectives, especially in circumstances where a sudden, albeit moderate, increase in border crossings is labelled as an emergency by those Member States tasked with its management.

#### *4.4.2 Heterogeneous Strategies and the NGO-Isation of EU Member States*

As mentioned earlier, the EUTF is administered by the EU Commission through its local delegations in aid-recipient countries. The implementation of the projects is operationalised through a "delegated cooperation", whereby public development

cooperation agencies belonging to individual Member States bid for and can be selected as contractors for large-scale aid-funded projects. This process produces a process of NGO-isation, or the tendency of Member States to act like NGOs insofar as they become the implementing actors of EU development funding. This marks a stark difference in the use of aid for border externalisation. When Southern European countries started introducing "soft" methods in border control, this process relied on bilateral cooperation – with Spain, Italy or France directly interacting with Morocco in the defnition and implementation of development projects. Fifteen years later, the process of delegated cooperation resemanticises this process in two ways. First, it involves Northern European countries in the funding and disbursement of aid. The European Commission grants technical-fnancial operations with development-aid in those geographical areas where experience and international recognition are concentrated. The three major countries that implement delegated cooperation funds with Morocco are Germany, France and Spain. Second, by moving cooperation from the bilateral to the multilateral level, the EUTF puts EU Member States in a position of subordination to the fnancial power of the EU. The EUTF increased the budgets of development actors, including not only NGOs and IOs but also development agencies handling the development budget of EU Member States (such as Enabel, Expertise France, GIZ or the AECID). In Spain, for example, delegated cooperation complemented the depleted funds of Offcial Development Assistance (ODA), which had seen a dramatic reduction in 2008 and had left many actors involved in the implementation of development cooperation projects in a precarious situation (Olivié & Pérez, 2019).

The process of delegation, however, makes EU Member States less autonomous, insofar as their agencies have to write projects matching the guidelines and complying with the procedural mechanisms of the EU. Reliant on funding that is allocated based on a competitive process within pre-established frameworks, EU Member States are less able to implement their own border externalisation agenda, and have to subordinate it to that of the EU. Delegated cooperation has blurred the state nature in the solutions provided by development cooperation by putting the EU regional vision in the foreground. Although all the projects analysed broadly aim to contribute to the overarching objective of migration control, it is extremely diffcult to identify a difference in the agenda of Southern European or Northern European countries or a distinctive migration regime for each individual Member State – partly because the projects they implement have very short timelines and scattered objectives, and partly because any difference has been diluted through the alignment with the EU global approach guaranteed through its delegations.

Belgium, through its development agency Enabel, for example, implements a 4.6 million euro project on the legal empowerment of migrant people (European Commission, 2017a) and a eight million euro project on the deployment of migration policies at the regional level (European Commission, 2018) –essentially targeting foreigners that have beneftted from the regularisation process in Morocco. Germany, through the GIZ, works on a 8.6 million euro project (European Commission, 2017b) on South-South migration cooperation. Spain implements one 5.5 million euro project through its development agency, AECID, and a 44 million euro project through the *Fundación Internacional y para Iberoamérica de Administración y Políticas Públicas* (FIIAPP). The two projects, however, have starkly contrasting objectives: AECID project focuses on the fght against xenophobia and discrimination; the FIIAPP project, instead, focuses on strengthening the capacity of the Moroccan security apparatus in border control, including the purchasing of equipment for the control of maritime and borders (European Commission, 2020d). The fact that the same EU Member State implements two actions (funded by the same funding body) which simultaneously *fuel* and *fght* racism highlights the neurotic politics characterising border externalisation. Morocco is also beneftting from a number of regional projects funded by the EUTF. The impact of regional projects on individual target countries is often diffcult to single out, which makes the coherence of the European migration regime even more challenging to decipher.

The only common thread linking a number of different projects is the material predominance of security concerns in the allocation of aid budgets. Currently, Morocco is the sole recipient of 7 projects implemented through the EUTF, for a total budget of 178.9 million euro. 80% of this budget refers to two security-related projects, one implemented by FIIAPP, and a second one, "Support to the actions of Moroccan authorities against the networks facilitating irregular migration fows", directly managed by the Moroccan Ministry of Interior as a form of budget support (European Commission, 2020c). This very substantial security budget is complemented by a further 30 million euro envelope allocated through the Border Management Programme for the Maghreb region (BMP-Maghreb), a regional programme handled by ICMPD. The intentionality of the EU in migration control is reduced to the availability of funding to implement security-related projects, a rationale that is then imposed on the implementing actors through the process of delegated cooperation.

#### *4.4.3 The EUTF and Morocco's Migration Diplomacy*

European pressures to secure their borders abroad have been met and resisted by Morocco's own migration control strategy. The approval of a border surveillance strategy by the Moroccan Ministry of Interior in 2002 was followed the next year by the approval of Law 02–03, a repressive migration act criminalising irregular immigration and emigration from the country. Branded as the clear evidence of Morocco's transformation into "Europe's gendarme", Law 02–03 actually reconciled a more diversifed set of interests: on the one hand, the appetite of the EU and its Member States to externalise their borders through the cooperation of North African countries; on the other hand, the willingness of Morocco to gain geopolitical leverage after a long time of political isolation in the continent and marginality in EU relations. The explicit security vocation of the law refected in the harsh policing methods that characterised the following decade of migration governance in Morocco: until 2013, the presence of black migrant people from West and Central Africa in the country was controlled through the deployment of vast and frequent arbitrary arrest campaigns, often complemented by deportation to the no man's land at the border with Algeria. The areas surrounding the Spanish enclaves of Ceuta and Melilla were particularly subjected to these violent policing methods.

In September 2013, Morocco reformed its approach to migration governance by announcing the adoption of a new migration policy, more respectful of migrants' rights and of Morocco's international human rights engagements. Applauded by the EU and its Member States (Gazzotti, 2021), the new migration policy included the announcement of a regularisation campaign in 2014, followed by a new one in 2017; the adoption of a National Strategy for Immigration and Asylum (NSIA) in December 2014; and the launch of three law projects on migration, asylum and human traffcking. The new migration policy earned Morocco further support from the EU and its Member States, as by centering the integration of migrants locally into the country's strategy, it clearly aligned with border externalisation imperatives. But it also allowed Moroccan authorities to ease relations with their African neighbours by offering a more dignifed treatment to their citizens living in the country after a decade of systematic abuses. The implementation of the policy, however, did not match the aspiration of the NSIA: although violence against migrants signifcantly decreased and over 20,000 people were issued residency permits during the regularisation campaign in 2014 (and a similar number was regularised in 2017), the two laws on migration and asylum have not yet been approved, the inclusion of migrants into state services still languish, and arbitrary arrest campaigns remained commonplace in border areas. Since the summer of 2018, migrants living in other areas of the country have also become the targets of arrest campaigns and violence by Moroccan police forces. Migrants' detention and dispersal to the South of Morocco were also recorded after the beginning of the Covid19 pandemic, when strict quarantine policies and the shut-down of the economy made migrant people particularly vulnerable not only to the virus, but to violent border control practices enacted by an anxious security state.

Reading the implementation of the EUTF in Morocco only through the action of Global North actors offers a very partial picture of the hardening of the Western Mediterranean border. It is true that the EUTF has partially homogenised the action of EU Member States – with Northern European countries taking a more active role, and projects responding to the same guidelines and operations. Morocco, however, directly shapes the functioning of the EUTF, as it steers the priorities of the EU and its Member States according to its own geopolitical agenda. The allocation of over 175.7 million euro for border security to Moroccan authorities between 2018 and 2019 cannot be explained solely as the result of the EU willingness to externalise its borders. Representatives of the Moroccan Ministry of Interior have long been outspoken about the substantial costs that Morocco was shouldering to watch out its borders and prevent irregular migration from its territory.4 In August 2018, Mustapha

<sup>4</sup>The expansion of the IOM Voluntary Return Programme in the mid 2000s, for example, was also partially the product of the complaints made by the Moroccan government about the disproportionate fnancial responsibility that Morocco was shouldering regarding the deportation of irregular migrants back to their countries of origin. Although the IOM had executed some ad-hoc voluntary

Khalf, the spokesperson for the Moroccan government, rebuked accusations made by Spain and the EU about Morocco's supposed laxism in border management by arguing that Morocco was doing huge efforts on the control of irregular migration, shouldering responsibilities beyond its capacities. Pointing to the lack of support from the EU, he specifed that Morocco "makes huge sacrifces alone and with its own means and that the fgures concerning [fnancial] support remain below the efforts and the sacrifces allowed by the Kingdom [of Morocco]" (GADEM, 2018, p. 48). That same year, Khalid Zerouali, director of Migration and Border Surveillance in Morocco's Interior Ministry, declared to the Spanish press agency EFE that Morocco was spending over 200 million euro per year in border patrol, including the deployment of 13,000 members of different security forces.

The concerns raised by Moroccan authorities about the disproportionate division of economic responsibilities in border control seem to have clearly bore a weight in the rationale underpinning the allocation of the EUTF budget to Morocco. The factsheet (European Commission, 2020c) of the project "Support to the actions of Moroccan authorities against the networks facilitating irregular migration fows", for example, specifes that facing the challenges related to border control required Morocco to "deploy major efforts at all levels and to continue allocating important human and fnancial resources [to border management]". The EUTF-funded project thus aimed at "supporting these essential investments for the fght against crossborder criminality under all its forms, on the whole Moroccan territory". The factsheet then clarifed the rationale for the budget and for the activities linked to the project, directly invoking the costs that Moroccan authorities were shouldering in relation to border control operations:

In terms of costs, in its fght against illicit smuggling, Morocco deploys major efforts at all levels through a strategy of border control that needs an important investment in terms of equipment and human resources (more than 30.000 elements of different security forces). The implementation of this action envisages that Moroccan authorities reinforce the resilience of control devices through important investment in terms of additional human resources as well as aerial, maritime and land instruments of detection, interception, communication, precocious alert and mobility. *The budget estimated by Moroccan authorities for this investment is 3.5 billion euro in the 2020–2027 period. Taking into account the 8-year period, this gives a yearly amount of euro 435 000 000* [translation ours, emphasis added]*.*

The factsheet then concluded that "the engagement of the Moroccan government in the framework of this action is suffciently relevant and credible to be supported with an important programme of budget support, which allows to contribute to a better management of migration and the fght against migrant smuggling and human traffcking, including a reinforcement of integrated border management".

return interventions, Moroccan authorities considered IOM's spot assistance to merely be "symbolic", to use the words of Moroccan offcers. In a conversation with American diplomats, Khalid Zerouali, director of Migration and Border Surveillance in Morocco's Interior Ministry, argued that IOM support was insuffcient in complementing the substantial economic effort that Morocco was making to repatriate irregular migrants (American Embassy of Rabat, 2006).

Acknowledging the capacity of Moroccan authorities to infuence the allocation of the EUTF allows us to nuance claims on the existence of a compact "European" migration regime. First, Moroccan authorities seem to consider aid more as a partial reimbursement for the gigantic costs they have to incur to fulfl migration control duties rather than an incentive or a conditionality to cooperate with European countries. This stands in stark contrast with scholarship contending that North African and Middle Eastern countries use migration as a bargaining chip to accumulate economic resources. In other words, Morocco seems to be partaking in border control cooperation more to gain and maintain political relevance in the Western Mediterranean rather than to accrue fnancial incentives, given that the amount of money that the country invests in migration management seems to be much higher than the contribution of the EU. Second, the infuence that Morocco is able to play on the implementation of the EUTF speaks to the agency that countries labelled "origin" and "transit" manage to exercise in border control cooperation. Despite being on the receiving end of border externalisation policies, countries such as Morocco manage to display signifcant agency in shaping the implementation of migration containment projects funding by the EU or its Member States. Morocco manages to do that either by actively shaping the project outline and budget, or by selectively engaging with the EU and its Member States, or by posing caveats to European action, as in the case under analysis. The shape that the EU externalisation regime takes is therefore also signifcantly infuenced by Morocco's reaction to border pressures.

#### **4.5 Conclusion**

In this article, we understood border externalisation as a transnational social feld where different pressures – established migratory patterns, the appetite of the security state to control the border, the diplomatic interests of countries in the Global North and in the Global South, and the fragmented strategies implemented by Northern and Southern European countries – play out in the power relations that solidify at the frontier. Taking the case of the EUTF specifcally, we refected about how the EU externalisation agenda comes to life in aid-recipient countries. Development projects can thus be explored as a transnational social feld where border politics, migratory regimes and aid policy overlap and clash. The EUTF thus becomes a vantage point to explore the contradictory logics of EU border politics, and to deconstruct the concept of externalisation. The EUTF constitutes the most interesting operational response of the EU to the migration crisis, as it signifes how European actors conceptualise the transformation of geopolitical spaces (Zardo, 2020). It also reshapes what externalisation means on the ground, and it shows how the European migration regimes can advance and backtrack in the Mediterranean scenario.

Our fndings challenge the existence of a structured European externalisation front. We show that externalisation has always been a battlefeld where the strategies of Northern and Southern European countries have been anything but coordinated. Prior to the establishment of the EUTF, this was particularly visible insofar as Northern European countries were basically absent from the debate, with Spain being the most active actor in the feld. After the establishment of the EUTF, the boundaries between the strategies of Northern and Southern European states blur into the technicalities of delegated cooperation, whereby the agency of individual Member States seems to submit to the functioning of the EU, driven by a clear, yet contradictory, objective to advance securitisation policies. The NGO-isation of Member States, in particular, highlights how lack of programmatic vision in development cooperation policy, which can be easily hijacked by the European border externalisation aspirations. The implementation of a specifc border externalisation programme is reactive and inconsistent in nature, driven by dynamics of temporary reaction to punctual crises and the NGO-isation of Member States.

The introductory section of this article referred to the summer of 2015, when almost one million people crossed the Aegean Sea to reach Europe. We conclude this article by referring to the 'border crisis' which occurred between 17 and 19 May 2021, when over eight thousand people, including many unaccompanied minors, crossed the border from Morocco to Ceuta (Eldiario.es, 2021). This latest episode highlighted the fragility of the European migration system, based on bordering, containment and control as the pillar of externalisation politics. This clashed with the migration diplomacy displayed by Morocco, based on the conversion of migration fows into a political weapon in the Mediterranean borderscape. This contradiction is particularly evident in transborder regions and its surrounding territory, where mobility has always been part of the economic and social dynamism. Migration control suffocates the plurality of exchanges that make cross-border life (and the life of cross-border communities) sustainable. Geopolitics takes embodied shape at the border, and transforms migration control into an essential instrument in diplomatic confrontation. Morocco is the target while an ally of these policies. When accepting them or not, the decisions the Kingdom makes can affect and even modulate the EU agenda. In turn, Morocco's migration regime is an increasingly decisive piece within its international agenda (Jiménez-Alvarez et al., 2020). The strategy of individual EU Member States is not therefore only affected by the supranational policies of the EU, but also by the political manoeuvring of countries in the South, affected in turn by their own regional dynamics. The "crisis" of Ceuta in 2021, produced in a context of Moroccan diplomatic escalation to obtain international recognition of sovereignty over Western Sahara, showed us once again how migration and border control can become weapons through which countries in the South subvert the power relations with their diplomatic partners in the North.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 5 The Genealogy of the External Dimension of the Spanish Immigration Regime: When a Bricolage National Policy Becomes a Driver of Europeanisation**

**Lorenzo Gabrielli**

#### **5.1 Introduction**

Immigration policy is a privileged prism of analysis used to understand the dynamic of policy and norm diffusion and circulation, as well as regime-formation in Europe. In particular, the Spanish immigration policy regime, and more specifcally its external dimension, constitute a relevant case to understand how a national policy may infuence the development and the implementation of an immigration regime at the EU level.

Until the 1980s, labour was a major Spanish export but by the 2000s, had become a major import, indicating Spain's gradual transformation into a country of immigration. The timing of this change indicates an increase of migration issue's importance in Spanish politics, much later than in "traditional" migrant destinations in Europe. However, the transformation of immigration as a main political issue in Spain was swift and particularly intense (Gabrielli, 2011a). In Span, the issue of immigration assumes a special dimension because of the country's geographical location on Europe's external border as well as that of its enclave territories, Ceuta and Melilla, the only two terrestrial borders between the EU and Africa. European accession, the entry into the Schengen Area, the progressive securitisation of immigration at internal level, and the more recent evolution of immigration fows have decisively pushed Spain to take on a gatekeeping role within the EU. In the span of 30 years, the country has emerged as a central actor in the European externalisation of migration control toward third countries in Africa.

This chapter will assess the genealogy of practices, logics, and organizational strategies that fostered the development of the Spanish immigration regime, and in

© The Author(s) 2023

L. Gabrielli (\*)

GRITIM –Pompeu Fabra University, Barcelona, Spain e-mail: lorenzo.gabrielli@upf.edu

C. Finotelli, I. Ponzo (eds.), *Migration Control Logics and Strategies in Europe*, IMISCOE Research Series, https://doi.org/10.1007/978-3-031-26002-5\_5

particular, the external dimension of its migration policy.1 The progression and development of the external dimension of the Spanish immigration regime is evident in the prism of multilevel relations between the Spanish government and the EU. The main research question investigated concerns which types of infuence and model circulation exist between the national and the supranational/intergovernmental level.

The study of the different phases of the construction of the Spanish immigration regime, from its exogenous inputs to a more endogenous development (Gabrielli, 2011a), points out that Europeanisation2 of immigration policy is not always a topdown and unidirectional process from the EU to Member States. In this respect, beyond all changes attributable to the EU that take place on the national scene (Börzel & Risse, 2000; Radaelli, 2003), interactions concerning the external dimension of immigration policies may work in an opposite way: a bottom-up infuence from Member States to the EU.

In the analysis, the concept of regime is applied following Finotelli and Ponzo (see Chap. 1) and Cvajner et al. (2018, p. 17), as a tool to explore the constancy of a "cluster of migration policy action" as a result of complex and dynamic negotiation between different actors with their own specifc logics and interests. This idea is crucial in understanding how Spanish actors modulate and exert control and infuence on Europeanisation in a functional way as to pursue its own interests during each specifc period.

Since the late 1990s, several scholars broke the silence on the Europeanisation of immigration policies from different angles: Baldwin-Edwards (1997) on the emerging European immigration regime and the implications for Southern EU countries; Huntoon (1998) on the implications of EU immigration policy for Spain; Guiraudon (2000) on the interplay between Member States and the EU; Geddes (2001) on relations between the EU and state sovereignty concerning the issue of international migration. However, a more recent analysis on this topic is needed, especially in the wake of research published by Wihtol de Wenden (2019), motivated by a lack of knowledge on the more complex confgurations of Europeanisation embedded in immigration policies, and particularly in their external dimension.

To examine the genealogy of the Spanish migration regime's external dimension, the three phases are differentiated, each of which is characterized by its own specifc

<sup>1</sup>The 'external dimension' is the component of migration governance that goes beyond national borders, and exists at the crossroads between migration governance, foreign affairs and international relations. It generally addresses a transfer of migration management and control outside Europe through different instruments, and implemented through collaboration with third countries, or with regional or sub-regional organisations.

<sup>2</sup>The process of "Europeanisation" is defned by Radaelli (2003, 30) as the "processes of a) construction b) diffusion and c) institutionalization of formal and informal rules, procedures, policy paradigms, styles, "ways of doing things" and shared beliefs and norms which are frst defned and consolidated in the making of EU public policy and politics and then incorporated in the logic of domestic discourse, identities, political structures and public policies". Here, Europeanisation is analysed in terms of the external dimension of immigration policies.

dynamics concerning both the Spanish approach to immigration and border control, and the role of Spain in European policymaking on this issue.

First, the chapter focuses on the "top-down Europeanisation" of migration policy, where Spain can be considered a passive receiver of European norms and standards, pushed by the conditionality of Spanish accession to the EU and later to the Schengen Area. In this frst period, 1985 to 1999, European conditionality in immigration and border control impacted the national regime of immigration policy. In this context, Spain swayed between its European obligations and national apathy, which was related to the very low relevance of the issue within its national political scene.

Second, beginning in the 2000s, Spain became an active player – jointly with other Southern EU countries such as Italy – in fostering changes and new developments for the EU regime, mainly through a strong emphasis on border control towards irregular crossings. Spain's increasingly important role in European policymaking on immigration fostered a more bottom-up Europeanisation approach.

Third, through the development of the external dimension of its immigration policy in the mid 2000s, Spain emerged as a model and inspiration for the external dimension of migration policies implemented at the EU level in the 2010s. Paradoxically, during this phase of strong infuence, Spain effortlessly obtained political rents from previous initiatives, operating within a previously designed external dimension of its migration policies. In this third period, Spain swayed between a political marginality under the *Partido Popular* (PP) regime (amid fallout due to the 2008 global fnancial crisis) to a more symbolic, progressive stance under a coalition government of leftist political parties. However, the hegemony of the repressive focus of immigration at the EU level was not questioned despite the shift of power between political parties in Spain.

Methodologically, this study at hand is based on a long-term qualitative policy analysis developed during the two last decades through an extended document analysis based on administrative documents, grey literature, media, as well as through qualitative interviews with stakeholders, to understand and interpret policies as well as practices. The research within this chapter considers both elements, taking into account the very strong contradictions between formal policy and legal framework on one hand, and more informal practices deriving from their implementation3 on the other.

<sup>3</sup> In the feld of external dimension of migration policy, this is due to several facts: the implementation happens outside the Spanish and European territory, fact that diffcult scrutiny and accountability, and the negotiations occur in the feld of international relations, often using diplomatic tools determining then a growing informalisation of the agreements (Cassarino, 2007).

#### **5.2 Top-Down Europeanisation: Spain's European Obligations and National Apathy**

The existence of substantial interdependencies between Madrid and the EU has been clear since the very beginning of immigration policy in Spain, when the frst immigration law was approved in 1985 (LOEX – Ley *Orgánica* 7/*1985)*. The birth of a legal framework for migration in Spain was not the result of an endogenous political need or social demand to regulate the phenomenon – immigration was very marginal at that time, and the issue had a very negligeable relevance on the national political scene (Dios Pintado, 2005) 4 – but was framed as a precondition for Spain to join the EU (Gabrielli, 2011a). Moreover, the idea of Spain as a transit country for migrants led some EU countries to have outweighed concerns about the facilitated entry of immigrants.

Spanish immigration policies developed according to a restrictive model of immigration policy, strongly inspired by the concerns of key EU member countries at this time. A strong dichotomy characterised Spanish immigration policy in this frst phase, similar to other Southern EU countries, wavering between a more "European-inspired" legal framework, repressive of immigration, and permissive national practices that overcame the lack of a formal channel of recruitment for foreign labour – coupled with cyclical regularisation processes (Arango, 2000; Peixoto et al., 2012).

Once the LOEX was approved in 1985, European pressures were temporarily mitigated due to the low intensity of migratory fows into the country during the second half of the 1980s. European interest in Spanish immigration control was revived when Spain signed the Schengen Agreement in June 1991. At that time, European pressure targeted the issues of visa policy and border control, given that the Spanish border was now a key external border in the Schengen Area. Spain's entry into the Schengen Area implied additional responsibilities, mainly refected in the introduction of the entry visa requirement for Moroccan, Algerian and Tunisian nationals since 1991.5

Spain's admittance to the EU and to the Schengen Area in 1996 initiated a physical integration of Spain into the European regime of immigration control. As noted by Guild and Bigo (2003), Southern European countries, i.e. Italy, Greece and Spain, found themselves in an ambiguous position *vis-à-vis* their European partners (mainly France and Germany) who were the most interested in the issue. On one hand, they wanted to avoid criticism for their *laissez faire* policies, and on the other hand, they were ready "to take symbolic measures", while they were "not prepared to launch very costly and probably not very effective programmes to control their

<sup>4</sup>This for two reasons: frstly, the phenomenon is quantitatively small and its visibility extremely limited; secondly, when developing the legal framework of immigration in Spain, the issue of immigration is very weakly politicised.

<sup>5</sup>Tensions arising with Morocco on this issue were mitigated by a targeted extraordinary regularisation program launched the same year.

coasts" (*ibidem*, p. 13). In this context, with signifcant European funding during the 1990s, the Spanish government began to build and fortify border fences in its territories Ceuta and Melilla, as well as install the Integrated System of Exterior Surveillance (SIVE) along the Mediterranean corridor, starting from the Strait of Gibraltar (Gabrielli, 2011a).

At the same time, Spain began developing the external dimension of its migration policies, concluding with a readmission agreement with Morocco in 1992, allowing readmissions of citizens from signatory countries but also from third country citizens. After signing, Morocco declared that it would not implement the agreement, and readmissions were to be negotiated on a case-by-case basis until 2018, when the agreement was fnally "re-activated" according to declarations of the Spanish government (Ferrer-Gallardo & Gabrielli, 2018; Gabrielli, 2015; Serón & Gabrielli, 2021).

In summary, between 1985 and 1999, the effect of the Europeanisation of the Spanish migration regime and its external dimension occurred in a rather "passive" way in Spain, in the sense that the government limited itself to seeking a balance between its European obligations and its own interests.

#### **5.3 The Transition: Spain's Ascent to an Active Role in the Development of the External Dimension of the European Immigration Regime**

An important change took place in political debates in 2000, refected in legislation that affected both Spanish immigration policy as well as the Europeanisation in the external dimension. At that time, the visibility of immigration in the political debate was the most evident and the development of the external dimension of Spanish immigration policy gained considerable momentum. Europeanisation became a less vertical process and the Spanish government was no longer a passive receiver of the political guidelines in Europe. Progressively, it became a central player in shaping immigration policy – or at least some of its dimensions – in Europe. This shift, following a key change concerning migration on the internal political level, was facilitated through a strong focus on irregular crossing and border control and externalisation of migration control.

In 1999, after approximately 15 years of political and legislative inactivity, the Spanish Parliament initiated the frst real debate on immigration. After a broad consultation process with social actors, a new consensual and pragmatic legal framework was developed.6 Shortly before the scheduled approval in the Parliament, the

<sup>6</sup>This law recognises that immigration is a structural element of the Spanish reality including the integration of immigrants, which is considered the fnal objective of all public administrations. The law aims to avoid, or at least remedy, the situations of irregularity through an ordinary individual regularisation procedure, and by excluding expulsion as sanctioning system. The law also establishes equality rights between nationals and foreign 'regular' residents (except the right to suffrage

centre-right government decided to oppose the law, justifying the sudden change of opinion with the fact that EU will not support such a progressive policy. The new migration law (4/2000) was fnally approved in January 2000, thanks to the support of all the opposition parties needed for a majority. However, this pushed immigration to the forefront of the political arena. The reform of the law was one of Aznar's priorities in the campaign of March 2000, breaking an existing and tacit pact between different political forces to not use immigration as an electoral tool, and instead, using a securitisation narrative, which was already common at the EU level. The absolute majority of the PP formed as a result from the March 2000 election strengthened the securitising trend.

After 15 years of inactivity, the approval of a more restrictive legal framework around immigration (Law 8/2000), the second in the same year, framed immigration as a threat to security. Deeply affected by the election defeat, an unexpected "community of discourse" between the PP and the *Partido Socialista Obrero Español* (PSOE) resulted in a restrictive approach on immigration, targeting border control, visa policies, immigrant detentions and deportations. This determined a "depolitisation" of the issue, limiting the debate to marginal elements of the policy and reduced the migration issue to a security threat (Cuttitta, 2018; Ritaine, 2003). The consolidation of immigration as a security issue fostered a growing role of the Ministry of the Interior (Zapata-Barrero2003), not just in internal matters, but also in the feld of external dimension (Casas-Cortes et al., 2016), driving a stronger emphasis of police cooperation with third countries.

Concerning relations at the EU level, the Spanish alignment with the hegemonic securitisation of immigration at the EU level foster a less top-down process of Europeanisation (Guiraudon, 2010). This can be considered the prelude to the gradual transformation of Spain into a central player in the immigration policymaking in Europe. Driven by the need for a political and, above all, fnancial support for its initiatives to strengthen border control apparatus and developing cooperation with third countries, Spain became increasingly involved in the development of immigration policy in Europe. Immigration control at borders and its externalisation became pivotal elements through which the Spanish government pushed its priorities into the European agenda.

This became more evident during the Spanish presidency of the EU in the frst half of 2002. In a global context framed by the 9/11 terrorist attacks in the United States and a signifcant pivot to securitisation in international policy, Spanish Prime Minister J.M. Aznar had a favourable environment to push his immigration agenda through to the European Union.

On June 21 and 22, the issue of immigration was at the forefront of the Seville Council, especially in regards to relations with third countries, formally planting a seed to further develop the external dimension. At the Seville summit, Spain proposed to link cooperation on immigration by third countries (at this time, mainly

and access to public offce), recognising the right to family reunifcation, the right to free legal assistance in any administrative procedure that may involve expulsion, and the right to healthcare and education for foreigners in an irregular situation (Gabrielli, 2011a).

the sanctioning of third countries through suspending aid when their cooperation is not considered satisfactory by EU countries. Although the most radical component of Spain's proposals, the possibility of establishing a negative conditionality through sanctioning third countries had been rejected. The Spanish Presidency had played a fundamental role in fostering the development of the external dimension and giving a different connotation of the EU security-driven focus, increasing the relevance of irregular crossings at the borders in the community agenda7 (Pinyol Jiménez, 2007; Terrón, 2004). Furthermore, a particularly close community of interests existed in Italy at this time (Ritaine, 2003) considering the centre-right Berlusconi government supported Aznar in applying pressure to European institutions on the basis of ideology but mainly for pursuing the common interests of both Spain and Italy, as gatekeepers at the Southern European border. Progress was particularly signifcant in certain repressive aspects of immigration policy which had the approval of the Spanish government: support in maritime border control through the launch of Operation Ulysses coordinated by Spain and sharing the fnancial burden of border control.

As a result of this infuence and execution of its agenda, Spain was regarded as a more relevant player in the EU immigration policymaking on border control and externalisation.

#### **5.4 Bottom-Up Europeanisation: Spain's Key Role in the Construction of the External Dimension of Immigration Policy**

Despite the growing control apparatus implemented at the increasingly militarised borders,8 the control of immigration by Spanish public authorities was questioned due to a growing media visibilisation of borders crossings. In the summer of 2005, an attempt to jump over the fences surrounding the enclave of Ceuta resulted in fve deaths and become a watershed of the country's recent migration history. In 2006, 32,000 people came into Spain via the Canary Islands, which was widely covered in

<sup>7</sup>For more information, see the "Global Plan to Combat Illegal Immigration and Traffcking in Human Beings in the European Union", adopted by the JHA Council (28 February 2002) which highlights the need to strengthen maritime border controls and to carry out joint control operations. The Commission also stresses that "external borders still appear, rightly or wrongly, as a weak link that may affect the level of internal security of the Member States, especially in an area without internal borders" (European Commission, 2002). The JHA Council of 13 June 2002 approved the "plan for the management of the external borders of the Member States of the European Union" control.

<sup>8</sup> In this respect, it is suffcient to recall the implementation of satellite-based maritime vigilance systems and the SIVE (Integrated System of Exterior Surveillance) in the Strait of Gibraltar, on a large part of the Andalusian coast and subsequently, around the Canary Islands.

the European media. These highly visible and symbolic events are seen as catalysts that gave a decisive boost to the development of the external dimension of Spanish immigration policy, by means of an intense diplomatic action in Africa to enhance relations with third countries and facilitate the externalisation of immigration control.

The strong focus of the Spanish government on these fows of migration, despite the limited numbers of the arrivals from sub-Saharan Africa compared to the total number of foreigners entering in Spain in the same years (Gabrielli, 2015), is related to increased media exposure and therefore to their symbolic relevance (Gabrielli, 2021), but also on the European political scene. The over-visibility of specifc entry points at Spanish borders allowed the government to put in place a visible action towards immigration and immigrants, while not overlooking the labour needs of the Spanish economy which was in dire need of foreign workers. This allowed the government to increase its legitimacy *vis-à-vis* EU institutions when campaigning for more monetary support.

Initially, the impulse to externalise control targeted Morocco but soon afterward, the externalisation of the control of migratory fows spread to other countries. The displacement of migratory paths towards the Canary Islands, migrants departing from the West-African coast pushed the government to act, resulting in the presentation of the "Africa Plan",9 an attempt to re-fuel diplomatic activity with Sub-Saharan countries. Now, the external dimension of Spanish immigration policy developed very quickly, alongside directives regarding the "Africa Plan" in 2005 but was further developed as a case-by-case bricolage, depending on migration routes in third countries and their demands in order to collaborate with Spain with forced returns, controlled maritime avenues, bilateral patrols, liaison offcers and information-sharing.

Spain negotiated bilateral agreements and more informal cooperation with Morocco, Mauritania, Senegal, Gambia and Cape Verde – covering all crossing points into Spain – as well as with several countries along the main land migration routes and origin countries in West Africa (see Table 5.1). Spain developed a variable geometry in its cooperation with these countries in order to deploy a buffer of mobility and to increase the scope of deportations: using both formal and informal agreements on readmission and redocumentation,10 but also through stipulations of broader agreements, including control of migration departures and "transit migrations", as well as operational cooperation in exchange of information exchange,

<sup>9</sup>The Africa Plan of 2006–2008, launched in May 2006, sought to establish a new, deeper framework of Sub-Saharan African relations through several lines of action: reinforcing Spain's political and institutional presence in Africa, fostering cooperation between Spain and African countries in the "regulation of migratory fows", and a more active participation of Spain in EU Strategy for Africa, which involved other more conventional development aid objectives (MAEC, 2006).

<sup>10</sup>The process of targeting the engagement of origin countries to quickly verify the nationality of migrants without documentation, and producing new temporary documentation that facilitated their forced return. This was the case with many migrants from Senegal that arrived in the Canary Islands in 2005, forcing the deployment of offcers to the islands to help in redocumentation efforts.


**Table 5.1** Migration agreements (Readmission or broader issues) between Spain and Sub-Saharan African countries (2000–2010)

Source: Author (In the highly informal framework of migration agreements, it was essential to carry out complex and in-depth reconstruction work through press sources in order to understand not only the negotiation process of this externalisation, but also the content of the agreements and the link with development aid concessions)

liaison offcers and joint patrol operations (Andersson, 2014; Casas-Cortes et al., 2016; Gabrielli, 2008, 2011a).

A central element of the Spanish external dimension was the "good will" of African counterparts, especially concerning readmissions, but since the mid 2000s, considerably widening the scope of cooperation in the exchange of information, presence of liaison offcers, participation to bilateral patrols, and the control of "transit migrations" (Gabrielli, 2011a, b). These "second generation" agreements have been implemented through strong conditionalities, established Spanish control of migration, as well as other elements of the bilateral relation. The involvement of African countries, thus, becomes a discriminating element when it comes to facilitating trade, boosting foreign investment and allocating development aid – all of which are considered the tools used to foster a cooperation that would otherwise be very unfavourable to participating African countries. In some cases, the creation of formal recruitment channels for seasonal workers were also used, as in the case of Senegal and Morocco. In Senegal, the implementation was temporary and quantitatively negligible, while in the case of Morocco the recruitment of woman seasonal workers for red-fruit collection had been more longstanding (Arab, 2018).

In the case of cooperation with Morocco, negotiations were more complex due to historical relations between the two countries and the balance of power (Carrera et al. 2016; El Qadim, 2015; Ferrer-Gallardo, 2008; Gazzotti, Jiménez Álvarez and Espiñeira, this volume). The instruments used were broader, on issues such as the trade agreement, or on the case of Western Sahara, including Spanish support at the EU level. In the case of Sub-Saharan African countries, the main tools of negotiation and implementation of bilateral agreements are generally linked to development aid (Gabrielli, 2011a, 2016; Kabbanji, 2013).

#### **5.5 Spain: A Model for the External Dimension of Immigration Policies in Europe?**

Through the development of the external dimension of the Spanish immigration policy, Spain emerged as an infuence and model to other European Member States, or at least a paradigm of an effective externalisation action for its effort with Sub-Saharan Africa. In this regard, Spain is considered a strong inspiration for the development of EU action (Gabrielli, 2017b; Serón & Gabrielli, 2019). In general terms, Spain is not a pioneer – as Italy was the frst Member State to implement this externalisation strategy. However, several reasons may explain why the Spanish immigration policy regime, and in particular its external dimension, became so inspiring for EU institutions when developing the recent wave of externalisation.

First, Spain is the main European actor to have signed agreements with Sub-Saharan African countries on the externalisation of control in the same manner as Italy which established agreements with Mediterranean countries such as Albania, Tunisia and Libya (Paoletti, 2012). In this case, there is no direct route from Sub-Saharan Africa: France has historically been a major actor in concluding bilateral agreements with Sub-Saharan African countries, which are mainly former colonies. However, the French efforts since the mid 2000s mainly addressed readmission and was linked to co-development and development aid (Panizzon, 2011).

Second, Spanish action in externalisation matters is inspiring for the EU because of a de facto introduction of a "migration conditionality" in relation to third countries, which links cooperation in migration issues with development aid, as well as with other aspects of relations such as foreign direct investment (FDI), visa quotas, etc. "Migration conditionality" was clearly emphasised in negotiation and implementation of the external dimension on behalf of the "goodwill" of third African countries. In exchange for their cooperation, the countries that accepted externalisation were able to obtain different compensations. Regarding migration control, Spain supported by the EU, provided material and fnancial assistance through the provision of control equipment and police and border guard training, but also in entry quotas for seasonal workers, FDI and more.

Third, despite growing analytical evidence underlying the real effect of these policies,11 there was and still is a diffuse perception by policymakers that Spanish outsourcing was fnally an effective model to reduce immigration. Since the peak of migrant arrivals in 2006, pirogues crossings to the Canary Islands are almost nonexistent (at least until 2019 and 2020) and informal crossings – in Ceuta and Melilla and the Strait of Gibraltar – seemed to remain relatively low in the following years.

In April 2016, Spain's Interior Minister Jorge Fernández Díaz, explained that the EU was copying the "Spanish model" in terms of its migration policy. At the December 2016 European Council, Spain's Prime Minister Rajoy underlined the continuity of the "Spanish model" for migration policy, as promoted by Rodríguez Zapatero, which was of particular interest at that time for the EU due to the eagerness to negotiate migration agreements with countries such as Senegal, Mali, Niger, Nigeria, and Ethiopia. The director of Frontex himself pointed to Spain as an inspiring example and leader in this regard, praising the effectiveness in "closing" the Canary Islands route after arrivals arose in 2006.12

Whether or not Spain is an ideal "model", it is clear that the Spanish externalisation of migration control was a great source of inspiration for the recent wave of externalisation promoted in the EU after the Arab Spring and after the "refugee crisis" in mid 2010. This period was marked by a relevant relaunch of supranational initiatives at the EU level.13 Since the "Arab Spring", EU policy towards third countries has been revitalised, maintaining the target of specifc sectoral lines, but executing them with new instruments and under the guise of a new geographical focus.14 For example, through the Global Approach to Migration and Mobility (GAMM) in 2011, but above all, through the Valletta summit of November 2015 which

<sup>11</sup>A larger geographical scope including all Mediterranean space and a longer timeframe should be considered to evaluate results and consequences. For instance, immediately afterwards implementation of Spanish externalisation, sea crossings from Libya to Italy increased again, indicating a reorientation of routes between Africa and the Mediterranean (Gabrielli, 2017a), and also that the effects would have to be analysed on a broader level, both geographically and temporally. At the same time, the "collateral" effects of these policies on the rights of migrants and refugees in transit spaces, as well as the violence to which they are exposed, should also be considered (i.e.: Andersson, 2016; Gabrielli, 2011b, 2016).

<sup>12</sup>El Español (2016, December 16), "Rajoy exporta a la UE el 'modelo Zapatero' de acuerdos migratorios": https://www.elespanol.com/espana/politica/20161215/178483123\_0.html

<sup>13</sup>However, the bilateral initiatives of EU countries – mainly towards the Mediterranean and Africa – have been and remain predominant in the external dimension of migration policy (Gabrielli, 2016; Kunz et al., 2011), with *ad hoc* support of the EU.

<sup>14</sup> In geographical terms, the external dimension of migration policy shifts its focus towards the central and eastern Mediterranean, given the current context of migratory routes and maintains its interest in sub-Saharan Africa as a "buffer space" for mobility. Eritrea, Sudan, Chad and Niger consolidated themselves as key targets to flter mobility before Libya, while Turkey, Lebanon and Jordan have also become priorities to further reduce the infux of refugees and migrants from Middle East, Central Asia and East Africa.

established the EU Emergency Trust Fund (EUTF) for Africa (Gazzotti, Jiménez Álvarez and Espiñeira, this volume)– and the EU-Turkey Statement of March 2016.15

In the Spanish case, the EU has decided to respond to perceived migration and refugee "crises" in a reactive and short-term manner through an increasingly intense and geographically widespread externalisation of mobility and border control. The objectives of cooperation that the EU and its members have sought to implement over the last decade are closely reminiscent of Spanish action: cooperation in redocumentation and readmissions, externalisation of mobility and border control. These instruments, when used in the negotiation seem very similar, with the obvious differences of geopolitical weight in negotiating on behalf of the EU as a whole with the different third countries.

The GAMM of 2011 reminds to content within migration agreements signed by Spain and several Sub-Saharan African countries, both in terms of the areas of cooperation (readmission, cooperation in controlling fows and strengthening border control) and the negotiation instruments (development aid and visa quotas).

Moreover, migration conditionality has become a recognisable element of the external dimension of migration policies in Europe (Gabrielli, 2011b, 2016). The negotiation with Tunisia and Libya, or with Mali, Ethiopia, Nigeria, Niger and Senegal in the framework of "migration partnerships" is representative of this trend (Gabrielli, 2016; Serón & Gabrielli, 2019). This is also the case in the EUTF for Africa, where the cooperation with Ethiopia facilitated a loan of USD 500 million to the Ethiopian government. In this regard, Francisco Carreras, the head of cooperation efforts at the EU Delegation in Ethiopia, clearly confrmed that "migration issues are at the heart of our support to countries".16

The EU seems to have gone further in several dimensions, frst by recognising the possible use of negative conditionality (EU Commission, 2016). This is the possibility of openly sanctioning a third country's "failure" to cooperate with the EU on migration, whereas, until now, only positive conditionality, i.e. incentives were recognised. Second, concerning the "migration-development nexus" and its translation into the EUTF for Africa, the mantra seems to be further reinforcing the process of "capacity building" within third countries in terms of border control and mobility (Gabrielli, 2016; Serón & Gabrielli, 2019).17

<sup>15</sup>For more information, see https://www.europarl.europa.eu/legislative-train/theme-towards-anew-policy-on-migration/fle-eu-turkey-statement-action-plan

<sup>16</sup> "Europe pays out to keep a lid on Ethiopia migration". IRIN (2016, October 24) https://www. irinnews.org/analysis/2016/10/24/europe-pays-out-keep-lid-ethiopia-migration

<sup>17</sup>This is the case, for example, of the funding of an electronic passport in Senegal in 2007, as well as the new implementation of the ECOWAS biometric card system in the same country in 2016, through an EU-funded project carried out by the IOM. Similar projects are also encouraged in the case of Mali. The digitisation of population registers and the identifcation system is an issue that can hardly be considered as a real priority for the development of local societies. This is also the case of some programmes of the EUTF for Africa Trust Fund, such as GAR-SI Sahel (Groupes d'Action Rapides – Surveillance et Intervention au Sahel or WAPIS (West African police information system) managed by INTERPOL, among others.

At times, development aid alone is not enough to encourage cooperation from third countries (Gabrielli, 2011a, 2016; Kabbanji, 2013), leading to the mobilisation of other instruments. The externalisation of migration control, both at the bilateral and multilateral levels, becomes increasingly embedded in a broader framework of political dialogue or intrinsically linked to other external relations issues: special visa regimes, visa quotas, diplomatic support, investment pledges, trade agreements, investment, fnancial aid packages and security (Jurje & Lavenex, 2014; Kunz, Lavenex, & Panizzon, 2011; Lavenex & Kunz, 2008). In the EU-Turkey Statement18 of 2016, several clauses closely resembled Spanish externalisation – in particular, the evident issue linkage between externalized tasks of buffering mobilities and accepting deportations on one side, and monetary support, visa issuance and an improved political and commercial relation from the other. The novelty is that, in this case, all policies are clearly visible, as they are explicitly mentioned in the agreement.

As in the case of Spain's externalisation, outsourcing creates an increasingly strong European dependency on neighbouring countries, which have since become key players in mobility control systems (Greenhill, 2016; Zaragoza-Cristiani, 2016). When a third country situated along major migratory routes wants to pressure Europe or its Member States to gain advantages in other policy areas, it can utilize and deploy the issue as very effective leverage. This was the case in the arrival of more than 10,000 migrants to Ceuta over the course of 1 day in May 202119 as well as by the Turkish President Tayyip Erdogan's opening of the Turkish border with Greece on February 28, 2020 when he declared that "hundreds of thousands have crossed, and soon we will reach millions",20 and lastly, by human crossings at the Poland-Belarus border in November/December 2021.

The exercise of "bricolage" in developing the external dimension of immigration policy and tailor-made agreements depending on soft-law instruments (memorandum of understanding and the exchange of offcial communication) appear to be an effective model within the EU due to the specifc evolution of migration routes and avenues of negotiation. Spain's contribution has become central to the defnition of Europe's external dimension, towards Africa and beyond. In terms of Europeanisation processes within immigration policies, Spain's role changed from that of a "normstaker" to that of "norms-maker". This evolution deserves to be taken into account

<sup>18</sup>See https://www.consilium.europa.eu/en/press/press-releases/2016/03/18/eu-turkey-statement/

<sup>19</sup>The Spanish position concerning the issues presented by Western Sahara, a territory occupied by Spain until 1975 and since, two-thirds of the territory is under de facto control by Morocco, and one-third controlled by the self-proclaimed Sahrawi Arab Democratic Republic with the support of Algeria. In early 2021, a Spanish hospital received the leader of the Polisario Front in disguise who was ill with Covid, which triggered important consequences, such as major arrivals of migrants on May 15–16, due to a "relaxation" of Moroccan forces controlling the common border (some sources also suggest a more active role in fostering crossings).

<sup>20</sup>For more information, see https://www.dw.com/en/erdogan-warns-millions-of-refugees-heading-toeurope/a-52603580

when analysing the circulation of policy models and tools, as well as in consideration of the forms that the Europeanisation process can take.

#### **5.6 Conclusions: Reaping the Benefts?**

The analysis of the external dimension of the Spanish migration regime since its inception in 1985 has resulted in a new perspective necessary to grasp the different dynamics of Europeanisation in the outsourcing of migration control towards third countries.

As the research has shown, this process has taken different forms at different times, both in Europe and in Spain. Until the early 2000s when the issue of migration suddenly appeared on the national political scene, Europeanisation acted as a top-down mechanism, mainly producing the transfer of a very restrictive legal immigration framework from the EU to Spain. Spain's entry into the Schengen Area brought about a change in visa policy, but did not alter the already-existing ambivalence between the restrictive legal framework and the policy actually implemented by public authorities. There are, therefore, two contradictory forces shaping Spanish immigration policy: on one hand, Spain's role as a Member State within the EU and guardian of the external border of the Schengen Area, and on the other, the national dependency and demand for foreign labour.

The Spanish government's autonomy of action in the area of migration is thus emerging. This is a constant element in the different phases of the research analysis, as Spanish actors modulate Europeanisation in a functional way in pursuit of their interests.

In 2000, the migration issue was suddenly transformed into a major political debate and became a key segment of the national security agenda in government rhetoric, whereas previously it had been considered a labour market issue. The Spanish government's adherence to the securing of migration issues drove an empirical change in the Europeanisation dynamic. This process is thus beginning to change its axis, pivoting from top-down to a bottom-up since the political transfer is bi-directional due to the emergence of Spain as an increasingly important actor in European policymaking on immigration. Due to its geographical location and its own internal political dynamic, Spain, together with Italy, has applied political pressure on other European Member States to support its tasks of controlling the common European border in the face of migratory fows.

With the development of the external dimension of Spain's immigration policy, the country has become a leading inspiration for the further development of the issue at the EU level. Paradoxically, the Spanish regime was developed in an improvised way, through a bricolage process following a main driving force: a reactive and crisis/emergency-based logic. More specifcally, Spanish inspiration has come at three levels: the tailor-made nature of the agreement depending of the conjuncture and characteristics of third countries, as well as informality and fexibility; the objectives of migration cooperation (re-documentation and readmissions, externalisation of border and mobility control, operational cooperation in terms of information exchange, liaison offcers and joint patrol operations); and the negotiation instruments (issue linkages between externalisation and other feld of relation, special development aid, and through a growing migration conditionality).

Just after Spain joined the EU, it was the recipient of the standards produced by Europe, and 20 years later, it became a pillar with a central role in defning European guidelines on the control of migratory fows. This is a particularly signifcant and rapid change, given that Spain is now at the heart of the defnition of European immigration and border policy, especially concerning Euro-African relations.

Ironically, during the 2010s, while the external dimension of the Spanish immigration regime was establishing its infuence at EU level, Spanish action occurred in the background. While the main crossing routes of migrants and refugees in the Mediterranean were mainly addressed to Italy and Greece – despite some conjunctural "crisis" in Ceuta and Melilla in mid-May 2021 – Spain took a more discrete position in EU policymaking and reduced its bilateral efforts towards third countries. In particular, budgets dedicated to the bilateral external dimension were cut, especially in the feld of development, cooperation (Serón & Aimé, 2020) and foreign affairs, while those in the feld of security and police cooperation through the Interior Ministry were maintained (Serón & Gabrielli, 2021). In this period, immigration slipped off the political agenda and Spain swayed between a political marginality under the PP government of Mariano Rajoy in 2011–2016 (Montilla Martos et al., 2017) to a more symbolical, visual and progressive stance under left coalition governments (with punctual support to Search and Rescue NGO activity in the Central Mediterranean, as is the case in the welcoming of the Aquarius boat in the summer of 2018). The onset of the COVID-19 pandemic has seemed to further reduce the relevance of the issue, at least during the onset.

Moreover, the multilateral efforts of the EU in the externalisation of migration control, as well the growing engagement of other EU Member States, present Spain with an opportunity to recollect as much political, diplomatic and economic benefts as possible, despite its minimal effort. Spanish political know-how in these felds allows the country to reap the benefts of past efforts, and receive important funding through leading and managing several projects of the EUTF for Africa, the GAR-SI Sahel endowment of over 41.6 million euros and the government-controlled *Fundación Internacional y para Iberoamérica de Administración y Políticas Públicas* (FIIAPP) foundation together with the Guardia Civil.

Nevertheless, this situation changed suddenly in 2020, when around 23,000 people arrived to the Canary Islands over the course of the year, and again in May 2021, when high diplomatic tensions with Morocco on the Western Sahara issue arose and led to more than 10,000 people crossing the fences that separate the Spanish enclave of Ceuta with Morocco. These new episodes, framed once again as "migration crises", have pushed the Spanish government to give a new impetus to external action and to resume a more active role at the EU level regarding policymaking on immigration. It will undoubtedly be interesting to analyse in the future what the consequences will be for the external dimension of immigration policy.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Part III Internal Controls**

# **Chapter 6 Challenges and Ambiguities of the Policies for Immigrants' Regularisation: The Portuguese Case in Context**

**Jorge Malheiros and João Peixoto**

#### **6.1 Introduction**

Regularisation of undocumented immigrants is part of the Portuguese panoply of mechanisms constructed to deal with migration since the early 1990s. Despite the change in the paradigm of the regularisation processes that took place in 2007 (from extensive extraordinary regularisations to case-by-case ones), the maintenance of such procedure and its relevance as a mechanism that migrants have to accede rights seems to point both to the ineffciency of the formal immigration channels and a certain normalisation of irregularity in Portugal – even if the existing clues point to a decrease in the number of irregular foreigners in comparison to the situation experienced in the beginning of the 2000s. This picture supports the idea of a systematic lax attitude towards informality and migration control, which corresponds to components of the supposed common migration regime of the Southern European countries (Finotelli, 2009).

In the frst part of this chapter, through frame analysis of the evolution of regularisation mechanisms in Portugal since the early 1990s, we try to uncover the motives behind the successive devices and to discuss the political interactions that supported them, from the political consensus dominant until 2012 to the evidence of fragmentation and politicization taking place afterwards. For this, we look to the legal instruments issued between 1992 and 2007 that opened extraordinary windows of regularisation for immigrants in Portugal. Then, we complement this analysis with an overview of the additional legal diplomas issued after 2007 that, in different ways, impact on regularisation issues, now in a novel framework based on

J. Malheiros (\*)

J. Peixoto

Centre of Geographical Studies (CEG), University of Lisbon, Lisbon, Portugal e-mail: jmalheiros@campus.ul.pt

School of Economics and Management (ISEG), University of Lisbon, Lisbon, Portugal e-mail: jpeixoto@iseg.ulisboa.pt

C. Finotelli, I. Ponzo (eds.), *Migration Control Logics and Strategies in Europe*, IMISCOE Research Series, https://doi.org/10.1007/978-3-031-26002-5\_6

case-by-case analysis. The reading of the laws has been complemented with an analysis of the Parliamentary voting behaviour of the parties and with statistical elements concerning the number of people that applied in each regularisation. In order to frame the process into the economic evolution of Portugal in this period, some data have been collected, namely the unemployment rate.

The aforementioned inputs allow us to confront the issue of regularisation – a key element in the Portuguese immigration policies – with elements pulled from other European countries, based in specialized bibliography, media references and analysis of some documents concerning regularisation processes. This enables us to position regularisation into the debate about the divergence or convergence of migration regimes, considering specifcally the Southern European (including Portugal) and the North-Western European ones.

In the second part of the chapter, we move from regularisations and migration policies to the broader perspective of migration regimes, a concept that has been marked by a fuid and even polysemic use. We use the ideas of Rass and Wolff (2018) and Cvajner et al. (2018), who see migration regimes as a process combining regulation and action in a migration governance prospect, involving a whole set of interrelated actors bounded by necessary and/or contingency interactions. This includes the notions of unequal power relations and access.

With this in mind, we align our perspective with the conceptual guidelines of this book. We develop an argument that challenges the aforementioned categorisation of the European migration regimes (which also includes a third one: the Central and Eastern European – Arango, 2012), calling upon different commonalities and distinctions that may change with the contexts. Even if shared contextual elements such as recent political histories, economic restructuring processes, dominant labour market features, and particularities of welfare state regimes frame migration regimes and allow a basic clustering of European countries, we argue, using the Portuguese case, that these regimes produce continuums, and their borders are more fuid than rigid. Specifc political cycles are relevant in understanding the "regularisation options," namely at a juncture marked by increasing politicization of immigration issues.

#### **6.2 Regularisations in Portugal: In Search of a Policy**

#### *6.2.1 A Series of Policy Measures*

Portugal is often considered within the overall framing of the Southern European case, yet its specifc context remains distinct (Baganha & Peixoto, 1997; Malheiros, 2012).

Contemporary foreign immigration to Portugal began in the 1980s, in part due to the general factors that explained its increase elsewhere, but also due to specifc national circumstances resulting from the country's move from dictatorship to democracy in 1974, and the following decolonisation. The bulk of the new foreign infows came from the African ex-colonies, particularly Cape Verde, later joined by Brazil. An immigration landscape built around a common language and former socio-political ties was a landmark of the country from the beginning. Democracy, the new immigration context and the path towards the adhesion to the European Community (which materialised in 1986), explained the need to draft the frst immigration law, launched in 1981. Part of its rationale derived from the prospective EU obligations, which required a strict control of international borders. But, as will be further described in this article, the series of policy initiatives that were enacted afterwards are largely specifc to the Portuguese case.

Immigration policy in Portugal has been the object of several studies which highlight its main traits and framework (see, among others, Baganha, 2005; Fonseca et al., 2005; Carvalho, 2009, 2018; Peixoto et al., 2009; Acosta Arcarazo, 2013; Padilla & França, 2016; Sampaio, 2017). The object of this section is not to review in detail all such developments, but rather to focus on the measures targeting irregular migration.

The policies enacted to tackle irregular immigration were exemplary in the need to face new challenges, using both already-tested mechanisms and novel policy initiatives. The complete list of regularisation measures adopted by the Portuguese governments is presented at Table 6.1. The policy solutions have varied, and many have been adopted regardless of the political orientation of the governments (left or right wing). Between the early 1990s and the mid-2000s, the principle of extraordinary regularisations was adopted. They differed from one other: some consisted of


**Table 6.1** Main regularisation measures

Source: Adapted from Sabino et al. (2010)

general amnesties, others relied on economic conditions and others depended on bilateral agreements. Since 2007, extraordinary regularisations were replaced by an ongoing case-by-case model, which remains as a solution for the structural problem of irregular migration until today. The ongoing regularisation model has survived several governments and major economic and social crises, including the fnancial turmoil of 2011–2014 and the Covid-19 pandemic.

The frst two policy initiatives were classic extraordinary regularisation processes and somehow mark the formal recognition of the importance of irregular migration. By this time, the migration turnaround of Portugal started to be visible: the decrease of emigration due to a shift in economic growth in the main destination countries of Europe since the mid-1970s and the subsequent adoption of restrictive policies was coupled with an increase of foreign immigration, largely a result of the decolonization process that took place in 1974–75. After the initial wave of *retornados*<sup>1</sup> in 1974–1975, successive waves of immigrants coming from the ex-colonies, now turned foreign citizens, came to Portugal. Many entered without an appropriate visa or overstayed, thus becoming irregular migrants. The improvement of the economic condition of the country, particularly after joining the European Community in 1986, favoured this movement. A large coalition of interests, described ahead, created the conditions for the regularisations of 1992–1993 and 1996, granting legal status to approximately 39,000 and 35,000 individuals, respectively.

A third process of regularisation, the largest until today, occurred in 2001. This time it was far from a classic regularisation process. It started with the acknowledgement of a new wave of foreign immigrants arriving since the mid-1990s, resulting from a new period of economic growth. Unlike the former infows, African immigrants coming from the ex-colonies were not dominant, but rather the Brazilians, and a new wave from Eastern Europe, primarily Ukraine. Some of these immigrants entered irregularly, but most of them arrived with tourist visas and overstayed. Despite their status, they were often recognized to be vital to fulfl labour shortages in sectors under expansion, particularly construction and personal services. The 2001 law did not grant automatic legal residence to these new immigrants. Instead, it created new "stay permits", which in practice corresponded to work visas conceded after arrival. The condition was presenting a labour contract or a promise of contract to the authorities. After 5 years of renewal, the new stay permits were reconverted into full residence permits. Under this legal framework, almost 184,000 immigrants were regularised.

The fourth and ffth processes of regularisation were also of a non-classic type. The fourth resulted from a special bilateral relationship with Brazil (Padilla, 2007). The "second wave" of Brazilian immigration was occurring since the late 1990s, the largest infow ever coming from this source (Malheiros, 2007). A special agreement was signed between the two countries in 2003, allowing the regularisation of approximately 16,000 Brazilians working in Portugal (much more than the irregular

<sup>1</sup>Portuguese or people with Portuguese ancestry that left the Portuguese colonies in Africa during the decolonization process and "returned" to Portugal.

Portuguese immigrants then living in Brazil, also entitled to regularisation). In 2004, a similar type of measure was extended to non-Brazilian immigrants, thus corresponding to a ffth regularisation. The objective was extending to all immigrants the rights that had been granted to Brazilians beforehand. In both cases, the regularisation depended on specifc conditions, namely presenting a labour contract or proof of having made payments for social security for a given period – not considered problematic given most irregular immigrants were employed.

From 2007 onwards, a new paradigm emerged. Instead of granting regularisation on an extraordinary basis requiring the enactment of special processes for given periods of time, the government created a mechanism of on-going regularisation, which could be carried out at any moment. The procedure had existed, in fact, beforehand, as it had been created in 1998 when a new immigration law was approved (Baganha, 2005). However, its application on a wide-scale basis only began in 2007.

The most important instrument of the new law was Article 88 (Acosta Arcarazo, 2013). Instead of the requirement that a valid visa be presented to obtain legal residence, the law accepted that a residence permit could be granted, for work purposes, under specifc circumstances. These included having a stable work relationship proved by a contract, a trade union, or an offcial entity (including immigrants' associations with a sit at the Consultative Council for Migration); having entered and stayed legally in Portugal (although this requirement could be dismissed after a penalty); and having registered and paid contributions to social security (a usual situation among irregular immigrants). Article 88 targeted subordinate employees, the most common situation, yet independent workers were also entitled to such benefts, under Article 89. In all cases, the access to legal status was not immediate: it depended on a personal interview. According to some authors (Acosta Arcarazo, 2013), this requirement was done in order to avoid a massive "pull effect" over further immigrants still abroad.

Besides immigrants engaged in work relations, the new law also created provisions for the ongoing regularisation of other individuals. These included victims of traffcking, thus respecting the EU directive on the theme. It also included children in specifc circumstances, such as minors born in Portugal from holders of residence permits, or who attended a pre-school, primary, secondary or professional education; and immigrants in specifc circumstances, such as adults with foreign parents born in Portugal yet who remained in the country from under 10 years of age and onwards. Also included were individuals with long term medical needs; and foreign citizens requiring exceptional responses, including reasons of national interests, humanitarian reasons; and other public interest motives.

Although former laws, mainly since 1998, had mechanisms that allowed the granting of legal residence to irregular immigrants in exceptional circumstances, it is widely recognised that the 2007 law was very progressive in this domain, capturing a wide array of situations non-existent beforehand. The number of immigrants that have beneftted from these provisions is not known. However, all sources consider the numbers to be quite high. For example, it is possible that by 2010, nearly 50,000 immigrants had already been given legal status under the new law (Acosta Arcarazo, 2013).

More than a decade after its approval, the 2007 law is still in place today, despite the changes introduced from 2012 to 2019. Notably, the law still contains the provisions concerning regularisation that were present from the beginning, even if some changes were introduced in Articles 88 and 89. These changes had the objective of extending or facilitating the process of regularisation, although they came with a different rationale – examined in the next section. For instance, the change from wage earner to independent worker was allowed; the creation of innovative entrepreneurial initiatives was rewarded; overseas students mobility stimulated; and the need for proof of legal entry in the country was discarded. The progressive approach enacted in 2007 was kept even during the fnancial turmoil of 2011–2014, when a centre-right government led the country, and was further enlarged after 2015, when a Socialist Party government, supported by an extended Parliamentary left-wing coalition, led the country.

#### *6.2.2 The Changing Alignment of Interests: The Erosion of the Political Consensus Around Immigration?*

The reasons for the policy choices relating to irregular infows were diverse. It has been argued that the two frst general amnesties, in 1992–1993 and 1996, resulted from a broad coalition of interests, which included pro-immigrant and pro-human rights associations, trade unions and several political parties spanning from left to right (Peixoto et al., 2009). The left-wing parties were the most proactive on this issue and advocated the rights' dimension, yet the right-wing parties were sensible to the irregular immigrants' profle. The majority of them were African immigrants coming from the ex-colonies, after the decolonisation process. The responsibilities inherited from a long colonial past, the conscience of the problems felt by the new independent countries, as well as the cultural continuities with the new immigrants (mostly Portuguese speakers), are explanations for the new policy options.

The regularisation enacted in 2001 had a clear economic rationale. The economic expansion of the time was partially linked to the EU membership and consequent European funds. This explained the accrued labour needs in the low-skilled economic sector, from construction to the service industry, and in the highly-skilled segment, such as professional services. Many of these sectors, particularly the former, were active employers of immigrants, either those already in the country or others recently arrived, such as the Brazilians or the Eastern Europeans. The recognition of such labour needs, the pressure from employers and, of course, the admission of a rights-based policy, were the main factors behind the new "stay permits" policy. These permits were a temporary solution to a labour problem – although it turned quickly into a permanent one. The new 2001 law was also innovative when it created the principle of "labour quotas", based on economic needs, to drive new

immigration infows; the creation of such quotas is said to be the result of the negotiation between left and right-wing parties to approve the law.

The 2003 regularisation is directly tied to the bilateral political relationship with Brazil. Former diplomatic problems between the two countries were related to migration (such as the diffculty for Brazilian skilled immigrants to exercise their profession in Portugal, for example dentists, and the increased number of Brazilians scrutinized at Portuguese airports – Feldman-Bianco, 2001). The economic dimension of Brazil and its political importance also certainly played a role in the agreement, as it was focused on Brazilians already working in Portugal (and Portuguese working in Brazil), who were fast becoming the main source of foreign labour in the country.2 When civil society actors raised the attention of similar needs among foreign groups other than Brazilian and pushed back against the exceptionality of the 2003 regularisation, the idea became generalized that no such privilege could be granted to just one country. The principles of the 2003 regularisation of Brazilians were thus extended to all immigrant workers in 2004.

As previously discussed, there was a change in paradigm from 2007 onwards. Policies after this time shifted from recurring to extraordinary regularisations, to a policy of regular and ongoing ones. The change was motivated by internal and external factors (Sampaio, 2017). Among the latter, the resistance of other EU countries to mass regularisations is of foremost importance (Finotelli & Arango, 2011). The change in the EU mood has also led other countries, such as Italy or Spain, to change their policy approach. The mechanism adopted in Portugal persisted over the years, with only some small changes which simplifed and enlarged its scope.

Ordinary regularisations under the 2007 Immigration Law were established by a centre-left government, led by the Socialist Party (PS). It fell within the context of economic uncertainty that preceded the harsh years of austerity and fnancial bailout in 2011–2014. During the fnancial turmoil, the political guidance of the country geared towards a right-wing leadership (PSD-Liberal Party and CDS-Christian Democrats), with an economic programme marked by late neoliberal principles negotiated with a Troika of international borrowers (IMF, ECB and EC). Even if the conditions for the application of the regularisation principle were changed slightly, the overall policy measures were not called into question. After 2014, the arrival in power of a left-wing government, led by the Socialist Party (PS), this time with Parliament support of the radical left (PCP-Communist Party and BE-Left Wing Block), not only maintained the regularisation mechanism, but even slightly enlarged the possibilities of mobilising it. Finally, no attempts to change it were made since the pandemic started in 2020.

It may be hypothesised that, notwithstanding the change of paradigm in 2007, many of the motives that explained regularisations remained the same. On the one hand, the economic rationale and particularly the acknowledgment of the need to fll vital labour shortages, was almost always present. According to some authors, there

<sup>2</sup>Brazilian immigration to Portugal increased substantially in the fnal years of the 1990s and early 2000s, then becoming the largest group of foreign nationals in the country.

was a hidden "expansionary approach" of the state behind these policies (Carvalho, 2018). Articles 88 and 89 of the law were clearly directed to an *ex-post* admission of the immigrants already living in Portugal and fulflling the country's labour market needs. The law, in this point, was a formal recognition of a *de facto* integration. However, on the other hand, the defence of immigrants' rights was subjacent to these initiatives. Regardless of a possible instrumental approach, the regularisation provided formal rights to immigrant workers from which they were initially excluded. In addition to workers, rights were awarded to immigrants not belonging to the labour force, including victims of traffcking, children born in Portugal who attended formal education, and immigrants present in the country from when they were less than 10 years old.

Further explanations might help explain Portugal's persistence of immigrants' regularisation policies. The need to control immigration, uncovering situations of irregularity and invisibility, has certainly been an underlying motive to politically address the issue (Malheiros, 2008). The need to eliminate unfair competition between immigrants and natives in the labour market was another factor that explained the persistence of such policies. Irregular migration can be accepted by unscrupulous employers, but has the collateral effect of damaging the working conditions of the native labour force – in addition to excluding immigrant workers from constitutional rights.

The positioning of different stakeholders may also help to explain the political options. Since the late 1980s, many coalitions have been enacted – although the degree of cohesion and proactivity of the actors have varied. Immigrant and human rights associations have been on the frontline of the battle for regularisations since the very beginning (Horta, 2010). Trade unions have also been a constant part of this movement, being active in most processes. Following ideological principles including the protection of workers' rights and the need to fght social dumping, the main trade unions in Portugal have been always vocal in this domain (Malheiros, 1998; Kolarova & Peixoto, 2009). Less visible were employers, though several observations indicate that they were behind the pressure to admit and regularise immigrants, clearly a vital resource to the functioning of low wage, labour-intensive sectors (Peixoto et al., 2009; Carvalho, 2018). The affordability and passivity of immigrant labour force was maybe an extra reason for their adherence. The Catholic Church was also among the more active actors in these policies, combining the supportive action in the civil society with the formal engagement of its representatives in offcial entities inspired by Catholic principles, including the High Commission for Migration (Esteves et al., 2003; Peixoto et al., 2009).

One of the main points highlighted in this domain was the broad political consensus around the theme that existed for long (Peixoto et al., 2009; Sabino et al., 2010), at least until the reform of 2007. The fact that immigration, and particularly irregular immigrants, have not been the object of high politicization has contributed to that consensus. However, it is possible to argue that after 2012 and especially after 2015, the prevailing dominant logic of consensus has been replaced by more explicit divergent discourses that are leading to systematic left-right divides in Parliament voting.

In fact, between 1992 and 2007, from the six bills on regularisation issues that were approved by the Portuguese Parliament, only two received votes of rejection and none of these expresses a left-right divide (see Table 6.2). After 2012, seven legislative changes were introduced in the 2007 Immigration Law, displaying a


**Table 6.2** Position of the main political parties regarding regularisations

Source: Sabino et al. (2010) and own elaboration

different panorama in terms of political alignments. During the period of the rightwing government (2011–2015), three bills of change were approved enhancing the links between migration, capital and skills, and also strengthening the securitarian principles. The right-left divide in the voting emerged, with the government parties voting "yes" and the radical left-wing ones voting "no". The centre-left Socialist Party joined the government majority and voted favourably the three diplomas.

This was followed by a second period of change, after 2017, which was marked by a shift in the policy guidelines and in political alignments, with a readjustment of the left-right divide that actually accentuated. The four bills of change approved in this period were geared towards the protection of foreign minors, the simplifcation and the enlargement of the case-by-case regularisation procedures. From these bills of change, only one did not receive favourable voting from all left-wing parties. The rationale behind these changes providing formal recognition and access to economic and social rights are in line with left-wing ideology, which tend to be less nationalist (*strictu sensu*) and securitarian, and also with the humanitarian principles of immigrant NGOs and Catholic Organisations.

Despite a wide coalition of interests around regularisation of immigrants following 2007, policy evidence from the last decade points indeed to some erosion. First, the fragmentation of the Parliament composition began in 2015, with the election of the frst MP from an animalist party (PAN – People, Animals, Nature). Four years after, the number of PAN MPs passed to four and three additional new parties elected one parliamentary each. Among these is CHEGA, a far-right political party, with an aggressive nationalist and anti-system discourse, supported by conservative and identity arguments that explicitly assume a xenophobic view of certain minority groups, particularly gypsies (Madeira et al., 2021). The fght against "illegal migration" is also a programmatic priority of CHEGA, involving more pro-active expulsion measures. Thus, the presence of a far-right nationalist and populist party in Parliament, for the frst time in Portuguese democracy, is the second major change in the recent political spectrum. Third, the neoliberal trends and the austerity policy of the 2011–2014 period led to widespread impoverishment and increasing inequality, justifying the voting shift towards centre-left and left-wing parties which became the majority in Parliament.

At the same time, some evidence of erosion of the most traditional parties of the system, who are present in the Parliament since the establishment of democracy in the mid-1970s, is contributing to a more polarized political system. The inability of these parties to respond to the growing economic diffculties of a large proportion of the population, who feels marginalized by the political system and threatened in identity terms (Ferrão, 2019; Fukuyama, 2018), helps explain political polarization and the adhesion to far-right parties with xenophobic and anti-immigrant discourses, not only in Portugal but also in several other European countries. The electoral attraction of these parties, namely CHEGA, is threatening traditional right and centre-right parties and pushing them to more conservative and authoritarian discourses, which move away from former agreements and risk jeopardising existing consensus around issues such as immigration.

Despite the changes, the main point to conclude is the persistence of the regularisation principle until today, based in an ongoing case-by case policy procedure from 2007. As discussed in this section, the imperatives behind such policy were diverse – and sometimes contradictory. They included an economic rationale based on the country's labour needs, a rights-based and humanitarian perspectives and broad geopolitical interests. However, as some observers have noted, the mechanism of ordinary regularisation has been far from automatic or transparent. In fact, the authorities have always maintained the ability to decide in favour or against the immigrants' requests, under motives that are not entirely clear – a problem that has been the object of scarce research and has often brought criticism. A recent work carried out by Costa (2020), applying the theory of "street level bureaucrats" to the operational procedures of the law, prove how the regularisation process is hermetic. Although discretionary applications of the law are common in every bureaucracy, the absence of clear guidelines causes frequent delays and leads to unclear procedures.

#### **6.3 The Portuguese Case in Perspective**

As seen above, in 2007 the Portuguese policies for immigrants' regularisation experienced a major shift, going from a rationale of extensive extraordinary processes to a logic of ordinary case-by-case ones. Despite this change, the maintenance of the legal possibility of ex-post regularisation seems to point to the recognition of the inability to adjust formal immigration channels to the migratory pressure and labour needs. Actually, the widening of the regularisation routes and the simplifcation of procedures adopted in 2017 apparently confrms the self-acknowledged inability for Portuguese authorities to regulate *ex-ante* immigration fows. Indirectly, this would also mean that Portugal should be among the EU countries with higher levels of irregular migrants.

In order to close the debate that has been developing along this chapter, we would like to challenge the two ideas expressed in the previous paragraph (the lack of capacity to regulate immigration fows and the high levels of irregularity in the European context) and to frame them in the wider context of the European responses to irregular migration. In other words, we want to address the merits and weaknesses of the Portuguese immigration regime, which may be considered a part of the wider Southern European model (Peixoto et al., 2012) and is supposedly marked by a lax migration regime and ambiguity towards irregular migrants (Finotelli, 2009). These features apparently correspond to some distinctive elements that characterize the South-North divide in relation to European migration regimes.

Arguments in favour of a common Southern European migration model have been advanced in the 1990s (see, for instance, King et al., 1997) and reiterated again more recently (Peixoto et al., 2012). Though the term employed by these researchers was "model",3 we may consider it equivalent to the notion of a migration regime, mentioned at the beginning of this text and used by Arango (2012) to establish the differences between the three basic regimes in Europe: Southern European, North-Western European, and Eastern and Central European. The differences between these clusters of countries emerge from a range of dimensions (from specifc admission policies to integration schemes, passing by regularisations and labour relations) included under the notion of immigration policies, that are framed in the general demographic contexts and socioeconomic features that characterize each of them.

Having taken into consideration this approach, Portuguese immigration policies tend to be interpreted within the framework of the Southern European migration regime (Arango, 2012), eventually constituting a particular subsystem, together with Spain, classifed as the Iberian variant (Malheiros, 2012). A systematic reading of the Portuguese case requires a combined analysis of the fuid management of fows (including regularisation) and integration principles, something that has also been underlined by Finotelli (2009) while studying the Italian and German migration regimes. The political consensus established around migration policies that dominated until recently implicitly assumed that fast labour market adjustments and precariousness, combined with high levels of informality,4 demanded a possibility of *ex-post* adjustment to labour needs (Malheiros, 2012; Carvalho, 2018). Because recruitment channels were everything but effcient (for instance, the labour quota system that lasted between the 2001 and 2007 laws never functioned), the solution was to rely on legal entries, irregular overstaying and "extraordinary" regularisation, as evidenced in the previous section. The consequence of this was both an exposure of foreign workers to labour exploitation and the risk of increasing unfair competition with domestic workers, leading to potential processes of social dumping.

This picture clarifes the bases for the political consensus around immigration policies, including the rationale of *ex-post* regularisation: on the one hand, the sectors more concerned with employers' interests and economic competition assumed the advantages of this format of workers' recruitment as a form of cost reduction and fexible adjustment; on the other hand, the sectors that privileged social protection and humanitarian approaches found in regularisation the way to equality of rights and citizenship, an issue that was also supported by trade unions in their fght against social dumping. This consensus was established in a period of economic expansion in the 1990s, and lasted while the Portuguese economy displayed ambiguous signs of contraction and expansion between 2004 and 2011. Once

<sup>3</sup>Though different authors use the terms "regime" and "model" to describe similar immigration frameworks, we have opted in this text for "immigration regime". The term "model" will be applied to the regularisation schemes, often designated as "regularisation models".

<sup>4</sup>Using the shadow economy estimates as a proxy to informality, the values found for Portugal place the country in the second highest position (after Italy) among Western European countries. The weight of shadow economy in Eastern European countries is estimated at higher levels (Schneider, 2009; Kelmanson et al., 2019).

socioeconomic decline has become continuous and extremely severe, the economic conditions for consensus around immigration decreased, followed by the aforementioned turnover in the political spectrum that also contributed to the emergence of a new frame.

In fact, the sinking of Portuguese economy, specially between 2011 and 2014, not only led to a substantial reduction in immigration infows and even in the stocks of legal immigrants (Oliveira & Gomes, 2016), but also generated a clear reduction in irregular migrants. Using the number of foreigners found to be illegally present in Portugal as a proxy for irregular migrants, its volume follows a path that seems to adjust itself in an imperfect way to the economic short-term cycles (it declined with the harsh economic and fnancial crisis of 2011–2014; it clearly increased in the initial years of the subsequent recovery and declined again with COVID-19 in 2020 – see Fig. 6.1), despite being also infuenced by other factors.

Having taken this into consideration, it becomes possible that irregular migration levels justifying policy measures such as extraordinary regularisations (the 1990s and mid-2000s typical procedure) or case-by-case ones (the system implemented after 2007), are more a function of economic cycles and a response to a certain type of economic demand (predominance of labour intensive industries) than the result of the implementation of policy mechanisms facilitating legalization. This is contrary to what is often stated. In other words, ex-post regularisation processes do not seem to lead to a signifcant increase in the number of undocumented immigrants or to produce what some authors designate as *call effect* (Orrenius & Zavodny, 2003; Fanjul & Gálvez-Iniesta, 2020). Empirical evidence does not seem to point to substantial increases in these numbers, though the transformation of international

**Fig. 6.1** Number of irregular migrants 2011–2020. (Sources: EUROSTAT (MIGR\_EIORD) (own elaboration))

migration in a global industry involving legal and illegal activities that generate millions of euros may put the focus of immigrant traffckers and labour recruiters in the areas that have the most fexible schemes of entry and regularisation (Baganha, 2005).

Another element that needs to be addressed is the idea that the Portuguese migration regime is part of the Southern European "common" migration regime and therefore different from the migration regime of North-Western European countries (NWEs). Though we share with Arango (2012) and Peixoto et al. (2012) the perspective that common contexts (stage of the migration cycle, economic restructuring processes, dominant labour market features, particularities of welfare state regimes, etc) lead to similar migration policies, several issues point to the limits of over-generalizing the supposed commonalities among countries of the same geopolitical space. In this line, authors such as Finotelli (2009) and Baldwin-Edwards (2012) have pointed the limits of a single and unitary Southern European migration regime.

An argument that challenges this idea concerns the relevance of interdependencies within the European Union in relation to the logic of migration fows and even its management. For instance, if politicians and civil servants of Northern and Central European EU countries explicitly criticized the extraordinary regularisations that Southern European countries made in the 1990s and early 2000s (Finotelli & Arango, 2011), this should be taken more on the side of political rhetoric than on side of effective policies. Actually, the facilitation of circulation from East to West in Europe5 that contributed to the increase in irregular migration, associated to overstaying processes in countries such as Italy or Portugal, corresponded to a process that interested NWEs countries, such as Germany, due to its intention of extending economic and geopolitical infuence to Central and Eastern European States. Only in the period of economic downturn associated with the world fnancial crisis of 2008, were statements made and EU directives approved on pushing the abandonment of extraordinary regularisations, forced return and increasing workplace inspections (Malheiros, 2012). Therefore, more than a clear divide between two very different regimes, we have a pipeline between them, a continuum more than a break.

Furthermore, the principle of general mass regularisations has also been applied in a more limited manner and with some specifcations by countries of Central and Northern Europe in the 1990s and 2000s. One example was the French operation of 1997–1998, who received 152,000 applications, of which 87,000 regularised. Another one corresponded to the German Arrangements for Right to Continued Stay (*Bleiberechtsregelungen*) of 2007, which allowed the suspension of the deportation order of approximately 50,000 irregular foreigners. Austria, Luxembourg and Netherlands also regularized undocumented workers (Kraler, 2009). These were mostly single processes in the period, and had a smaller impact than the

<sup>5</sup>For instance, visa requirements for the circulation of non-EU Eastern Europeans have been progressively relaxed along this period.

"equivalent" that took place in Southern Europe, requiring more years of irregular residence in the country. The intrinsic nature and fnal goals, however, were similar. Circumstances dictate the differences in frequency and specifcities, pointing to the idea that not only were Southern European responses less exceptional when viewed from this comparative perspective, but also the use of extraordinary regularisation mechanisms can still be exceptionally mobilized in contemporary time both in and out of Southern Europe, as the recent Irish case of 2022 demonstrates.6

The transition to "case-by-case" regularisations was also recommended in the 2008 European Pact on Immigration and Asylum, which explicitly rejects mass regularisation programmes. Though the majority of these processes privilege humanitarian reasons to justify regularisations (Kraler, 2009), some require labour market bonds to apply for a legal status and a few even consider employment as a preferential issue, such as France, Spain (*Arraigo Social*) or Portugal. The particular situation of Spain7 and Portugal eventually appears closer to the cases of France and Switzerland8 than to the cases of Greece or Italy, who do not allow case-by-case regularisations. This contributes to a fssure in the rationale of a single Southern European migration regime. The Portuguese law of case-by-case regularisation is however probably the most generous among the EU Member States, as it requires only 1 year of work and contributions to Social Security, demands a work contract or a promise of work contract, and allows irregular migrants to present their demands on an accessible portal.

The aforementioned evidences and its interpretation place the justifcations for irregular migration fows more on the contextual economic side than on the specifc policy side. Additionally, contradictions between political rhetoric and the explicit and implicit practices of managing migration fows are evident. Finally, the socalled South European migration regime seems less exceptional then is often

<sup>6</sup> Ireland opened a regularisation process between the 31st of January and the 31st of July 2022 for non-EU foreigners who have been undocumented for at least 4 years at the start of the scheme, or for at least 3 years for families with children under 18. (https://www.citizensinformation.ie/en/ moving\_country/moving\_to\_ireland/rights\_of\_residence\_in\_ireland/permission\_to\_remain\_for\_ undocumented\_noneea\_nationals\_in\_ireland.html). Just before the process started, estimates pointed to 17,000 non-EU foreigners living irregularly in the country and in conditions to respond to the scheme (https://www.independent.ie/irish-news/undocumented-migrants-in-irelandoffered-once-in-a-generation-amnesty-40775476.html).

<sup>7</sup>The *Arraigo Social* (Art. 124.2 of Royal Decree 557/2011, of 20th of April, which regulates Organic Law 4/2000, on Rights and Freedoms of Foreigners in Spain and their Social Integration) establishes the possibility of regularisation of immigrants who prove (i) that they have stayed for a minimum period of 3 years in Spain, (ii) have actually joined the labour market, by presenting a job offer for at least 1 year, and (iii) have a family bond or present a "report of roots", proof of their social integration in the country, issued by the Autonomous Community of habitual residence (Finotelli & Arango, 2011; Costa, 2021).

<sup>8</sup>Under certain conditions, such as having lived for several years in Switzerland and displaying a "good integration'' (which incorporates the notion of generating means to survive), irregular foreigners have possibility of applying for regularisation at the Canton level, that articulates with the Federal level (Der Bundesrat, 2020).

presented, emerging a clear need to repair regime inconsistencies that are transversal and common to several EU countries.

#### **6.4 Final Remarks**

Irregular migration has become a worldwide component of contemporary migration. Global income gaps, new labour market dynamics, refugees' waves, securitarian concerns and nationalist pressures, all lead to a vast crisis of control in this area (Cornelius et al., 2004). The dimension of the problem depends more often on geographic position, lack of adequate entry channels and type of economic demand (including the size of informal economy) than effective border policing. The challenge is tackled in various ways in different geographies: whilst some countries prefer to ignore the problem, others enact deportation strategies, and some create mechanisms for regularisation – although most jump from option to option along the time. When opting for regularisation, a change from extraordinary processes, based on economic or humanitarian grounds, to case-by-case mechanisms, has been pushed by some international organisations including the EU and effectively implemented by some countries. Whatever the choice is, responses cannot be considered innocent. Regularisations may be a way of satisfying employers, enabling economic growth, or conceding rights. They are an option which context and motives deserve scrutiny.

This chapter was devoted to examining the Portuguese case. As in other Southern European countries, Portugal witnessed considerable immigration fows since the 1980s. The type of economic demand was primarily based in labour-intensive industries, and the strength of the shadow economy fuelled the infows. Such as its Southern European counterparts, the policy responses were tentative and not capable of solving the endemic nature of irregular migration. Since the early 1990s several approaches were taken, which started as classic mass regularisations, turned to targeted regularisations and fnished as a case-by-case mechanism inscribed in the law. One of the features of the Portuguese case was the alignment of interests behind such policies, joining employers, trade unions, NGOs, the Catholic Church, leftand right-wing parties. Only recently, after the economic downturn of 2011–2014, some divergence emerged. Currently, the potential for erosion is strong and the continuity of the political consensus around immigration has been called into question.

The observation of the Portuguese case leads us to equate differences and commonalities with other EU countries. As other Southern European nations (Greece, Italy or Spain), the recourse to *ex-post* regularisations has been vast. However, as in the case of Spain (and also France or Switzerland), the system has evolved to a caseby-case basis, not applicable in Italy and Greece. This challenges the image of a common and singular Southern European migration regime, at least with regard to responses to irregular migration. Moreover, immigration to Portugal – and the subsequent irregularity – cannot be dissociated from the whole of EU dynamics, thus refecting a global chain of events, more than a singular autonomous capacity to attract migrants. In other words, irregular infows have largely resulted from the European framing of the Portuguese economic fabric and also from external geopolitical strategies, such as the openness of Germany and its policies towards Central and Eastern European States.

In conclusion, although the use of regularisation mechanisms as a way to address irregular immigration might be considered a characteristic feature of the so-called Southern European migration regime, it is marked by variations between countries within the region and has also been enacted in other regions of Europe. Whenever the migration pressure is strong, and whenever the context induces such a response (in economic or humanitarian grounds), regularisations remain a way of tackling the problem. Today, on a global and European level, there are no adequate channels to legally frame all potential migrants, nor are there suffciently tough enough borders to resist such infows. If some arguments remain to differentiate migration regimes of Southern and other European contexts, irregular migration and corresponding regularisation policies are not among the most prominent.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 7 Knowledge Production Through Regularisation and** *Ex-Post* **Regulation Strategies: Italy and Germany Compared**

**Claudia Finotelli**

#### **7.1 Introduction**

The gap between the restrictionist goals of migration policies and their expansionist outcomes represents one of the major dilemmas of immigration control (Cornelius et al., 2004). The consequence of this gap is often the presence of politically unwanted migration on the host state's territory in the form of irregular migrants and potential asylum seekers. Their presence not only emphasises a sense of a loss of control in nation-states, but also these states' diffculties in managing migration according to the expectations of their own populations. Since the 1990s, the Northern European recipe to reduce unwanted migration has mainly consisted of implementing measures to curb asylum fows by reducing welfare benefts for asylum seekers and strengthening external European borders. In contrast, new immigration countries in Southern Europe often had to solve the challenge of irregular migrants *a posteriori* through regularisation measures that granted residence permits to unauthorised residents on the state's territory (Ambrosini, 2018; Peixoto et al., 2012). It almost goes without saying that Northern European Union Member States viewed the Southern Europeans' *ex-post* regulation strategies with great concern. Regularisations were seen frst of all as the direct consequence of the Southern European soft underbelly, i.e. the "porous" southern borders of the European Union. Second, the Southern states were sharply criticised by Northern EU Member States since the frequency and degree of these regularisations was taken as further proof of Southern Europeans' "public ambiguity" towards irregular migration (Brochmann, 1993; Baldwin-Edwards, 1999).

Public criticism about regularisation processes was linked to scepticism about their long-term effectiveness. For instance, there was general agreement on the fact that regularisations had a sort of "pull effect", creating expectations of more or less

C. Finotelli (\*)

Complutense University, Madrid, Spain e-mail: cfnotel@cps.ucm.es

<sup>©</sup> The Author(s) 2023

C. Finotelli, I. Ponzo (eds.), *Migration Control Logics and Strategies in Europe*, IMISCOE Research Series, https://doi.org/10.1007/978-3-031-26002-5\_7

imminent regularisation and attracting an increasing number of irregular migrants, rather than reducing their numbers (OECD, 2000). In addition, researchers assumed that not all irregular migrants on the territory could fulfl the regularisation requirements, and that regularised migrants would usually fall back into irregularity after the expiration of their residence permit, because they would be unable to renew it in time (Reyneri, 1999). Finally, researchers also generally assumed that a considerable number of regularised immigrants would continue to be employed in the shadow economy, despite having legally obtained a residence permit through a regularisation process (Zincone, 2004). As argued by Papademetriou et al. (2004, p. 31), "evidence is meager and provides only spotty support for the benefcial labour market effects of regularisations."

More than other migration regulation tools, regularisations were used to highlight the "effcacy gap" (Czaika & de Haas, 2013) existing between Northern and Southern European control regimes, and these policies became the object of several confrontations between Northern and Southern European Member States. At the European level, the EU Commission repeatedly expressed its scepticism about such measures, questioning their capacity to reduce the shadow economy and arguing that they represented a "form of encouragement to illegal migration" (EU Commission, 2004, p. 10). In addition, some European countries clearly attempted to counter regularisations in other Member States. For instance, after Italy's 2002 regularisation, representatives of various Member States attempted to exclude immigrants regularised under such circumstances from the categories laid down in the European Directive on Long-term Residents from third countries (Finotelli, 2007). In 2005, both the German and Dutch governments heavily criticised a decision by the Spanish government to give a mass regularisation to irregular immigrants. The same Spanish government was also blamed for not having informed the other EU Member States about the process in a reasonable timeframe (Europapress, 2005). In particular, German and Dutch criticism was fuelled by a widespread fear that large numbers of regularised immigrants in Spain would subsequently enter other EU Member States, attracted by their generous welfare systems (Finotelli & Arango, 2011). As a consequence, Member States signed the European Pact on Immigration and Asylum in September 2008, where they agreed "to use only caseby-case regularisations rather than generalised regularisations carried out nationally for humanitarian or economic reasons" (European Council, 2008, p. 7). Since then, Italy has been the only Southern European country to maintain mass regularisations; Spain and Portugal turned to individual regularisations, which have garnered less media attention (see Malheiros & Peixoto, Chap. 6 in this book).1

The use of regularisations seems to have strengthened the stereotype of "weak" Southern European migration regimes, ostensibly characterised by chaotic and unplanned regulation mechanisms (Finotelli & Arango, 2011). Yet, in this chapter I argue that the function of regularisations should be assessed beyond a dichotomous

<sup>1</sup>Please note that Italy was able to keep its policy of periodic mass regularisation because EU Pacts on Migration are not treaties, and have no binding effect.

distinction between "weak" and "strong" migration control regimes. Instead, regularisations and *ex-post* regulation strategies should be seen as instruments to overcome weak internal controls and state ignorance on irregular migration, where ignorance is defned as the difference between what public authorities think they should know on unauthorised residence and what they actually know about it (Boswell & Badenhoop, 2021). Irregular migrants in the form of overstayers, *sans papiers* or non-deportable but rejected asylum seekers are probably the most evident examples of the expansionist outcomes of restrictive immigration policies. Such migrants represent a challenge for many national welfare states, since their presence cannot be registered by offcial statistical systems, while police identifcation and the subsequent deportation of the irregular migrants identifed is often diffcult to implement (Colombo, 2013; Sainz de la Maza Quintanal, 2015; Landini & Sciortino, Chap. 2 in this book). Recent Eurostat data is very illustrative in this respect: in 2019, only 29% of the immigrants ordered to leave a given state's territory were eventually returned to their country of origin (Eurostat, 2019). Against this backdrop, the goal of this chapter is to show that regularisations cannot be seen as a phenomenon unique to Southern Europe. Instead, they refect a rather widespread strategy to "repair" the inconsistencies of internal control regimes by producing state knowledge on the presence of unauthorised immigrants. To demonstrate this, I examine the Italian and German immigration regimes as comparative examples, since they represent two opposed migration patterns in Europe, with opposed approaches to unauthorised residence. Italy is a rather recent immigration country, which nevertheless accounts for at least half of all regularised immigrants in Europe and where expectations about the state capacity of law enforcement and knowledge production are low. In contrast, in the Federal German Republic, which has so far received the largest number of refugees in Europe, regularisations are rare and expectations about the state's capacity for law enforcement and knowledge production are high (Finotelli, 2009). Driven by the concept of regime proposed in the introduction to this volume, analysis focuses on organisational strategies and processes rather than simply on policy goals and outcomes in order to understand the functioning of regularisation processes beyond national typologies. The aim of this comparison is not only to discuss how these very different migration regimes produce knowledge on unauthorised residence on their territory, but also to highlight the importance of these measures for the stabilisation of a precarious immigrant population.

The analysis, based on research conducted in 2004 and 2021, draws on offcial statistics and government documents. The frst part of the chapter describes the use of regularisations in the Italian case; the second deals with *ex-post* regulation measures, such as old-case regulations and "exceptional leave to remain" (*Duldung*) in Germany; the last part of the chapter compares the two countries to show the functional equivalence of the different types of *ex-post* regulation measures that are used to bring irregular migrants back under the radar of the state.

#### **7.2 When the Exception Becomes the Rule: The Role of Regularisations in the Italian Migration Regime**

Italy's transformation to an immigration country was anything but easy. From the very beginning, the country had to adopt an extremely restrictive immigration policy approach, since "this was the price that this 'new' immigration country and future Schengen member had to formally pay for its European membership" (Finotelli & Arango, 2011, p. 499). As a consequence, irregular migration became a structural feature of the Italian migration regime, resulting from the combination of dysfunctional entry policies with an urgent labour demand in various economic sectors. According to Italian law, foreign workers had to be recruited before they entered Italy and were only admitted if previous labour-market checks had demonstrated the unavailability of Italian or EU citizens for the offered position. Such labour market checks, however, turned recruitment procedures into burdensome and ineffective processes. Moreover, the number of available entry slots annually included in the so-called "decrees on fows", Italy's annual entry quotas, often underestimated the real labour-market demand. Ultimately, the inadequate immigration legislation, together with growing overall migration numbers and an expanding informal economy, spurred an increase in irregular migrants (Finotelli & Sciortino, 2009). As a result, most employers started hiring foreign workers who had already begun living in Italy as irregular immigrants. In the 1990s, Italian governments used "decrees on fows" to regularise a given number of irregular immigrants every year, and "thus to "correct" the consequences of a dysfunctional migration regime (and the structural demand for low-skilled labour)" (Finotelli & Arango, 2011, p. 499).

However, resorting to "fake" quotas was not enough to curb the number of irregular migrants working in the informal economy. For this reason, Italian governments carried out fve mass regularisations between 1986 and 2002, issuing a total of 1,417,000 residence permits. The large number of immigrant workers regularised in 2002, half of them in the domestic service sector, brought to the forefront the crucial need to revise recruitment procedures and strengthen both the external and internal control apparatus. Indeed, in the last two decades, Italy has been able to strengthen its external controls, especially through the signing of *ad hoc* admission agreements with various countries of origin (Finotelli, 2018). In contrast, internal controls have remained weak due to the wide gap between the number of deportations ordered and the number carried out, as well as insuffcient labour market controls (Colombo, 2013). In addition, recruitment in sectors with high labour demand, such as domestic service, is still very diffcult though formal recruitment procedures. Not surprisingly, the diffculties in designing and implementing labour migration policies that can effciently match supply with demand have greatly contributed towards the continued use of regularisations and "fake" quotas in Italy (Sciortino, 2009). Similar to what occurred in the 1990s, the objective of the decrees was not to allow "new" workers to enter, but to regularise the irregular immigrants who had already been living in Italy as visa overstayers. In 2006, the Berlusconi government approved a so-called "maxi-decree" on annual entry quotas, allowing the entry of 470,000 foreign workers. The maxi-decree was followed by two new "decrees on fows" in 2007 and 2008, while a new regularisation process – albeit limited to domestic and care workers – was carried out in 2009. Another regularisation followed in 2012, which also targeted domestic workers, but was not limited to them. However, both the 2009 and 2012 regularisations were not only accompanied by a remarkable decrease in the number of applications with respect to regularisations in the past, but also in the number of residence permits ultimately issued. According to Colombo (2009), the decrease was not only related to the poor performance of the Italian bureaucracy, but also to changes in the type and dynamics of migration systems, which have shown an increasing presence of intra-EU mobility from Eastern Europe (see Devitt, Chap. 10 in this book). The most recent regularisation was carried out in 2020 to address the consequences of the Covid pandemic. Also in this case, the results of the process were signifcantly below expectations; it has experienced several bureaucratic delays with the consequence that only 60,000 residence permits, equating to 26% of the 230,000 applications submitted, had been issued one year after the beginning of the regularisation process in 2020 (Ero Straniero, 2021).

All in all, Italian governments have carried out eight regularisations since 1986. Each one was accompanied by the promise that it would be the last of its kind. Their objective was to regularise as many immigrants as possible in order to bring irregular migrants out of the shadows of residence invisibility and labour exploitation. Other than the 1990 regularisation, they all depended on the immigrant's employment status, which meant that foreigners could regularise their stay if they were able to regularise their employment situation.

In general, regularisations in Italy were very burdensome processes. Their implementation was often complemented by administrative memos, which often added details or changed requirements in the course of the ongoing process; these included the ability to accept expulsion orders for illegal residence as proof of the applicant's residence in Italy after a certain date (Finotelli, 2007; Asgi, 2021).

Undoubtedly, regularisations have profoundly shaped Italian migration policy to date. They have been equally carried out by both centre-right and centre-left governments, and have often benefted from the lobbying action of trade unions, business associations and – especially – Catholic organisations, which have also been described as the "strong lobby for weaker strata" (Zincone, 2011, p. 259) and whose lobbying action benefted from the presence of Christian parties, such as the UCD (Unione Cristiano-Democratici) in government coalitions. Clearly, the widespread support of regularisations reduced their public perception as controversial measures. This has also been refected in the media discourse, where coverage of the so-called "*sanatorie*" and "*condoni*" in the immigration feld tended to occupy a marginal position compared to other types of amnesties of this kind (Colombo & Sciortino, 2004). Analogous regularisations have often been carried out in several other felds, most often the construction sector or to reduce tax evasion, but also in less important sectors, such as that of the illegal possession of archaeological fndings or of exotic animals (Colombo & Sciortino, 2004). The implementation of immigration regularisations was generally accepted as necessary to deal with irregular migration and the weak penetration capacity of the Italian state on its own territory. Such acceptance was embedded in the economic legitimisation of immigrants – corresponding to the idea that if migrants work, then they deserve a right to residence – and the political culture and organisational structures of the Italian state.

Even though Italian voters accepted these regularisations as necessary policy measures, they were deemed by experts to offer only very precarious residence statuses to migrants. This was not only due to the short duration of the residence permits (with a maximum of 2 years) and precarious employment conditions, but also due to the discretionary power of the public offcials who renewed the permits. For this reason, it was generally assumed that the protracted renewal procedures, the precariousness of labour in sectors where immigrants were employed, and the frequent use of short-term labour contracts might have facilitated regularised immigrants' "return to irregularity" (Reyneri, 1999). However, research conducted to date seems to contradict this assumption. Data from 2002 showed that more than half of the immigrants regularised in Italy between 1986 and 1998 were still in possession of their original residence permit at the beginning of 2000 (Carfagna, 2002). Another study showed that the number of residence permits issued between 1992 and 2000 increased from 649,000 to 1,341,000. More than 60% of the permits issued during that time came on the heels of the 1995 and 1998/1999 regularisations (Istat, 2005). In addition, the 2002 regularisation caused a spike in residence permits registered in 2004 (724,000 more than in 2003). At the same time, the number of individuals who applied to more than one regularisation programme remained negligible (Carfagna, 2002). Considering the restrictive and dysfunctional entry and residence rules in Italy since the 1990s, we can assume that regularisations represented a major stabilisation channel, since the majority of the initial residence permits in Italy were issued after a regularisation process, paving the way for family reunion and long-term residence (Einaudi, 2007; Ponzo et al., 2015). Renewal regulations, together with the labour market structure, have facilitated residence consolidation in spite of an initial period of legal instability for regularised migrants (Vianello et al., 2019). Moreover, the periodic execution of regularisations has compensated for the lack of active (and effective) labour-migration policies. This is especially true for domestic service and small and medium-sized enterprises, whose representatives have often advocated more generous quota regulations and more fexible recruitment procedures. Most of all, however, the periodic regularisations have prevented the uncontrolled spread of irregular employment and have allowed "invisible" irregular migrant workers to be brought back "into the light" of state control. The internal control function of regularisations in a country such as Italy, where internal controls – including labour market controls – are very weak, is particularly relevant when assessing the importance of *ex-post* control mechanisms in the German migration regime, as will be seen in the next section of this chapter.

#### **7.3 The Path to "Real" Residence: Forms of Regularisation in the German Migration Regime**

#### *7.3.1 From the Suspension of Deportation to Residence Regularisation*

Germany is considered the strongest bastion against the "temptation of regularisations" in Europe. In contrast to Italy, regularisations are rather alien to the German political culture, and if the possibility of a regularisation is raised, it usually meets with little acceptance in public and political debate. Such attitudes are certainly linked to the public's strong expectations about the law enforcement capacity and knowledge production of the German state. Indeed, Germany (in contrast to Italy) has a rather effcient state monitoring system in which public administration offces are closely interconnected and are subject to a general reporting obligation (Boswell & Badenhoop, 2021). It is under such assumption that German policymakers, for instance, almost immediately turned down a proposal by several NGOs in 2004 to issue "papers for all" irregular migrants; after that, the Finance Minister at the time, Hans Eichel, proposed to declare the employment of irregular domestic workers as a crime and to control private households (Balser, 2004). Likewise, when immigration experts from the Federation of German Industries (*Bundesverband der Deutschen Industrie*) argued in favour of economic utility as a criterion to address the problem of irregular migration, it immediately met with political resistance (Alt, 2003). For all practical purposes, the use of regularisations in Germany has only been accepted under humanitarian premises. At the beginning of the 1990s, Germany was the largest asylum country in Europe. An amendment of the Basic Law (the equivalent of Germany's constitution) on 1 July 1993 not only triggered an increase in irregular entries and visa overstays, but also created the subsequent need to address the growing number of rejected asylum seekers remaining on German territory after their asylum claim had been rejected. According to data provided by the German Parliament, about 650,000 rejected asylum seekers were still staying in Germany in 1997. Some of them were still waiting for a result of their appeal to a negative decision on their application; the rest had received exceptional leave to remain (*Duldung*), i.e. a temporary suspension of deportation. Exceptional leave to remain may be granted for legal or factual reasons, or because of obstacles to deportation due to considerable specifc danger to life, limb and freedom. Nevertheless, the granting of exceptional leave to remain was predominantly based on legal or factual grounds. "Legal" obstacles to deportation included, for example, the threat of torture, the death penalty, and inhuman and degrading treatment, or in favour of the protection of marriage. The reasons for "factual" obstacles to deportation have ranged from an unverifed identity and a lack of deportation possibilities (such as the lack of an airport) to the inability of the person concerned to be transported, to the lack of a passport; this seems to have been the most frequent reason for exceptional leave to remain to be granted (Heinold, 2003).

**Fig. 7.1** Duldungen and Asylum Applications in Germany (1995–2004). (Source: Finotelli, 2007)

Since the restrictions of the German asylum system were put in place, the constant decline in the number of asylum seekers contrasted with quite a steady presence of foreigners with exceptional leave to remain (see Fig. 7.1). Most came from the former Yugoslavia; others came from Turkey, Afghanistan and Iraq (Finotelli, 2007). The number of persons with exceptional leave to remain at the beginning of the new century refected the problem of the whereabouts of rejected but not deportable asylum seekers, which is often considered in the literature as one of the preconditions for permanent, irregular stays. Foreigners' Offce (*Ausländeramt*) employees interviewed in various German cities at the beginning of the new century blamed the absence of the necessary documents on a lack of cooperation by foreign authorities (Finotelli, 2007). Many countries of origin had no interest in the return of their own citizens, either because they were politically undesirable or because they wanted to maintain remittances from abroad. In addition, the lack of readmission agreements also complicated or prevented speedy readmission procedures (*ibid*.).

Exceptional leave to remain documents were therefore considered by state authorities to be a useful tool to overcome temporary non-deportability. The government issued them for very short intervals; depending on the cases, these documents could have a duration from a few days to many months, and they were not necessarily linked with a work permit. Due to deportation diffculties, many *Duldung* holders remained in extremely precarious situations for several years, trapped in so-called *Kettenduldungen* (serial suspensions of deportation). Due to the high level of uncertainty they bring, the *Duldung* system has repeatedly come under criticism. In particular, the impossibility of deportation combined with the *de facto* residence of *Duldung* holders in Germany for a period of many years was also perceived as a challenge for integration. As the Expert Council on Integration and Migration described in a 2004 report, "a permanently high number of merely tolerated

refugees is associated with the organisational decline of integration potential" (Sachverständigtenrat für Zuwanderung und Integration (SVR), 2004, 106). As mentioned above, exceptional leave to remain does not allow a stable residence status, and thus gives immigrants only limited opportunities for social participation. At the same time, people with exceptional leave to remain are dependent on benefts under the Asylum Seekers' Benefts Act (*Asylbewerberleistungsgesetz – AsylBLG*) and are largely paternalistically cared for by the state in the form of in-kind benefts, since they are not always allowed to work.

The presence of a large number of long-term *de facto* residents who were in judicial limbo forced policymakers to act. The words of an employee from a Foreigners' Offce exemplify this, who described *Duldung* holders as migrants whom "nobody wants, but the state can't get rid of" and added: "At some point they are there, and I have to think about what to do with them" (Head of a Foreigners' Offce, April 2004). For this reason, the German government decided to introduce residence regulations for rejected asylum seekers who could not be deported and who had been staying in the Federal Republic for years with exceptional leave to remain. Before the Immigration Act of 2005 (*Zuwanderungsgesetz*) came into force, these regulations were made by the immigration authorities of the respective local federal states (*Länder*) in agreement with the Federal Ministry of the Interior in accordance with Paragraph 32 of the Foreigners Act of 1990. According to this procedure, the Ministries of Interior of the *Länder* implemented seven "old-case" regulations between 1990 and 2000 (Kraler et al., 2009). These regulations addressed rejected asylum seekers who had found a job and belonged to specifc ethnic groups. In any case, the possession of exceptional leave to remain represented an essential precondition to apply for this regulation*.* Despite the media interest in these measures, the number of cases processed with a positive outcome remained low, while the number of *Duldung* holders in Germany remained high. According to information provided by the German government, old-case regulations carried out between 1996 and 2002 allowed the country to issue about 60,000 residence permits (*Bundestagsdrucksache* 14/9916 of 30 August 2002). Yet in 2003, according to data of the German Statistical Institute there were still 226,569 *Duldung* holders in Germany, half of whom were rejected asylum seekers, and half of whom were migrants who had entered the country irregularly and could not be sent back for legal or factual reasons (Finotelli, 2007).

To resolve this situation and to adapt the new immigration legislation to reality, the German government introduced the instrument of the "right to stay" (*Bleiberecht*) into the new Immigration Act of 2005; this included the Residence Act of 2004 (*Aufenthaltsgesetz – AufenthG*), which itself replaced the old Foreigners Act of 1990. The new Residence Act kept exceptional leave to remain as a regulation tool for those foreigners who could not be deported for legal or factual reasons (Paragraph 60a (2)). Nevertheless, the new law also introduced new possibilities for immigrants to normalise their residence status. According to Paragraph 25 (5) of the 2004 Residence Act, the local Foreigners' Offce authorities may grant a temporary residence permit to foreigners who are subject to an enforceable obligation to leave the country if their departure is impossible for legal or factual reasons, and if their deportation has already been suspended for 18 months. However, a prerequisite for obtaining the "right to stay" is that the foreigner concerned is not prevented from leaving Germany due to their own fault. In particular, the foreigner is deemed to be "at fault" if they have provided false information or deceived the German authorities with respect to their identity or nationality, or have failed to meet reasonable requirements to remove the obstacles to departure (Paragraph 25 (5) Sentence 2, Residence Act of 2004).

Soon after the approval of the new Residence Act, the state-level interior ministries promoted an additional far-reaching old-case regulation, which was however not laid down in the new law, but remained a Decision by the Conference of the Ministers of the Interior (also called the *IMK-Bleiberechtsbeschluss*). According to available data, 71,857 people applied for a residence permit, 19,779 of whom received a residence permit by September 2007, while an additional 29,834 were granted exceptional leave to remain to be able to look for a job and fulfl the income requirement (Kraler et al., 2009). Overall, since 2006, about 80,000 people have obtained a "right to stay" according to one of these regulations (Diakonie Deutschland, 2021). Indeed, one of the most criticised aspects of this *ad hoc* regularisation was that applicants had to prove they had suffcient income for themselves and their family members – despite the fact that *Duldung* holders in Germany are not always allowed to work. Consequently, the government had to progressively relax the application criteria to apply for this exceptional "right to stay" (Kraler et al., 2009). In 2011, the German government introduced the option to issue a residence permit to people between 15 and 20 years old who could demonstrate a high level of integration (Paragraph 25a *AufenthG*), while in 2015, the German Parliament approved an Act on the New Regulation of the Right to Stay and the End of Residence (*Gesetz zur Neubestimmung des Bleiberechts und der Aufenthaltsbeendigung*); this turned into Paragraph 25b of the *AufenthG* and established the rule that *Duldung* holders could obtain a residence permit if the condition of effective integration was fulflled. To assess the level of "effective" integration, Foreigners' Offces have to consider, for instance, the applicants' means of subsistence, school attendance in the case of minors, and language knowledge, as well as knowledge and respect of the social and legal norms characterising social life in Germany (Lower Saxony Ministry of Interior, 2019). This was new in the German migration regime, because it created a link between long-term integration and formal residence, even when some of the other minimum application criteria could not be fulflled (Diakonie Deutschland, 2021). Finally, the German legislation gave foreigners the possibility to obtain exceptional leave to remain for education purposes (Paragraphs 60 b and c *AufenthG*), giving the holder the possibility to subsequently have this transformed into a residence permit according to Paragraph 23a *AufenthG*.

#### *7.3.2 Lights and Shadows of Ex-Post Regulation Tools in Germany*

Exceptional leave to remain (*Duldung)* is an instrument to suspend deportation; it is not equivalent to a residence permit from a legal point of view. Moreover, *Duldungen* are not regularisations, because they do not affect the holder's obligation to leave Germany, although they represent a *quasi*-right to residence by suspending deportation (Hailbronner, 2000). In contrast, old-case regulations follow a different logic. German authorities have always denied that old-case regulations are like regularisations, and that they could be useful in solving the problem of irregular migration (Kraler et al., 2009). Nevertheless, from a technical point of view, they can certainly be considered close to regularisations since their goal is to issue residence permits to migrants who should not actually be on state territory. Against this backdrop, oldcase regulations have been described as a two-phase regularisation process, which frst provides for the suspension of deportation and then, as a second step, for a residence permit valid for 2 years (Hailbronner, 2000). In this vein, old-case regulations contribute to stabilising the situation of a precarious immigrant population in a similar way as in Italy.

The legal similarity between Italian regularisations and old-case regulations fts with the argument that regularisations cannot be solely considered a peculiarity of Southern Europe, although they have indeed been implemented predominantly in Southern EU Member States as a systematic measure to control irregular immigration (de Bruycker & Apap, 2000; Baldwin-Edwards & Kraler, 2009). Nevertheless, a comparison of *ex-post* regulations in Italy and Germany from a functional rather than a legal perspective can provide new insights into the functioning of migration regimes and the ways states handle the challenge of unwanted migrants.

Regularisations in Italy dealt with irregular immigrants whose presence was not formally known, and who were therefore referred to as *clandestini*. For Italy, regularisations allowed the government to recover control over a segment of the immigrant population that had escaped state monitoring. Second, taking into account that a large number of migrants living in Italy today obtained their frst residence permit after a regularisation, regularisations not only contributed to overcoming state ignorance, but also paved the way for the long-term legal integration of immigrants. In contrast, old-case regulations in Germany dealt with foreigners who had already been registered, and were therefore already known by the German state as people with exceptional leave to remain. The function of registration ascribed to Italian regularisations would therefore be more conceivable in the case of exceptional leave to remain, which formalises a foreigner's stay without granting him or her a regular residence permit. As with regularisations in Italy, issuing such documents can be understood as the state's response to the challenge posed by the unlawful residence of foreigners that cannot be terminated for the short term (Bommes, 2006). Put differently, *Duldungen*, like regularisations, have allowed to overcome state ignorance on the problem of irregular migration. This means that the German regime also shows a certain degree of public ambiguity in dealing with irregular migration insofar as *de facto* unlawful residence in Germany is tolerated, and in a formalised way through exceptional leave to remain (Bommes, 2006). Like Italian regularisations, *Duldungen* have allowed policymakers to repair the dysfunctionalities of the German migration regime to a certain extent, and for this reason they have been retained as a disguised regularisation tool ever since. If they cannot be considered regularisations *de jure*, they are certainly regularisations *de facto*. Yet, due to the precariousness of the *Duldung* status, they have contributed little to promoting legal residence stabilisation as in the case of Italian regularisations. Such stabilisation functions took place in the second step of old-case regulations, which made legal stabilisation depend on the pre-existence of socioeconomic integration as it occurs for Italian and other types of regularisations in Southern Europe (see also Malheiros & Peixoto, Chap. 6 in this book). Notably, the German requirement of effective integration as a precondition to take advantage of old-case regulations also includes cultural aspects, such as (oral) knowledge of the German language; this does not apply in the Italian case, where regularisations only depend on whether the migrant can demonstrate economic integration – having employment and paying into the welfare system.

The rather large number of refugees with exceptional leave to remain, in spite of the recent changes in legislation, confrms the continued use of *Duldungen* even today, as the evolution of fgures between 2013 and 2021 seems to demonstrate. Initially, the changes introduced during the frst decade of the new century, together with deceasing asylum fows due to changing political conditions, contributed to reducing the number of asylum seekers and refugees with exceptional leave to remain from 650,000 to 134,000 in 2011. However, the refugee crisis of 2015 triggered a new increase in the number of asylum seekers, which jumped again, to over 725,000 applicants. By the end of 2017, the number of asylum seekers and refugees with exceptional leave to remain had returned to 511,000 (*Bundestagsdrucksache* 19/3860 of 30 June 2018), while 166,000 of the 229,000 foreigners with a deportation order in 2017 had received a *Duldung* (*ibid*.). The reasons for exceptional leave to remain ranged from medical reasons to the requirement to care for relatives, to a lack of travel documents, or impossibility of deportation due to the unbearable situation in the country of origin. 43% of the *Duldungen* issued by the end of 2017 were granted for "other reasons", such as cases of people with relatives who could not be deported, people in the appeals process, or young people who were doing an apprenticeship (*Bundestagsdrucksache* 19/633 of 5 February 2018). According to recent data, 236,000 of the 281,000 unlawful residents in 2020 had been granted exceptional leave to remain (*Bundestagsdrucksache* 19/28234 of 6 April 2021).

Clearly, the evolution of *Duldungen* between 2013 and 2021 demonstrates that these instruments have come to play a crucial role in the German immigration regime in dealing with the persistence of obstacles to deportability without granting a formal residence permit (see Fig. 7.2).

The German authorities are well aware of this important function. As Boswell and Badenhoop (2021) have shown, German state offcials have perceived a kind of

**Fig. 7.2** Duldungen and Asylum Applications in Germany (2013–2021). (Source: Bundestagsdrucksachen. 19/1371; 19/633;19/28234)

"schizophrenia" between straightforward law enforcement against irregular residence in Germany and the ethical (and practical) problems related to the detention and return of irregular migrants. Boswell and Badenhoop noted that "the practical and ethical challenges involved in pursuing *sans papiers* and the political attention devoted to asylum led the federal authorities to focus their efforts on the *Geduldete* [migrants with exceptional leave to remain]" (*ibid*., p. 348). This runs contrary to the Federal Ministry of the Interior's description of *Duldungen* as an "instrument of fne-tuning" to manage the challenge of those asylum seekers who should not be on the state's territory, but who cannot be deported (Bundesministerium des Inneren, 2005, p. 4). What is less evident, but no less important, is that this instrument also allows the state to keep control of a social problem that would otherwise widen the gap between what the German state is expected to know and what it actually knows.

#### **7.4 Conclusion: The Signifcance of Regularisations and** *Ex-Post* **Regulations Across Regimes**

The comparison between Italy and Germany shows that both migration regimes have had to deal with the consequences of their "restrictive orthodoxy" to the extent that they were faced with politically unwanted migrants who remained. Both countries reacted to this with *ex-post* regulation tools. Both Germany and Italy have implemented specifc measures to produce state knowledge on the presence of unauthorised migrants on their territory. However, comparing the two countries also confrms that the way states overcome their ignorance about social problems are deeply embedded in nation-state contexts, and that such cross-national variations are related to public expectations regarding the oversight capacity of nation-states (Boswell & Badenhoop, 2021). In Italy, public acceptance of regularisations is deeply embedded in the Italian political culture, where expectations about the oversight capacity of the state are traditionally low. Second, regularisations were able to receive widespread economic legitimisation since they were traditionally presented as a way to trigger the emergence of irregular workers from the shadows of the informal economy. Third, they also implied low political costs for the political elites due to the structure of the Italian welfare regime, where immigrants can only become regularised after they pay a certain amount into social insurance funds (at least since 1998), and only small amounts of unemployment beneft are paid for a limited period of time. Against this backdrop, the sudden unemployment of regularised immigrants would not have had any signifcant consequences for decision makers, since the burden on the welfare system from regularised immigrants was limited in time.

In Germany, by contrast, regularisation would represent a public refutation of the oversight capacity of the state, since expectations regarding the state's law enforcement abilities are higher than in Italy. Moreover, mass regularisations for economic reasons would also have contradicted Germany's public image as a reluctant immigration country or, more recently, the image of a country open to high and mediumskilled migration, but less prone to accepting low-skilled migrants. Finally, and in contrast to Italy, the implementation of a mass regularisation would have carried higher political costs because of the public salience of the debate on unemployment benefts and social aid for the long-term unemployed. At the same time, recent research has shown that German public authorities seem to be less interested in "producing knowledge" on *sans papiers* on account of ethical and practical concerns (Boswell & Badenhoop, 2021). Due to the unpopularity of mass regularisation measures, together with the challenge represented by the implementation of straightforward internal controls, *ex-post* regulations in Germany are mainly anchored in the structures of the asylum system, and do not have an economic, but rather a "humanitarian" background. As the above shows, *de facto* regularisations such as issuing *Duldungen* are bound to the structures of the *non-refoulement* ban, and they show how modern states cannot escape from the embedded liberalism of their norms. The existence of these barriers built into modern welfare states contribute to the *de facto* consolidation of residence, forcing states over time to recognise that *non-refoulement* is an open-ended principle. Against this backdrop, the granting of exceptional leave to remain or a residence authorisation has so far mainly taken place within the framework of an individual case assessment, rather than as a collective solution, as in the case of Italian regularisations.

All differences notwithstanding, regularisations and *ex-post* regulations in both Italy and Germany have allowed the state to recover control over a social problem such as irregular migration, which, by defnition, escapes state monitoring. In this respect, the comparison between the two countries and their *ex-post* regulation tools reveals similar regime logics behind the different handling of state ignorance on unauthorised residence and legal inclusion in the long term. This functional equivalence further blurs the North-South divide in immigration by showing that "public ambiguity" about unauthorised residence is not only a Southern European feature. It further demonstrates that both the Italian and the German migration regimes are ultimately the result of "continuous repair work through practices" (Sciortino, 2004, p. 33) that allows states to produce knowledge when traditional internal controls fail. Finally, *ex-post* regulations have progressively contributed to the stabilisation of the immigrant population in both countries over the last three decades. In Italy, most regularised migrants are now long-term residents. In Germany, *Duldungen* were considered a stopgap solution before being able to obtain formal residence through old-case regulations and right-to-stay policies. However, *Duldungen* were granted to a smaller number of migrants compared to Italy and have been linked to more precarious conditions than regularised migrants in Italy. In this respect, *Duldungen* can be certainly understood as an instrument to make a particular segment of immigrant population more visible, but not more stable – stability instead corresponds to old-case regulations.

#### **References**


Bundesministerium des Innern. (2005). Zuwanderung – Das Neue Gesetz.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 8 Differently Similar: The Quest for Migration Control in the Netherlands and Spain**

**Gabriel Echeverría**

#### **8.1 Introduction**

Analysing the approach and effcacy of EU countries in controlling international migrations, the Netherlands and Spain have often been portrayed as opposite examples. The former as the top of the class when coming to strict control enforcement and effective migration deterrence, the latter, in good company with countries like Italy, Greece or Portugal, as an example of weak control measures and inconsistent results. Yet, the dichotomist hypothesis of a North/South divide that stays behind this type of interpretation, frequently associated to a not too veiled moral judgment on the good and the bad, is diffcult to maintain even after a frst scratch to the surface (Finotelli, 2009; Finotelli & Kolb, 2017; Echeverría, 2020; Ponzo, 2021).

The comparative analysis of migration control realities – something very different from the examination of migration control political agendas, governments' offcial statements, or even law provisions in the paper – reveal a much more complex and nuanced scenario. Migration control is neither a static nor a top-down, linearly determined, process. In this feld, countries have displayed and keep displaying a very dynamic, at times erratic, conduct that generates both convergence and divergence with the others. Moreover, in each context, the effcacy of control efforts, the translation of the political and legal desires into practice, in other words, the overall outcomes of the adopted strategies appear to be infuenced by a variety of interactions – not always easy to disentangle and assess – that are related to the confguration and functioning of societies as a whole. Against this backdrop, rather than as good or bad enforcers, effcient or ineffcient implementers, strict or bland masters of migration phenomena identifable with any geographical or cultural area, the European states emerge more as pragmatic actors that, considering their history, geography, socio-economic structures, administrative capacities and ever evolving

© The Author(s) 2023

G. Echeverría (\*)

Universidad Complutense de Madrid, Madrid, Spain e-mail: gechever@ucm.es

C. Finotelli, I. Ponzo (eds.), *Migration Control Logics and Strategies in Europe*, IMISCOE Research Series, https://doi.org/10.1007/978-3-031-26002-5\_8

internal political struggles, look after the best, temporary, adaptive and inevitably sub-optimal solution (Sciortino, 2000; Bommes & Sciortino, 2012; Echeverría, 2020).

Yet, assuming this perspective poses several conceptual and methodological challenges to the research in this area (Wimmer & Glick-Schiller, 2003; Bommes, 2012; Luhmann, 2012; Boucher & Gest, 2015). On the one hand, the idea that migration control is not a process fully determined by the states, but rather the outcome of complex social interactions determined by the interplay of social subsystems (economy, politics, welfare, communications, religion, etc.) requires the revision of dominant state-centric paradigms and the adoption of more sophisticated tools that assume complexity as the starting point and put the concept of society at its centre. The frst implication of this shift is the dismissal of easy, clear-cut categorizations and the adoption of a radically differentialist view. Whereas is certainly possible to identify patterns and common trends in specifc areas, each country is different from all the others, and requires an *ad-hoc* account. The second is that no state can be selected as a benchmark or model of migration control to be used as a measure for the compliance or rightfulness of all the others. Instead, the objective should be that of using comparisons to explain both the peculiarities of each case and the emergence of similar trends.

Another important conceptual challenge is to better understand the role and functioning of states and therefore, also, to produce more realistic appraisals of their capacities and "intentions" (Boswell, 2007). Too often, states are considered as monolithic, almighty actors endowed with a clear, coherent, time-stable stance. Something that enables a sort of personifcation of the state that is well represented by sentences commonly used like: "Greece doesn't want to control migrants" or "Denmark has decided to stop migrations" or "Spain has lost the control of its shores", etc. Also in this case, a more complex understanding and more sophisticated tools are necessary. States are internally fragmented institutions, conglomerates of functionally differentiated structures each one provided with an own perspective, functioning and apparatus. The governance of migrations, as adequately captured by concept of "migration regime" (Sciortino, 2000; Cvajner et al., 2018), is the result of the non-arithmetical sum of all the provisions, directly or indirectly affecting migration, that each of these structures implement and of the way in which all the other social actors react.

On the other hand, focusing on methodological issues, a more accurate understanding of migration control demands for systematic, multi-level, multi-perspective, and multi-disciplinary comparative effort. Analyses cannot limit themselves to the comparison of legal provisions, budgetary allocations or general statistics explicitly directed to migration control. They need to dig under the surface to unveil the actual impact on migrations of the whole state conglomerate. This, however, requires the combination of quantitative and qualitative research approaches and of different disciplinary standpoints at different levels. An effort that is inherently more timeconsuming and inevitably less effective if asked to provide defnitive conclusions.

Exploiting the results of an original, multi-sited, qualitative research among Ecuadorian irregular migrants in Amsterdam and Madrid, this chapter comparatively analyses the effects of migration controls in the two cities, focusing on internal controls. The irregular migration phenomenon, somehow the nemesis of migration control, is inquired from below – through the lived experiences of migrants in the two contexts – in search for hints able to throw some light onto the migration control realities in the Netherlands and Spain. From the 60 in-depth interviews and the collected ethnographic material, two ambiguous realities emerge, with differences and similarities, degrees of convergence and persistence of variance, complex enough to escape a clear-cut description in terms of opposites.

#### **8.2 The Quest for Migration Control: Policies and Implementation**

After the oil crisis of 1973, most European receiving countries observed a proliferation of policies, mechanisms, administrative structures, and legal frameworks dedicated to dealing with the control of international migrations. The real or perceived sense of failure signalled by the migration crisis of the 1990s intensifed the development and implementation of newer and increasingly-sophisticated policies. This perpetual escalation of control measures, on the one hand, and migrants' countermeasures on the other, is far from being concluded in our days. The main consequence for research has been a corresponding proliferation of studies, taxonomies, and classifcations in the attempt to analyse and better comprehend a constantly evolving landscape.

Regarding the classifcation of migration policies, a frst important distinction is the one between immigration policy and immigrant policy (Hammar, 1985). Immigration policies include those directed at controlling and selecting or deterring migration fuxes. Within this broad group, two main sub-groups can be distinguished: external control policies and internal control policies (Cornelius et al., 1994; Brochmann & Hammar, 1999; Cornelius, 2005; Broeders & Engbersen, 2007; Doomernik & Jandl, 2008; Van Meeteren, 2010). The second group, the one of specifc interest in this chapter, includes three main type of policies: (a) policies directed at making irregular residence diffcult and costly through labour market controls, for example, employer sanctions, employers' deputation to check for identities, labour site inspections (Brochmann & Hammar, 1999; Cornelius,2005; Broeders & Engbersen, 2007; Broeders, 2009) and policies aimed at the exclusion of irregular migrants from public services (identifcation checks in order to use services) (Van Der Leun, 2003; Van Meeteren, 2010). (b) Policies directed towards the identifcation, detention, and expulsion of irregular migrants (identifcation and surveillance systems, random checks in public spaces, administrative detention, readmission agreements) (Schrover et al., 2008; Engbersen & Broeders, 2009; Schinkel, 2009; Van Meeteren, 2010). (c) Policies directed at the regularization of irregular migrants (collective and individual regularization, *de jure*, and *de facto* regularizations) (Papademetriou, 2005; Schrover et al., 2008; Boswell & D'Amato, 2012; Chauvin et al., 2013).

Since the early 1990s, many scholars highlighted the existence of a gap between the laws and policies stated on paper and what they effectively achieved in "reality" (Cornelius et al., 1994). This awareness stimulated an intense debate over the need for a more comprehensive understanding of policies and their interaction with social life. Within this debate, a group of scholars underlined the necessity to shift the focus from policy formation or policy classifcation to policy implementation (Guiraudon & Lahav, 2000; Van Der Leun, 2003; Castles, 2004). Whereas many studies existed on laws, explicit regulation, policy documents and decision-making processes, scarce attention had been given to their implementation as well as to the resilience of lower-level counterforces (Van Der Leun, 2003; Lahav & Guiraudon, 2006). As pointed by Van del Leun, a large body of literature not directly concerned with the study of migrations, had already "warned against straightforward ideas about the process of implementation of public policies" (Van Der Leun, 2003, 28).

The shift of attention to implementation dramatically increased the complexity of the picture. If focus was on laws and regulations, researchers could refer to the offcial documents and statements by politicians and administrators. Enquiring into implementation, instead, forced them to get out of the libraries and adopt qualitative strategies to fnd and recompose the pieces of the puzzle.

Notwithstanding these diffculties and the relatively recent attention given to implementation, the efforts made in the last two decades have produced signifcant results. On the one hand, theoretical attempts have been made to develop frameworks of analysis. Since every national context produces distinctive practices of implementation, two questions have been raised: (a) what determines the specifc mode of implementation? (b) How is it possible to explain differences? Four aspects have been suggested as crucial to understand different practices: the peculiar national regulatory styles and traditions; the organizational culture of bureaucracies and the degree of discretionality; the grade of isolation of bureaucracies from external pressures; the social attitude and toleration towards informality (Scott, 1998; Heyman & Smart, 1999; Guiraudon & Lahav, 2000; Jordan et al., 2003; Van Der Leun, 2003; Lahav & Guiraudon, 2006).

On the other hand, researchers have analyzed policy implementation in different countries with the purpose of detecting possible common trends. Lahav and Guiraudon (2006) have indicated an on-going shift of focus in the implementation of policies. While before the migration crisis of the 1990s, controls were limited to border enforcement and were implemented exclusively by states' central institutions, after that, controls have been moving "away from the border and outside of the state" (Guiraudon & Lahav, 2000). This process has followed a threefold strategy: a shift outwards, with the adoption of remote control policies; a shift upwards, with the development of international frameworks for control; a shift downwards, with the delegation of control duties to the local institutional level. Another group of scholars have observed a slow but constant shift in the logic of policy implementation (Engbersen, 2001; Broeders & Engbersen, 2007). Broeders characterized this shift as the alternation between two contradictory logics of exclusion: exclusion from documentation/registration and exclusion through documentation/registration (Broeders, 2009). The frst logic intended to exclude irregular migrants, denying

them the possibility to acquire the documents necessary to access public services. While it may have been effective in fencing migrants' access to welfare, this logic did not prevent the growth of irregular migration and was ineffective for expulsions. The main objective of the second logic was precisely to make expulsions effective trough identifcation technologies and surveillance systems (Engbersen & Broeders, 2009; Leerkes, 2009). The correct identifcation of migrants was the main condition that origin states asked for, in order to accept their citizens back once they were expelled. While the frst logic has been implemented principally in Northern European countries, the second logic has been central to the European Union common policy and seems to be gaining importance in the rest of receiving counties.

Finally, several scholars have suggested the need to look beyond policies closely related to immigration control to fully grasp migration management (Finotelli, 2009; Garcés-Mascareñas, 2012). "Labour market, macro-economic, welfare, trade and foreign policies […] affect fundamental economic migration drivers" so "their infuence might actually be larger than specifc migration policies, which perhaps have a greater effect on the specifc patterns and selection of migrants rather than on overall magnitude and long-term trends" (Czaika & De Haas, 2013, p. 5). It seems possible to conclude that only the joint analysis of the interaction and the implementation of migration and refugee policies, labour market policies and welfare policies allows for a full picture of the framework within which migration control takes place and evolves.

#### **8.3 Internal Controls in Amsterdam and Madrid: A Case Study**

This section is based on broader research project (Echeverría, 2020) whose objective was to compare the experience of irregular migrants in two receiving contexts and to assess the differences and similarities that characterized the two cases. The aim was to offer empirical material for the theoretical refection on the practices and effects of migration control in two different European countries.

The chosen case was that of Ecuadorian irregular migrants in the cities of Amsterdam and Madrid. The empirical study combined ethnography and the collection of 30 in-depth interviews with irregular migrants in each context. The main research questions that prompted this study were: What have been the main structural characteristics affecting migration in the two contexts (migration history, migration regime, economics, welfare state typology, public and political opinion)? What was the experience of Ecuadorian irregular migrants within the two different contexts in the different aspects of their life (legal trajectories, work, housing, healthcare, controls, etc.)?

In this chapter, I will focus on three topics more closely related to the internal migration control. In particular, the experience Ecuadorian irregular migrants had in relation to work controls, *ad-hoc* identity and documentation controls and housing and healthcare access controls.

#### *8.3.1 Work Controls*

The work experience of Ecuadorian irregular migrants in Amsterdam and Madrid has presented several and evolving differences. In both cities, during the considered years, migrants experienced important changes in their working opportunities.

In Amsterdam, it was possible to recognize two very distinct moments. The frst, that lasted until the early 2000s, was characterized by the abundant availability of jobs in numerous sectors (construction, services, industry, cleaning). Even though it is not too marked, a certain gender distinction was observable with irregular man fnding good opportunities in the construction sector, in the port or playing and selling handicrafts in the streets, and irregular women in the cleaning sector, both in offces and private houses. In many cases, the migrants had more than one job. The second, from the mid-2000s on, was characterized by a progressive reduction of the available sectors. For irregular migrants it became increasingly diffcult to fnd working opportunities in sectors other than private-house cleaning. As emerged from the interviews, this change was largely due to a restrictive turn on the part of the authorities. The increased inspections on the working sites and the higher fnes in case of misconduct made it inconvenient for employers to hire irregular migrants.

The work in the hotel slowly reduced. A lot of people were looking for jobs and the employers preferred to hire those with papers. Maybe in the high season they would still take you on, but only because they really needed workers. But now it is diffcult because you must have papers, even during the high season… The market is dead, dead, dead… The only thing possible to survive now, because we still survive [the irregular migrants], is to work in houses… Why? Because they are private… it is private people that want you. There papers are not required, and there nobody stops you.

The second reason given by the migrants was that new groups of migrants who had papers started to fll the labour market and take the jobs they used to get before.

The frst to come were people from the East… they have papers, they started to work where we worked before. Then a lot of migrants from Southern Europe started to come… They go where the economy is still good. Many Spanish Ecuadorians [Ecuadorians with Spanish nationality] are coming, because they have papers and fnd work…

Yet, the shift towards the cleaning service in private houses was also part of a strategic option of the migrants themselves. Working in private houses offered several advantages in terms of salaries, fexibility of hours and security.

When the controls started to become tougher, I was scared. I said: no! I won't look for jobs in hotels and restaurants anymore… In houses it is much better… There the people know you, they give you the keys, you go, you respect your schedule… the hours you have to work and you leave… You don't see anybody and it is impossible that they come to check you. Moreover, they pay you more… a lot more. In the hotels and restaurants they used to exploit me. So much, so much! They paid me 6 euros, in my houses I don't get less than 10… Can you imagine the difference?

Notwithstanding the gradual reduction of sectors and increasing controls, Ecuadorian irregular migrants have generally judged their working conditions and opportunities in Amsterdam as good.

At the beginning it is hard, you don't know anyone… and it is hard. But then you start to know people, to make friends… They talk to you, they help you. Here there is a lot of work… once you start, you fnd more and more. There is plenty of work. And you can make money. Here the problems are others, the house, the papers, but there is work for everybody.

Even if labour controls became increasingly severe through the 2000s, both migrants and employers in Amsterdam have always been alert to the possibility of inspections. For this reason, a number of strategies have been developed both to employ irregular migrants and to escape possible inspections.

Regarding the frst aspect, e.g. the irregular employment of migrants, some sectors, for instance, port services, industrial cleaning and construction, appeared to have more inspections and required specifc strategies. The decisive factor seemed to be the size of the business. When the employer was a medium-sized or big company, a contract was usually needed. The strategies, therefore, were basically aimed at bypassing this limitation. The two main options were: for migrants to rent or borrow the papers of a regular migrant; for employers, to hire more than one worker with a single contract.

Once I worked in the port. We had to unload and load Russian ships… There you worked with the name of someone who had papers… Every morning when you arrived they told you: if the police of the port come, you have to say that this is your name… And you tried like hell, had to keep repeating to yourself who you were: Juan Charles, Juan Charles, Juan Charles… Sometimes they asked you just to check if you were alert. The "owner" of the job charged you a commission…

To avoid controls migrants and employers develop specifc strategies, which depend on the type of work. An important aspect, in all cases, is to try to pass unnoticed, especially when working on exposed sites.

A lot of friends have been caught because they were working outside. You must always work inside because if you are working outside they can always ask you for your working permit.

Many migrants agreed on the fact that in Amsterdam it is customary that inspections arrive because someone calls the police. Therefore, it is always advisable to be as little eye-catching as possible. *possible.*

Many times there were inspections when I was on the building site. I had to hide, go to the roof. They frst enter and ask, if they don't see anything weird, nothing happens, but if they see something suspicious, they call and more inspectors arrive. Many times, they come because there has been a complaint. Here [in the Netherlands] there are many complaints. A group of friends of mine, they were Brazilian, they were working on a building site in the street and they were listening to music that the people here do not listen to. When I work outside, on the street, for example painting, I always listen to a Dutch radio. If you want to listen to your music, use headphones and that's it. If you are showy, you fail! But if you learn to be discreet, there's no problem.

In restaurants and hotels, the owners always told the migrants where to hide in case of a labour inspection or gave them other instructions so as not to raise suspicions. In certain cases, they had a way to alert the workers back in the kitchen or in the corridors, about the arrival of inspectors, so that they had enough time to hide.

We knew what to do in case of inspections. In those years [before 2003] it was very unusual… but once we had an inspection. A Moroccan, who was the oldest worker, took me by the hand and we climbed from the stairs up to the roof. He told me to be careful because up there it was all greasy since it was where the extractors were released… After a while we heard something like a little bell, it was the cook beating with a knife on the metal… It meant the inspectors had gone… I didn't even see the guys of the inspection, their faces, what they looked like, how many they were. That was the only time, because before there were few inspections…

Also in Madrid, it has been possible to distinguish two very different phases regarding irregular migrants' working opportunities. The frst phase, which lasted until the end of the 2000s, was characterized by a great availability of working opportunities in many sectors. Although this was the case for both men and women, a rather marked sectorial division was registered. Men were mostly employed in the construction sector, women in private-house cleaning and the care sectors. The second phase, which started in 2008, with the beginning of the economic crisis, was characterized by a sharp reduction of the working opportunities that deeply affected all migrants. For irregular migrants, in particular, it became extremely diffcult to fnd any occupation. The changed scenario especially affected men. The most important sector where they had found opportunities, e.g. the construction sector, literally collapsed.

After one week that I had been here, a friend of mine took me with him to the building site. The boss said: perfect, you can start right away. After that, I always worked in construction. I had to adapt, to learn all the names, because in Ecuador we call the tools with other names… My boss helped me to get the papers… The frst year without a contract I earned 900 euro, then when I got the papers I started earning 1200 euro. It was very good. One day, in 2009, the owner of the company came and said to us: that's it. There is no more work. He closed the company and that was the end… Now there is nothing… for 2 years I have been doing little things to survive.

For those who were still in an irregular administrative situation or those who "lost" the papers (because unable to renew them or in case of administrative withdrawn), it became increasingly diffcult to fnd opportunities to work. Among the interviewed, few had been able to keep their previous jobs in restaurants and in transportation; others started to fnd jobs in a sector that until that moment had been exclusively for women, e.g. house cleaning and care, while others relied on small occupations such as painting, gardening, electricity, etc.

I worked in that discotheque for more than fve years… I had to clean and prepare everything for the next day… In 2007, the things started to go badly… Two of my colleagues were fred… My boss was very nice to me and he said that I could stay for some time. In 2008, they fred my boss and me… Luckily, I had unemployment beneft for more than one year… Now I basically have not worked for 3 years … I mean, sometimes a friend calls me for 1 month or little things… I am thinking of going back to Ecuador…

For women the situation has got worse as well. Yet, the cleaning and care sector seemed to be still offering opportunities.

Until recently I was working with a cleaning company… The owner was helping me with the papers… I was there for two years. Two years ago, though, he said to me: 'you had better not work until you have papers'. The people working there were legal, but they hired me because my cousins told him about me. He agreed to hire me… But since things have become more diffcult, he told me to stay at home. Now, I have been unemployed again for 4 months. It is diffcult because everyone asks you for papers… For one hour or two that you want to work they ask you for papers…

The effects of the economic crisis on irregular employment were made even worse by a stricter control policy and the availability of workers with a regular status. Until the mid-2000s before the start of the economic crisis, migrants' descriptions reveal a very relaxed situation. Most of these migrants never experienced a control on a work site and the employers were not worried about hiring people with an irregular status.

There was no problem… You know, everyone was illegal… so you just went and you started working. I think they knew that nobody was going to check, because otherwise they would have been more worried….

In the restaurant where I worked for more than 8 years, we never had a control..

From the mid-2000s on, and especially in certain sectors, a gradual increase of controls was recorded. The employers, who until that moment had been basically unconcerned, started to ask more frequently for papers or to develop strategies to avoid possible controls. Accordingly, also irregular migrants had to develop their own strategies to get hired.

They kept hiring irregular migrants. It became only a little more diffcult. When I fnally regularized… My name say it is Xavier Ramirez, we went to the working site and there were three Xavier Ramirez… My boss said to me: don't work there… I asked: why? He said: because there are two others with your name. I said: but you pay me the day? Yes! Since he had my documentation, he could do that… I went back home but he had to pay me the day. Why? Because he had my papers. Before, I could not say anything because it was me who was the one working with the name of another… But now… When controls increased, a lot of people made money acting as intermediaries.

Within this new changed scenario, the opportunities for irregular migrants became very limited. Many migrants decided to move to another country or to go back to Ecuador. Those who remained tried to survive doing small jobs in the construction sector, transportation, or in services. For men, an option was to switch to the cleaning and care sector.

Overall, while the double scenario is similar in both Amsterdam and Madrid, the underlying reasons for the dichotomy appear different, and likewise the consequences. In Amsterdam, the causes of changes experienced by Ecuadorian irregular migrants appear to be mainly political, in Madrid mainly economical. In Amsterdam, the increasing number of inspections in many economic sectors caused a sectorial shift on the part of the migrants. Since working in sectors such as construction, services and industry became increasingly diffcult and risky, irregular migrants moved to the private-house cleaning sector. In Madrid, the effects of the economic downturn caused a general reduction of the working opportunities. For irregular migrants, it became very diffcult to fnd a job in any sector. The reason in this case, was not, or not principally, that there were more controls, but simply that there was no work at all. Those migrants who had been able to regularize their status and who were also unable to fnd any employment have confrmed this impression.

Regarding the working opportunities for irregular migrants, two further differences can be underlined. Firstly, in the frst phase, Amsterdam displayed a more even distribution of irregular migrants in different sectors (construction, services, industry, cleaning in private houses) and then, in the second phase, a concentration in one (cleaning in private houses). Madrid, instead, displayed a more marked concentration in some sectors (construction, cleaning, and care) in the frst phase and, in the second phase, a concentration in two (cleaning and care) but with scarce opportunities even there. Secondly, the cases revealed a different situation concerning gender distribution. While in both cases a certain sectorial difference emerged, in the case of Madrid this was much more marked.

An interesting facet regarding the topic under discussion concerns the care sector and, in particular, the care service for the elderly. While this sector has had a crucial role in Madrid, employing a vast number of irregular migrants and especially women, in Amsterdam employment in this sector has been completely absent. As pointed out by many migrants, in the Netherlands, the government offers several subsidized services to the elderly so that no working opportunities "under the table" are available in the sector.

Finally, also regarding the working conditions, the two cases have shown a differentiated picture. In Amsterdam, notwithstanding the necessary sectorial shifts, the working conditions for irregular migrants have generally and steadily been valued as positive. The large majority of the interviewed migrants told stories of relative success. They were treated well, were able to save money and to fulfl their economic expectations. In Madrid a distinction must be made. The interviewed migrants clearly distinguished in their stories between the pre-crisis and the crisis period. The frst was generally characterized, although with a slightly higher number of exceptions, by a great availability of opportunities, good working conditions and economic success; the second, by very limited working opportunities, unstable and underpaid jobs.

#### *8.3.2 Ad-hoc Identity and Documentation Controls*

Regarding the experience of internal controls that Ecuadorian irregular migrants had in Amsterdam and Madrid two aspects will be discussed: A. the experience of police (or other authorities) controls that migrants had in their daily lives; B. the actual fear that migrants had of being deported.

The results that emerged from the feldwork have revealed two different situations regarding internal controls in Amsterdam and Madrid. In Amsterdam, there have not been *ad-hoc* controls on irregular migrants in the streets or in public places. Migrants, therefore, did not feel under direct threat and did not usually feel scared about moving around and carrying out their normal activities. However, the rigid checks regarding respect for the rules that regulate most social activities, such as, walking in the street, riding a bike, using public transportation, have been an indirect form of control. Irregular migrants know that a simple mistake, an administrative fault of any kind, can lead to an identity check, to administrative detention. For these reasons, these people are usually very alert to the situation around them at all times and very self-controlled in their public activities.

No, no… in this country there are no controls in the streets… I mean, there are controls in the labour sites, as for all workers, but not for the papers. Then if there is something irregular, they can ask you for your papers, but they never come for the papers… Here in the Netherlands, only if there is a complaint or if you made a slip, can they check you… Otherwise no… they let you live in peace… Those who have been caught, it is because they were in a nightclub, they had been drinking too much and they started a fght outside. Others, pitifully, were caught with the bicycles… If you make a mistake and you are unlucky, a policeman stops you… Many of us have fallen for a red light, or for other little things… Those little things betray you…

Regarding the possibility of being deported, the impression gathered is that in Amsterdam this has become in the last decade a very realistic one among irregular migrants. In other words, most migrants seem to know that if they get caught, the most probable consequence is that they are going to be deported.

There was a change around 1999 or 2000… Before that, even if they caught you, it was rare that they deported you… But, after that, they started deporting everyone… We were scared, you heard about this and that… Many people who were deported before 2003 were able to come back a week later… In Ecuador, you simply ask for a new passport… They had a different number each time… So, there was no problem… But after 2003, they started asking for the visa… If they sent you back, you could not return.

In Madrid, there has always been the possibility of *ad-hoc* controls on irregular migrants. Until the second half of the 2000s, though, these were very limited, unsystematic, and largely without consequences. Migrants, therefore, described a very relaxed situation and a negligible possibility to be deported.

I was always outside… nothing happened… When I was illegal, the police stopped me on two occasions… The frst time I was waiting for the bus… A policeman came and asked me for papers… I had only a photocopy of my passport… I was scared, but nothing happened… He looked at the picture, looked at me… and said: it's ok! Don't get into trouble… And he left… The other time was the same…

After the start of the economic crisis, this scenario changed. The controls on irregular migrants became more frequent and systematic. Every street, metro station or public place could be the place for a potential raid. These controls, however, were rather intermittent.

A couple of months ago, there were many controls… it goes in spells. In *Metro Plaza de Castilla* [the name of a metro station], there were those paisanos [policemen in plain clothes] as they call them. There were many of them, checking for papers. And also in *Metro Usera*… I always go there, because one of my cousins lives there… And I saw that they were also stopping women… they were taking them away… I could not believe it… Because, before, they used to stop only men, but now also women…

When the raids started, there were *latino* radio stations that alerted the places where the police were… They alerted the illegal migrant… Be careful in that station, they are checking for documents there… It was the people who called to say where the controls were…

They increased in particular months or weeks and diminished afterwards. While the fear of being deported certainly increased in the second phase, for the interviewed migrants, this possibility remains rather unlikely. As pointed out by many, to be actually deported, it is usually not enough to simply not possess a residence permit; the irregular migrant has to have a criminal record.

Here in Spain, it is not that they get you without papers and they deport you. No! You need to have a criminal record… You need to have been involved in something like fghts, thefts, vandalism… Otherwise it is very diffcult… They can even take you to jail or the CIEs [administrative detention centres] but they let you go after a while… A change took place with the beginning of the economic crisis. From that moment on, they started to check and deport the people. I had been stopped before, but, I swear, it was as if they checked you only to check you… The second time, I think it was two years ago [2011], a policeman stopped me. You could see that it was not like before. Now, they stopped the people and sent those without papers away. Now they were checking in order to deport.

#### *8.3.3 Housing and Healthcare Controls*

The access to housing and healthcare for Ecuadorian irregular migrants in Amsterdam and Madrid has presented a number of important differences.

Regarding the frst issue, the access to housing in Amsterdam has been generally much more diffcult than it is in Madrid. This has been due to two main factors: A. the lower supply of housing opportunities; B. the existence of a strictly regulated and controlled system of public housing. The combination of these factors determined a very precarious situation for irregular migrants. The available options were generally unstable, expensive and at risk of frauds.

The house is the biggest problem here in Amsterdam. Imagine: there is no house for the Dutch, so what do you expect for irregular migrants? I have changed more than 10 houses in the last number of years… It is really bad.

Here it is very diffcult. If you don't have someone who knows you, it is very diffcult to get a good house. You cannot go and say: I want a room or a house. They ask you for your residence permit. The other option is to sublet a room in a government house, but in that case, you never know if a control may come. Those houses are very much controlled.

In contrast, in Madrid, after a frst moment in which the housing opportunities had been relatively scarce, the situation rapidly improved.

At the beginning it was diffcult. The *Spaniards* didn't want to rent you a house without the *nomina* [a working contract] and you could not have a contract without the papers… So, the only option was to rent rooms… You know, in Ecuador there is a lot of space… we were not accustomed to renting rooms, to living with other people that you don't know… It was hard.

As many Ecuadorians and migrants from other countries started to get their status regularized, they were able to rent entire houses or fats and sublet rooms to the newly arrived. This determined a quick expansion of the housing opportunities and, therefore, the availability of relatively cheap, stable, and safe housing for the irregular migrants.

As far as the second issue is concerned, a similar situation has been found. In Amsterdam, access to healthcare has been much more problematic for irregular migrants than in Madrid.

For more or less 10 years now, things have been very diffcult. We were not 50 or 60 who went to the hospital… We were 500… Each time that you go to the hospital it costs at least 300–400 euros… So that was a debt that the government has. I think it was for that reason that they changed the law. Now you cannot buy a medical insurance, you need the residence permit. There is a fund that the hospitals can use in cases of emergency for us… [by "us" she means irregular migrants]. So now, what we must do is to hope not to get ill, otherwise you have to go and pay the bill… Sofa, that friend of mine, broke her leg… I think she has 15,000 euro of debt with the hospital…

In this case, the determining factor was the different regulations regarding access to the public healthcare system. Whereas in Amsterdam, irregular migrants have been excluded from non-emergency care since 1998, in Madrid they could freely access the public system until 2012. Hence, while for irregular migrants in Madrid, the issue of healthcare was basically not a problem, in Amsterdam they had to fnd ways to overcome the existing limitations. The most common option, in case of serious medical problems, was to go to the public hospitals and pay the costs at market prices. For minor problems, there was the option of medical support offered by humanitarian associations or by private unoffcial doctors.

Luckily, I never had problem, so I didn't need to go to the hospital. I discovered how expensive it was when I had to give birth. One echography costs 200 euro… Here they don't give you anything, absolutely anything if you are illegal. Now I am legal, and I pay insurance, so I can go whenever I want. But when I was illegal I could not. I mean, I could but I had to pay…

#### **8.4 Internal Migration Controls in the Netherlands and Spain: Differently Similar?**

The results of the case study presented in this chapter, far from providing the fnal word on the functioning of internal migration controls in the Netherlands and Spain, offer however interesting material for refection on the differences and similarities that have characterized their approaches.

A frst remarkable point is the confrmation of the dynamic character of migration control strategies in relation to more general societal processes, directly or indirectly, linked to migration. In the Netherlands, there has been a clear shift in time from a more relaxed to a stricter enforcement of internal migration controls. Until the late 1990s, irregular migrants could easily fnd jobs in a variety of sectors and controls on the worksites were rather limited. Likewise, access to healthcare and other social benefts was possible for everyone without regard to his/her migratory status. From 1998 on, things changed. Work controls increased in every sector and, in a relatively short time, the opportunities for irregular migrants severely reduced. Moreover, the Linking Act (*Koppelingswet*), approved that year, established a direct nexus between the possibility to access public services, such as, social security, healthcare, education or public housing, and the holding of a valid residence permit. These important transformations can be related, on the one hand, to a long-term process of deterioration of the social and political acceptance of migrants and, on the other, to a slow transformation towards a more qualifed and formal labour market. The Netherlands, in this sense, does not appear to be generically more prone than other countries to strict migration control, but rather to have adaptatively modifed its approach in relation to transforming of societal demands. Until irregular migration was not especially unpopular, and its contribution engrained well into the functioning of the national economy, controls were rather relaxed.

Also in Spain, the experience of Ecuadorian irregular migrants allowed to observe a transformation in time of migration controls. Whereas until the outbreak of the economic crisis in 2008, migration controls were very limited, the sudden and abrupt collapse of economy and the concomitant change of the political forces in charge had an impact in this area. The new, rightist government inaugurated a rhetoric against irregular migration and promised a tougher hand. In the immediate, this translated into a sensationalist yet ineffectual policy of random identity controls in the public spaces. For the longer term, into a promise – never really fulflled – of more controls on the labour market and the exclusion of irregular migrants from the welfare. The conditions for irregular migrants changed indeed, but as a result of the practical disappearance of working opportunities due to the economic crisis. Many irregular migrants left the country and the phenomenon appeared contained at last. What at a political level some tried to "sell" as a migration control success was, in fact, the result of self-regulatory socio-economic dynamics.

A second point of interest is the very different approach to internal migration controls in the two countries. Once the change of attitude was decided, the Netherlands adopted a comprehensive, systemic strategy based on the social exclusion of irregular migrants and the effective expulsion of irregular migrants through identifcation. The new course required interventions in different social realms (economy, welfare, housing) and implicated all state apparatuses. In Spain, the new strategy, mainly based on identity checks, was more explicit and direct, but rather unsystematic. These differences can be related to several factors that bring into play complex political, cultural, historical dynamics. Firstly, an historically different grade and mode of state/society interpenetration in the two countries. In the Netherlands, the state apparatuses appear to be much more pervasive in their ambition, and therefore also in their capacity, to regulate social transactions in a myriad of social areas and at different levels. Three examples of this clearly emerged in the feldwork: the regulation of the housing market; the extension of the welfare system (which for instance has a dedicated program for elderly care); the capillarity of regulation compliance (all regulations) controls in the public spaces. Secondly, a different political culture that affects the way in which controls are implemented and accepted by the public opinion. The practice of random identity checks in the public spaces, for instance, is considered unacceptable in the Netherlands, whereas the exclusion of irregular migrants from healthcare not. The opposite takes place in

163

Spain. It could be cautiously inferred that this may be related to a different societal valuation of political and social rights. If addressed in moral terms, as often done in the public debate (and not only there), what can be observed is not the presence of a morality on one side and the lack on the other, but, instead, the presence of different moralities. Thirdly, it is probable that a different underlaying objectives sustained the design and implementation of control policies. In the Netherlands, which has a long migration history and a relatively formal labour market, also in connection to a large and bipartisan political consensus in favour with a hardening of controls, the main objective was to reduce the burden of migrants on the welfare system and to further advance in the formalization of the labour market. In Spain, which has a very recent migration history and a largely informal labour market, the main objective was to display toughness in a conjunctural situation of political transition, without excessively interfering into the free regulation of the economic dynamics largely perceived as functional.

All in all, the results presented in the chapter show two very different realities in relation to internal migration controls in the Netherlands and Spain. Such differences can be explained, in part, by the different objectives the two countries were pursuing, in part, by the different approaches and implementation strategies adopted. Observed from a more distance perspective, however, these marked differences reveal also, and somehow paradoxically, a degree of similarity. In both countries, governments' approaches appeared to combine pragmatist – tendentially less restrictive – stances to more ideological/moralistic – tendentially more restrictive – ones in relation to broader and dynamic societal contingencies. In both countries, although in different ways and with a different timing, it was possible to recognize an overall tendency towards more restrictive policies very much linked to the political acceptance of migration. Finally, in both countries, the persistence of the irregular migration phenomenon, notwithstanding the efforts against it, revealed the existence of structural limitations to migration control that are common to all countries and especially to liberal democracies.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Part IV Labour Migration Policy**

# **Chapter 9 "Selecting by Origin" Revisited: On the Particularistic Turn of German Labour Migration Policy**

**Jan Schneider and Holger Kolb**

#### **9.1 Introduction**

In 2005, Christian Joppke published a spadework in the feld of comparative migration research. In "Selecting by Origin" he examines a triumphant cross-country victory of universalist philosophies and admission policies over principles of ethnic selectivity across Western democracies. In particular, he describes the end of selective settler state constellations in the nineteenth and twentieth century, in which the legal framework for immigration and naturalization epitomized an instrument for (ethnic) nation-building, and a turn towards source-country universalism and ethnic equality. His analysis tracks the sustainable ousting of negative discriminatory admission legislation and practice in democracies such as Australia and the United States, i. e. bans in immigration or naturalization for particular groups as enshrined in the—virtually emblematic—US 1882 Chinese Exclusion Act. Furthermore, he shows that even Western liberal nation states, which have given preferential access to residence and citizenship to particular groups due to colonial, national, ethnic or ethno-religious ties follow a universalistic trend, with ethnic migration losing momentum.

The root image for his analysis is "to assume a built-in tension between universalistic and particularistic elements in all liberal nation-states, the liberal component commanding nonascriptive, universalistic criteria and equity in the selection of immigrants, the national component (sometimes) commanding the opposite, in the name of reproducing the particular beliefs that constitute a political community" (Joppke, 2005, p.18). While he considers blatant forms of negative ethnic discrimination as "stamped out" and "racist", with no empirical example remaining in twenty-frst century Western democracies, he claims the more subtle form of

J. Schneider (\*) · H. Kolb

Expert Council on Integration and Migration (SVR), Berlin, Germany e-mail: schneider@svr-migration.de; kolb@svr-migration.de

<sup>©</sup> The Author(s) 2023

C. Finotelli, I. Ponzo (eds.), *Migration Control Logics and Strategies in Europe*, IMISCOE Research Series, https://doi.org/10.1007/978-3-031-26002-5\_9

positive ethnic selectivity tend to take over the same function. As the "reverse side of prioritizing some is discriminating against all others, […] most ethnically selective immigration policies have presented themselves as positively selective policies, even if their primary intent was the exclusion of particular groups" (*ibid*., p. 23). The most straightforward examples of persisting ethnic preference migration exemplifed in his book are the regimes associated with a diasporic concept of dispersed members of the tribe, who are by law allowed to "return" (as in the case of Jews in Israel) or to "repatriate" (as in the case of ethnic Germans from Eastern and Southeast Europe and the former Soviet Union).

Other ethnic immigration policies exemplifed in the book differ in two ways from this ethnic migration in its purest form: for one, the salience of the ethnic factor is frequently reduced in the sense that origin is not the exclusive, nor the primary criterion for selection. Rather, it is "nested within or tagged onto other, nonethnic selection rules such as skills or family ties", while the "overarching frame is nonethnic"—which makes the particular policy "less objectionable from a liberal point of view" (Joppke, 2005, p. 222–223). For another, they do not put the oftentimes complicated question of ethnic kinship upfront, but rely on country of birth or citizenship in the formal sense of state membership as a criterion for positive discrimination. This model may still be based on "panethnic constructs of state-transcending community", as the case with the Portuguese and Spanish preference regimes depicted in separate case studies (*ibid*., p. 25). In other constellations—and thus particularly relevant for the analysis of the distinct regime for labour migration in Germany undertaken in this Chapter—it will even ignore ethnic sub-affliations and rely solely on the requisite state (or civic) membership. Again, one needs to differentiate: while such nonethnic preference regimes may be the result of diplomatic reciprocity (involving the defnition of "best friends" between nation-states), they oftentimes also involve the expectation of "assimilability" (Joppke, 2005, p. 23, 26; FitzGerald et al., 2018) of nationals of particular countries due a similar or the same language spoken, cultural or religious proximity, or migrant networks already present in the country.

The purpose of this chapter is to make a humble contribution to these considerations—not by putting into question any of the persisting subtle, indirect mechanisms of discrimination observable in admission policies. Rather, we maintain that there is slow but steady return of frst-level source-country selectivity and particularistic features within the German migration policy. Thus, we take Joppke's analysis as a starting point for a case study and apply it specifcally *vis-à-vis* a clearly delineated segment of Germany's immigration policy: the regime for recruiting labour migrants. In doing so, we aim to add another, more subtle layer to the debate over source-country particularism. Rather than analysing immigration policy writ large, we focus on country-specifc confgurations in Germany's system for admitting labour from third countries. In addition to the "hard" legal regime for admission, residence and settlement in the narrower sense (Groß, 2018) we also consider "softer" aspects such as information campaigns, infrastructure for language acquisition or scholarships, i. e. non-legal frameworks. Empirically and methodologically, the paper is based largely on an extensive analysis of legal and policy documents.

A special role is played by the political and legal genesis of the so-called Western Balkans Regulation, which in the meantime has been the subject of several studies both in terms of its political context of origin and its legal dogmatics.

We follow a rather narrow concept of the term "regime" by limiting it to the labour migration from third countries and identifying its constituent parts. Thus, our understanding of a *labour migration regime* is more or less congruent with what Cvajner, Echeverría and Sciortino (2018, p. 69) have described as the "internalist" perspective. This perspective center-stages the regulatory and institutional framework of the country of destination—one could also say: its public policy. The state tries to develop and maintain a distinctive profle in its law, policies and institutions across sectors, levels and actors. We are interested in whether this profle—implicitly or explicitly—has discriminatory effects, i. e. whether the legal and institutional framework plays out in favour of citizens from particular countries of origin and thus discriminates against others. In doing so, we try to take into account that "migration regime is usually not the outcome of consistent planning" (Sciortino, 2004: 32), but rather a complex set of circumstance, in which "public and private bodies, as well as decision-makers and administrative agencies, can coordinate (or at least try to coordinate) their expectations and produce and carry out governing decisions" (Cvajner et al., 2018: 13).

Following this introduction, Sect. 9.2 will give a brief account of Germany's labour migration regime from the 1950s until the end of the Millennium. In Sect. 9.3 we describe the rather rapid policy change in the area of labour migration since the year 2000 from ethnic particularism to almost complete source-country universalism. The main part of the analysis, Sect. 9.4, tracks both the more obvious and the less visible cracks within this universalist regime which have occurred since 2015 and asks, whether the labour migration regime in Germany is on its way back to ethnic selectivity. In Sect. 9.5, we then put the "re-ethnicised" labour migration regime into a comparative perspective and observe a tendency of Germany to adapt certain traits that have traditionally been considered as typical for Southern EU Member States, suggesting a tendency towards blurring boundaries between formerly described clear-cut Northern and Southern models in the European labour migration policy domain.

#### **9.2 Source-Country Particularism: Germany's Labour Migration Regime Before 2000**

The history of labor migration to the Federal Republic of Germany is almost as old as the country itself. For heuristic reasons we differentiate the post-war history of labour migration until the new millennium into two different periods which are separated by the recruitment ban of 1973 and look at each of these periods specifcally under the lense of the extent the country of origin of a respective applicant for labour migration to Germany is used as key criterion for the screening and selecting process.

#### *9.2.1 The Pioneer of Labour Migration Policy in Post War-Germany: Guest Worker Recruitment 1955–1973*

In 1955, just a few years after the founding of the Federal Republic of Germany and already a few years before the onset of the post-war economic upswing, a German-Italian agreement was reached on the recruitment of Italian workers to Germany. This was soon followed by agreements with other mainly Southern and Southeastern European countries. In most cases, the concrete initiative for the agreements came from the countries of origin which hoped that sending workers would relieve the pressure on their domestic labor markets and that remittances would improve their balance of payments (Thränhardt, 2002, p. 347; Knortz, 2008, p. 67–68). In the case of Turkey, with which an agreement was concluded in 1961, geopolitical considerations also played a role: the country became a member of NATO in 1952, and the possibility of sending workers to Germany was intended to contribute to the stabilization of the NATO partner (Steinert, 1995). Accordingly, the agreements on guest worker migration were a case of labour migration without a (primary) motivation in terms of labour migration policy.

In the absence of a corresponding legal standardization of labour migration (there was no immigration law or equivalent legislation), "migration control […] resulted from the recruitment agreements themselves (Conradt & Hornung, 2020, p. 171; our translation)". The decisive or even exclusive selection criterion in this phase of labour migration to Germany was thus the origin of an applicant or, more specifcally, the question of whether an applicant was a national of a certain group of countries with which—on the basis of different constellations of motives—special contractual relations existed in the form of bilateral agreements on labour migration. The guest worker migration thus can be understood as a prototype or even the purest form of a "selecting by origin"-approach in the feld of labour migration.

#### *9.2.2 Exceptions from the Ban After 1973: Migration by Ordinance and Bilateral Agreements in the 1990s*

With the recruitment stop of 1973 guest worker migration ended; labour mobility from the formerly sending countries partially became part of the free movement of workers. However, the structure of a labour migration policy that focused strongly (or even exclusively) on an applicant's country of origin in terms of the dominant selection criterion persisted. At the end of the 1980s, certain sectors of the economy suffered a shortage of labour (e. g., agriculture, hotel and catering), despite generally high unemployment in the country. This led to a partial rollback of the recruitment stop. Particularly in the years following the collapse of the Eastern bloc in 1989/1990, the German government used a familiar pattern of migration control and concluded bilateral agreements with numerous Central and Eastern European states to employ their nationals as seasonal, contract, guest and border workers. Sourcecountry particularism at this stage was also affected by foreign policy considerations in the wake of the upheavals of 1989: the permission for time-limited employment of workers from Eastern Europe was aimed at supporting these states in the transition from a centrally administered national economy to a system of market economy and channelling the migration pressure from these countries to Western Europe and, above all, to Germany.

The extensively reformed Aliens Act, which came into force shortly after reunifcation did not contain any specifc material provision on the basis of which gainful employment migration could have taken place. Possible constellations of labour migration to Germany (also beyond bilateral agreements) were summarized in the ordinance on exemptions from the recruitment ban (*Anwerbestoppausnahmeverordnung* (ASAV)) in 1998, but were narrowly defned and—with the exception of seasonal employment privileging nationals of a few countries—hardly relevant empirically. The ASAV was a fexible but binding regulative instrument subordinate to a formal law, which was adjusted several times according to labour market needs. It already contained, for example, the so-called "best friends" regulation, which allows a certain group of nationals to be granted permission to pursue any employment in Germany. The according white list included countries from which, on the one hand, low immigration fgures are expected, but with which, on the other hand, very good foreign and economic relations have existed for many years.

As regards the selective bias of the labour migration regime, Christian Joppke, in his meticulous comparative study on immigration, integration and citizenship in the United States, Germany and Great Britain, qualifed German foreigners law as "violat[ing] non-discriminatory norms and the principle of source-country universalism in drawing categorical distinctions between non-privileged foreigners and privileged foreigners […], which would never pass constitutional muster in the United States" (Joppke, 1999, p. 76). This was about to change as of the year Joppke published these lines.

#### **9.3 Triumph of Meritocracy: A Universalized Regime for the Twenty-First Century**

The turn of the century was an important landmark in the history of German migration policy. On frst January 2000 a new Citizenship Act came into force, which modernized Germany's *ius sanguinis* dominated nationality law. This year 2000, however, also witnessed an important reform in the area of labour migration, which is particularly relevant for the main interest of this Chapter, the alternating ways the respective country of origin is used as a criterion of labour migration policy. The Green Card of 2000 (see Sect. 9.3.1), a special regulation for foreign IT specialists, was soon followed by a comprehensive immigration policy reform in 2005 which

also affected the area of labour migration (see Sect. 9.3.2). In 2012 the German government used the transposition of an EU directive for a further reform of labour migration policy (see Sect. 9.3.3) and in 2020 the Skilled Immigration Act came into effect (see Sect. 9.3.4). Since it is far beyond the scope of this Chapter to comprehensively inform about these measures, all these instruments are exclusively scrutinized under the specifc lens of a dominance of particularism or universalism in terms of the countries of origin of the respective applicants.

#### *9.3.1 First Steps of Farewell to a Labour Migration Policy Based on Countries of Origin: The German Green Card*

The introduction of the Ordinance on Work Permits for Highly Qualifed Foreign Professionals in Information and Communications Technology (IT-ArGV), which became known to a wider public as the German "Green Card", marked no less than the beginning of a new era in German labour migration policy since it was the starting point of a universalist era of labour migration policy: with this sector-specifc recruitment tool, an applicant's qualifcations began to replace his or her country of origin as the primary criterion for selection decisions. The focus on IT specialists from India, which was present in the political and media discourse before and after the measure came into force, fell completely short of the legal reality, as the ordinance did not contain any restriction on nationals of certain countries of origin. The criteria were exclusively merits-based, the decisive factor being a university or technical college education with a focus on information and communications technology or – in order to take into account the cases of specialists without university certifcates, which are not uncommon in the IT sector in particular – an annual minimum salary of at least EUR 51,000 (Jurgens, 2010). In sharp contrast to particularistic measures such as the guest worker recruitment and the attraction of posted and seasonal workers, the Green Card opened a pathway for citizens of any country.

#### *9.3.2 Qualifcations in the Centre: The Immigration Act of 2005 as Paradigm Shift*

As a second and much bigger step the Immigration Act (*Zuwanderungsgesetz*) came into force in 2005 and replaced the old Foreigners Law with an all-new Residence Act (*Aufenthaltsgesetz*). Yet, its introduction did not mean a landslide reform in the sense that it starkly liberalized access for foreign workers (Schönwälder, 2013, p. 277–282). In retrospect, the provisions for the admission of skilled and highlyskilled migrants as well as for self-employment as they were laid down in the 2005 law appear almost bizarrely restrictive, and they have undergone numerous amendments and (liberal) extensions ever since. However, they continued the trend initiated in the context of the "Green Card": promoting a qualifcation-based selection decision irrespective of the ethnic or geographic origin of the candidate. Immigration for the purpose of gainful employment from any country of the world, which had been narrowly introduced for a specifc sector through the Green Card, was generalized by the Immigration Act. The key question of labour migration policy "Where do you actually come from?" got replaced by "What qualifcations do you bring?", laying the foundation for a nascent cross-occupational, skills-based labour migration policy, which made the question of access decisions in the area of labour migration primarily dependent on the individual qualifcation of the applicant (and thus—*ex negativo*—not any longer on the applicant's country of origin or nationality).

#### *9.3.3 Implementing the Blue Card – And Much More Than That*

This fundamental shift in the meaning of criteria applied in the context of selection processes specifc to labour migration, described here as a universalist turn, was continued and further strengthened in the implementation of the so-called EU Blue Card Directive in 2012. The legislator made a conscious decision to place the residence title, which had to be created by the directive's requirements, at the centre of German law by implementing the European requirements as liberally as possible (Cerna, 2013). The consequence of this decision was a further strengthening of universalist principles in German law, since the Directive defned a valid employment contract, a certain form of qualifcation, and a minimum salary as the only selection criteria. Country-specifc considerations remained absent (Thränhardt, 2014, p. 9–10). However, the law implementing the Blue Card Directive went far beyond the mere implementation of European law requirements (Laubenthal, 2014, p. 485–486; Thym, 2017, p. 366). It also for the frst time introduced a residence title for job-seeking, despite the fact that there were no Union requirements in this regard. This job searcher visa was—and this being decisive for the purposes of this Chapter—universalistic from the outset, i. e. not limited to certain countries.

#### *9.3.4 The Culmination of Selecting by Qualifcations: The Skilled Immigration Act*

So far, the last step in a history of liberalization of German labour migration policy spanning over the last two decades was the Skilled Immigration Act of 2020. The core element of this act was to align the legal position of vocationally and academically qualifed specialists and to strengthen the options for training and postqualifcation of third-country nationals. The new and expanded options for labour migration within the Skilled Immigration Act are universal in that sense that they fully abstract from a potential applicant's country of origin. The law can thus be regarded as the culmination of a gradual process of geographical universalization of labour migration policy which barely left any space for considerations based on the respective origin of labour migrants and thus as the (so far) fnal stage of a process of two decades of source-country universalism.

#### **9.4 Back to the Sixties: The Return of a Policy of "Selecting by Country of Origin"?**

The history of German labour migration policy of the past 20 years so far was described as a linear amplifcation of a universalist philosophy of screening and selecting labour migrants by their merits. Recent developments, however, have the potential to challenge this assessment and instead pose the question of a legal und institutional rollback in a way that labour migration policy increasingly focuses on the countries of origin (again), instead of further disregarding the origin and exclusively focusing on the qualifcation of individual applicants. These tendencies can be found in migration law (see 9.4.1) but also in the specifc features of the institutional framework, of the polity, and of the way Germany organizes and funds migration policy (see 9.4.2).

#### *9.4.1 Selecting by Origin in the Legal Framework: The Western Balkans Regulation*

Just a few months after the entry into force of the Skilled Immigration Act another important decision in the area of labour migration was pending at the federal level, the question of how to deal with a specifc segment of labour migration and concretely with a section in the Employment Ordinance, which became publicly known as Western Balkans regulation. This exclusive entry channel is the most telling with regards to the observation of a particularistic turn of German labour migration policy and thus warrants a more detailed analysis.

The Western Balkans Regulation, which favours nationals of Albania, Bosnia-Herzegovina, Kosovo, North Macedonia, Montenegro and Serbia, was not introduced as an answer to particular labour demand in Germany's industry, nor did it primarily involve considerations of cultural or linguistic proximity, nor any intergovernmental strategic cooperation on labour migration. Rather, its introduction resulted from a grand bargain in domestic politics about how restrictions in Germany's asylum policy towards these countries, particularly their inclusion into the list of safe countries of origin, could be "compensated" for (Bither & Ziebarth, 2018, p. 13–16). Thus, it was a political attempt to "trade the asylum channel for a labour channel", to the beneft of nationals of these important source countries (SVR, 2017, p. 77). The only preconditions for nationals of these countries are a work contract with a German employer and a labour market test which assesses whether a German national or another entitled foreign worker can be given priority in flling the position. Its relevance for German labour migration policy is twofold: on the one hand, empirically, because the norm rapidly developed to be one of the central immigration channels for labour migration (SVR Research Unit, 2019). On the other hand, because the approach of primarily (or even exclusively) targeting the origin of a potential labour migrant can push the guiding universalistic principle of labour migration policy of the past 20 years (a focus on individual qualifcations; see 9.2), overboard again.

Originally, a sunset clause ruled the regulation to expire on December 31, 2020. The new regulation adopted in October 2020, however, extended this countryparticular labour migration channel for another 3 years. Yet it limited the number of approvals to 25,000 per calendar year starting in 2021 (while the original regulation did not provide for such a cap). Thus, the Western Balkans Regulation 2.0 is slightly more restrictive than the frst edition. However, this cannot challenge the notion that there are signs of a general change on the horizon: towards a country-specifc reparticularization and thus a renaissance of the relevance of origin as recruitment criterion in the area of labour migration management. This potential blueprint of a new structure and basic understanding of Germany labour migration policy can be observed on two different but argumentatively interconnected levels, the spatial and the temporal.

In the context of the 2020 discussion on the future of the Western Balkans regime, the original nexus to the policy goal of reducing asylum migration largely got lost. The instrument was no longer understood as part of a migration policy package or as an element to correct or compensate policy developments in other areas of migration policy, but as a stand-alone instrument of labour migration policy – praised and appreciated particularly by the employers' associations for being an unbureaucratic exception from the complicated standard procedures of labour recruitment. This again very clearly indicates the character of a migration regime as "negotiated outcomes of multiple actors and organisations with different interests and different functioning logics" (Finotelli and Ponzo, in this volume). A telling example of reframing the structural foundations of the Western Balkans regulation from a regulatory outsider to a potential blueprint of German labour migration policy was the argument put forward by the Committee on Labour, Integration and Social Policy of the second parliamentary Chamber, the *Bundesrat*. Its plea for a renewed evaluation of the measure was based on the argument that this would provide further and more detailed fndings "in the event that the system is extended to other states or regions" (BR-Drs. 490/1/20, p. 5). In doing so, the Committee implicitly resumed a discussion that had been taken place in the years 2016 to 2018 in the course of the federal government's planned (and so far unsuccessful) inclusion of the three Maghreb states of Algeria, Morocco, and Tunisia in the list of safe countries of origin. The ideas of the Liberal parliamentary group in the German Bundestag, which is part of the federal government, go much beyond. In a position paper, they go so far as to

subject not only the Maghreb states but also India and Nigeria, and thus two of the most populous states in the West, to this new "selecting by origin" principle.1 Thus, at least at the level of political discourse, the consideration of decoupling the measure from the original spatial restrictions becomes visible and proposes a labour migration policy in the style of the of Western Balkans regulations in other regions.

The tendencies outlined above to remove the exceptional character of the regulation by transferring it to other regions fnds its expression in a temporal dimension. A good example for this was the initiative of the federal states of Rhineland-Palatinate, Bremen, Hamburg and Thuringia in the second chamber not only aimed at eliminating the restrictions decided upon in the context of the extension, such as the introduction of a quota of 25,000 consents per calendar year, but also at extending the regulation by making this option permanent. Even stronger as an argument for this development is the agreement of the parties forming the federal government in the 20th legislative period to "make proven approaches […] such as the Western Balkans regulation permanent". Admittedly, this is only a political declaration of intent and it remains to be seen whether this announcement will be implemented. However, it is a clear expression of how strongly particularistic regulations are now seen as a standard instrument of labour migration policy.

#### *9.4.2 The Institutional Framework: Labour Demand, Migration Control and Development Cooperation as Drivers of Particularism*

Public policies are never confned to the statutory legal framework and its execution. This is also the case in labour migration governance: there are spheres beyond the law, in which institutions and (funding) programmes can let "country-specifc regimes" emerge, or contribute to their manifestation. Following a "universalistic decade" after the year 2000 (see Sect. 9.3), Germany has continuously invested more in institutional arrangements that facilitate labour migration from particular countries. Three main underlying rationales may be identifed for this (Beirens et al., 2019, p. 12–17): *frst*, current or future prospected labour demand in specifc areas is the most pressing factor from a domestic perspective—the nursing and care sectors serving as the emblematic case in Germany. Faced with the double challenge that nursing diplomas acquired abroad often do not meet the requirements for recognition and that Germany's starting position in a ferce global competition for care workers is rather unfavourable (not the least due to the diffcult German language), the Federal government has engaged in strategies for labour recruitment. *Second*, an

<sup>1</sup>See "Für einen Paradigmenwechsel in der Einwanderungspolitik. Positionspapier der Freien Demokraten im Deutschen Bundestag" of September 2022 (https://www.fdpbt.de/sites/default/ files/2022-09/Fu%CC%88r%20einen%20Paradigmenwechsel%20in%20der%20Einwanderungspolitik.pdf; accessed 27 September 2022).

important impulse for bilateral labour migration arrangements stems from the intention to incentivize cooperation with "partner" countries in order to reduce irregular migration, secure borders, or improve the return of nationals from these countries. Offering pathways into legal migration for work or vocational training can mark a fipside in formal or informal migration partnerships or agreements, which are primarily geared at restrictive measures and the prevention or (irregular) migration. *Third*, facilitating the entry of citizens of particular third countries for work or training may be an instrument of international diplomacy, economic exchange or development cooperation. In the latter, the case for legal migration is made on the basis of expected benefts for countries of origin, such as remittances or skills transfer. Without claiming completeness, the following three subsections provide a sketch of the institutional frameworks that are highly source-country-selective, providing further evidence for our thesis of an increasing particularization of Germany's labour migration regime.

#### **9.4.2.1 Triple-Win Programmes, Bilateral Agreements and Skilled Labour Bridges**

The frst generation of migration partnerships launched at the level of the European Union were driven primarily by the goal of managing and limiting unwanted migration. Nevertheless, following a "carrot and stick" logic, they also brought organised migration into play for partner countries: EU Member States were to engage in circular migration schemes in order to "address their labour needs while exploiting potential positive impacts of migration on development and responding to the needs of countries of origin in terms of skill transfer and mitigating the impact of brain drain" (COM, 2007, p. 2). Within the EU partnership with Georgia, Germany started a programme which provided the frst structures for a development-oriented approach to labour migration, supporting migrant entrepreneurship, circular migration and the involvement of civil society actors (Nordhus, 2015). It also laid the foundation for a subsequent pilot that recruited around 40 skilled workers from Georgia for the German hotel, catering and care sectors.

In the years to come, care and nursing evolved as the main felds of recruitment. The German agency for development cooperation GIZ started a programme called "Triple-Win Migration" in 2013. It targets countries with a surplus of professionally trained nurses whose qualifcations can be (partly) recognised in Germany. Candidates undergo linguistic and intercultural preparation in their home country. Once they are in Germany, they beneft from a swift procedure for recognising their certifcates and some of them undergo further training to adapt to the required standards. Ever since 2013, more than 3000 skilled workers from Serbia, Bosnia-Herzegovina, Tunisia and the Philippines have taken up employment or adaptation training in Germany, most of them with the option of a longer-term or permanent sojourn.

In the run-up to the adoption of the Skilled Immigration Act (see 9.3.4), a "Skilled Worker Agency for Health and Care Professions" founded by the Federal Government started its work at the beginning of 2019 to further support the recruitment of care workers. The agency "helps to bring large number of qualifed international healthcare professionals to Germany [by] organiz[ing] administrative procedures of application with domestic and foreign authorities and services". For each recruitment case handled, the agency charges the commissioning companies (e. g., hospital operators, nursing services) a fee of just 350 Euros; for 2020 there were call-off quotas for more than 1200 skilled nurses, primarily from the Philippines and Mexico (German Bundestag, 2020). In this sense, Mexico could become the new source-country of choice, as the German Federal Minister of Health travelled to Mexico in September 2019 and concluded an agreement with his Mexican counterparts to speed up the issuance of work permits.

Beyond the care and nursing felds Germany has also set up training partnerships in sectors such as construction and hospitality. For instance, the German–Moroccan Partnership for the Training and Recruitment of Skilled Workers provided vocational training to young Moroccans in hotels and restaurants in Germany. Participants undergo 3 years of full vocational training and acquire a German language certifcate. To be eligible for this programme, applicants have to prove they have practical experience in the hotel and restaurant sector, but they are not required to present a vocational qualifcation. A more general German–Tunisian Mobility Pact was commissioned by the German Ministry for Foreign Affairs and implemented by the GIZ between 2012 and 2016 to facilitate labour migration of the highly skilled and arrange language classes and internships for Tunisians in German businesses with the goal to facilitate labour matches. Eventually, the German Federal Government concluded a higher-level agreement for cooperation on migration issues with the Republic of Egypt in 2017 ("Agreed Elements of Bilateral German-Egyptian Cooperation on Migration"; German Bundestag, 2018b). Among other things, it involved Egypt into what so far might be the largest and most ambitious pilot programme in terms of supporting mobility for labour and training on the African continent: the project THAMM ("Towards a Holistic Approach to Labour Migration Governance and Labour Mobility in North Africa"). It will run between 2019 and 2023 and is worth about 20 million Euros, 15 million of which are provided through the EU Emergency Trust Fund for Africa (Beirens et al., 2019, p. 46). The project's aim is to strengthen partner institutions for labour migration in Egypt, Morocco and Tunisia and support them in providing safe, development-oriented labour migration. The various programme lines are supposed to foster the mobility of up to 750 young people from the three North African countries who are interested in migrating to Germany for labour or to receive vocational training (*ibid*.). Bearing in mind that at least Morocco and Tunisia have been called up as candidates for the special extension of the ultra-liberal Western Balkans Regulation (see 4.1.1) one could go as far and qualify the various programmes to facilitate labour mobility, including THAMM, as functional equivalents to a legal provision, as they facilitate skilled but also semi-skilled labour migration, therefore providing an institutional framework for preparatory language classes and (partial) recognition of skills and certifcates.

This was only a cursory overview of a plethora of country-specifc programmes funded through the German Federal budget, which foster labour mobility or vocational training. In any case, it becomes evident that, below the surface of an overly universalistic immigration law, there is a solid layer of policies and frameworks for labour migration that evidence structural source-country particularism.

#### **9.4.2.2 Migration Advisory Centres**

Another trend to country particularism—ironically—results from the domain of return policy. Since the introduction of two funding lines in 1979 to encourage the voluntary return of rejected asylum-seekers and other foreigners mandated to depart with the help of the IOM (REAG/GARP) Germany has been maintaining a list of countries "of particular interest to Germany in terms of migration policy", which defnes nationals who are eligible for return subsidies. Thus, return and reintegration aid is highly selective, and both the country list and the amounts paid out to returnees have been subject to bureaucratic engineering between the Federal Ministry of the Interior and the Federal States (*Länder*), taking into account current political developments.

As a result of the so-called refugee crisis in 2015/16, the number of third country nationals forced to return rose considerably. Besides legislative changes it was agreed within the Federal Government to invest massively into return and reintegration support. In fact, the considerable use of funds in a budget line linked to the Ministry of Economic Cooperation and Development for the reintegration of rejected asylum seekers marked a paradigm shift in the understanding of return (Schneider, 2022). One important step was the establishment and expansion of migration advisory centres in relevant third countries to coordinate the various return and reintegration measures. These centres were run by the German agency for development cooperation GIZ (see 9.4.2.1). At the same time, the new paradigm acknowledged that these countries might as well be source countries for prospective labour migration. Consequently, besides counselling those returning from Germany and assisting them with reintegrating, the centres' offcial task is to give advice to the local population on channels for legal migration to Germany (German Bundestag, 2018a). In May 2021, there were 12 of these centres up and running with more than 100 staff, namely in Afghanistan, Albania, Egypt, Ghana, Iraq, Kosovo, Morocco, Nigeria, Pakistan, Senegal, Serbia and Tunisia, and it was planned to open one more in Gambia.

The GIZ has a long record of facilitating labour migration, particularly through the transnational exchange of experts and the large triple-win programme (see 9.4.2.1). Thus, the migration advisory centres (and the associated networks and services such as information on, and referrals to, language and training classes) set up in those 13 countries provide an ideal infrastructure to link them to the new migration programmes fostering labour and vocational training such as THAMM. They provide for an opportunity structure in particular partner countries (namely those which at some point had been major countries of origin of asylum-seekers), which are unavailable to people in other third countries (who might yet be interested to come to Germany for training or gainful employment). While this institutional framework certainly does not constitute positive discrimination in legal terms, such as the Western Balkans Regulation does, it still favours distinct (national) groups and is thus a subtle expression of source-country particularism.

#### **9.4.2.3 Recruiting Seasonal Workers**

As one last example of imminent source-country particularism in German labour migration policy we want to refer to the category of seasonal work. For many decades, this category has been of utmost importance in quantitative terms, with around 300,000 admissions per year (Palumbo & Corrado, 2020). Bilateral placement agreements are a reminiscence of Germany's recruitment particularism in the 1980s and 1990s (see 9.2.2), which became meaningless with the accession of Eastern and Southeast European countries such as Poland, Romania, Bulgaria, Croatia, Hungary and Slovakia to the EU. A changing geopolitical situation resulted in a new reservoir of workers, putting third-country specifc recruitment policies on hold. Until very recently, the demand for seasonal workers (mostly for the seeding, planting and harvesting) could be satisfed primarily with Poles and Romanians, i. e. through intra-EU mobility. However, demands for allowing the recruitment of workers in (designated) third countries were increasingly voiced as this reservoir dried up. Among other reasons, for many Central and Eastern European citizens more attractive employment opportunities have been opening up, either in other sectors in Germany or in their home countries. In fact, the precondition for admission of seasonal workers from third countries (Section 15a Employment Ordinance) is the existence of bilateral agreements between the German Federal Employment Agency and the labour administration of the country of origin in question. Upon provision of a work contract, a residence permit for up to 6 months can be issued for activities in the agriculture, forestry, gardening and the hotel and restaurant industry. But no such agreement had been concluded with any third country, and the provision in the Employment Ordinance therefore ran empty for many years (Offer & Mävers, 2016, margin number 10 to Section 15a Employment Ordinance).

This changed in 2020: even before the onset of the Corona-pandemic and the travel restrictions in March 2020, which severely aggravated the challenge of recruiting workers, an agreement was signed with Georgia to start a programme with 500 seasonal workers per year. The agreement was the frst to be struck as a result of a whole series of bilateral talks to initiate placement agreements with third countries, among them also Bosnia-Herzegovina, Albania, Moldova and North Macedonia (Lechner, 2020, p. 17), but also Ukraine. What was planned as a slightly cautious pilot scheme with 500 recruits for the harvesting season of 2020 at frst had to be postponed and was then boosted in the following year. After the Corona pandemic had put many farms in dire straits and a return to freer cross-border movement was on the horizon for the 2021 harvest season, Germany put its money where its mouth is: the quota was increased tenfold to 5000, immediately applicable. Given that the interest in employment opportunities in Germany is considerable in Georgia, the years of repeated seasonal employment ahead could be the starting

point for deepening labour migration relations in other sectors of the economy as well. The new seasonal workers scheme with Georgia (and potentially soon with other third countries) can thus be interpreted as the latest expression of a shift towards a much more country-specifc regime of labour migrant selection in Germany.

#### **9.5 Discussion and Conclusion: Blurring Boundaries and Unexpected Similarities Between Germany and Southern Europe**

The question whether the German regime for labour migration has departed from the twenty-frst century universalistic victory still awaits a fnal answer. However, the described developments in its legal and institutional framework are clear markers of the growing importance of what—following Joppke (2005)—we might coin *source-country particularism* (regarding confgurations in which nationals from particular third countries are preferred over others by law or regulation) or *preferential bilateralism* (delineating cases where contractual agreements, other *soft law* or a favourable institutional framework are benefting nationals of particular countries, thereby discriminating against others).

The analysis of country particularism in Germany almost inevitably brings to mind the migration regimes of Southern European states and their confgurations for labour recruitment, particularly in Italy and Spain, challenging also the North-South divide in immigration policies. Both countries, for quite some years now, display "hybrid" labour migration regimes in the sense that they are universalistic in principle but feature distinct particularistic pillars. The latter come primarily as *preferential bilateralism* and are typically rooted in a strategic *quid pro quo* logic of cooperation with third countries, which began to take shape with the increasing migration pressure faced by the Mediterranean EU Member States, particularly in the frst decade of the new century (Beirens et al., 2019, p. 39–42). Special labour migration quota or pathways for vocational training cast into bilateral agreements became part and parcel of broader arrangements for cooperative migration management, which—from the European vantage point—primarily aimed at reducing irregular migration and securing external borders and the functioning of the asylum system (Ferreira, 2019). Given the fact that the details about the specifc traditions of Southern European countries of resorting to source-country particularism have been elaborated in detail elsewhere it is not necessary to explain those in order to highlight some unexpected parallels between Germany and Southern Europe: Germany is displaying policy traits traditionally associated with the EU Mediterranean Member States and integrates various country particularistic structures into its labour migration regime. The recent agreement on the placement of up to 5000 seasonal workers from Georgia most closely resembles the southern pattern of selective cooperation. Even if Germany does not yet have any outspoken migration-related bilateral agreements at ministerial level in which fxed quotas are promised, the various state-sponsored country-specifc programmes can be understood as functional equivalents to this, as can the growing network of migration advisory centres in various third countries—even if their substantive focus has so far been on providing information and to support the reintegration of returnees. In one respect, Germany's country particularism even goes beyond that of the Southern EU Member States: with the Western Balkans regulation, positive discrimination in recruitment policy in favour of nationals of six countries was enshrined in the Employment Ordinance—a *quasi*-law for which amendments are typically the result of parliamentary resolutions in both chambers and not only decreed by the responsible Ministry of Labour and Social Affairs.

What remains unanswered (at least in this article), however, is the question of how to explain these frst indications of a new particularistic trend of German migration policy. Putting the German case study into a comparative framework might help to generate some explanatory hypotheses. In the case of Southern European countries cooperation agreements are certainly related to the imperative to intensify cooperation against irregular migration. In Germany the motivation is less clear. But it cannot be excluded that the newly emerging particularism also has its roots in a mix of geopolitical interests and intention to intervene on the routes of unwanted migration fows towards Germany—the Western Balkans Regulation (despite the fact that it originally was only motivated by consideration of domestic politics) gives rise to such an explanation. With regard to other geographical areas the motives for the emergence of country-specifc frameworks might be even more opaque, i. e. resulting from contingent opportunity structures, rather than from distinct clear-cut policy rationales. As in the case with the THAMM project in North African countries, foreign policy considerations, development cooperation, migration management objectives as well as domestic demand on the German labour and vocational training market all may play a role, thus forming a bundle of motives which constitutes a (more or less) coherent policy regime.

The renewed importance of the national origin of would-be migrants in Germany's post-2015 migration regime is not only at odds with the basic principles of its labour migration law, which is supposed to regard the interests of the Federal Republic of Germany in terms of its economy and labour market—and thus focuses primarily on qualifcations. Moreover, it is problematic from a normative perspective: origin is an ascriptive characteristic and as a primary criterion for labour migration requires much more justifcation than non-ascriptive, merit-based characteristics such as skills and qualifcations, which depend on performance and personal effort. The growing importance of *source-country particularism* in labour migration control is therefore irritating, besides the fact that such positive discrimination cast into the legal framework, as the case with the Western Balkans Regulation in the Employment Ordinance, can hardly be reconciled dogmatically with the norm structure of the migration policy regime as it has grown over the past two decades, aligning the latter to the primary criterion of qualifcation. Undoubtedly, it would be an exaggeration to state that Germany has already undergone a fundamental backlash in the sense that an applicant's ethnic or national origin has become the key

criterion for admitting labour migrants. However, with the Western Balkans Regulation as a potent and meaningful instrument of positive discrimination which defnes origin as a primary criterion, Germany has opened up a path that might and—according to the agreement of the parties forming the new government—will be used in a broader way in the future.

As far as the institutional framework beyond the mere legal sphere is concerned, our analysis has shown increased dynamics: Germany's labour migration-related activities abroad cover a growing number of third countries, and they have picked up in volume. As such, the migration advisory structures, training programmes and skilled labour bridges would be unsuspicious of discriminatory country particularism if they resulted from the free play of a transnational labour market and global value chains. Indeed, employers follow the economic principles of supply and demand and try to meet their labour needs according to the opportunities available to them. In times of growing sector-specifc domestic labour shortages, these options quickly reach their limits, as employers' networks in third countries through visits or diaspora connections tend to be limited (Beirens et al., 2019). If then, we argue, the state invests massively in labour recruitment and identifes such bilateral mobility-promoting constellations as part of the respective government policy (explicitly or implicitly linking them to other policy felds such as migration control, return policy or to development cooperation), it leaves the path of a universalistic, "country-blind" migration regime. In that same vein, carving a corridor for up to 5000 Georgian nationals for seasonal work in agriculture as of 2021 is a clear expression of positive discrimination and source-country particularism. Given the serious domestic policy discourse over a spatial extension of the Western Balkans Regulation towards countries in the Maghreb and the recent popularity of legal migration pilot projects and talent partnerships within endeavours towards a common European Union Framework for migration and asylum management (COM, 2020, p. 17–19, p. 24–29), one is to expect growing relevance of *origin* as a central criterion in Germany's labour migration regime—and thus a further approximation towards Southern EU Member States with a tradition of *preferential bilateralism* and country-specifcity in their migration systems: Ethnic selectivity in the area of labour migration policy might be increasingly utilized as a means to boost effectiveness in other sections of migration control. This would imply a return of a policy of "selecting by origin"—in a different shape as described in Joppke's analysis, but with the same potential of discriminatory power.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 10 The Admission of Foreign Workers to Italy: Closing the "Gap" with Northern Europe**

**Camilla Devitt**

#### **10.1 Introduction**

A common perspective in labour migration studies is that of a North-South divide in European labour migration governance, with Southern European countries exhibiting a distinct – and generally less effective – approach to the admission of migrants. In particular, Southern European systems are known for their immigrant quotas, permanent low-skilled labour immigration, high levels of irregular immigration and cyclical amnesties, while Northern European regimes tend to facilitate the entry of highly skilled labour immigrants, restrict the low-skilled to temporary programmes and oppose mass regularisations of irregular migrant workers (Baldwin Edwards, 1998; Castles, 2006; Finotelli & Sciortino, 2009; Sciortino, 2009; Pastore, 2010; Salis, 2012; Bonizzoni, 2018; Boucher & Gest, 2018; Colombo & Dalla-Zuanna, 2019; Geddes & Pettrachin, 2020).

This chapter explores the Italian labour immigration regime to assess whether and to what extent its admission policy and practice *has* traditionally diverged from the most common features of Northern European systems. I furthermore investigate whether, if it has been distinctive, it has remained so since the disruption of the international fnancial crisis of 2008 (the "Great Recession"). The study is based on an analysis of secondary literature on Italian and other West European labour migration regimes, along with OECD and EUROSTAT migration and employment statistics and Italian ministerial data on quotas (provided by the Ministry for Work and Social Policies) and Italian National Institute of Statistics (ISTAT) data on residency and citizenship. I fnd similarities and differences between the Italian regime and labour immigration regimes in Northern Europe between the late 1990s and 2008.

© The Author(s) 2023

C. Devitt (\*)

Trinity College, Dublin, Ireland e-mail: DEVITTCA@tcd.ie

C. Finotelli, I. Ponzo (eds.), *Migration Control Logics and Strategies in Europe*, IMISCOE Research Series, https://doi.org/10.1007/978-3-031-26002-5\_10

The similarities with Northern European regimes have increased since the Great Recession, with, for example, a more restrictive approach to infows of non-seasonal workers from outside of the European Union (EU) and a stronger reliance on the free movement of workers from Romania and various non-EU immigrant channels outside of work for low-medium skilled labour needs.

It is important to clarify at this point what is understood by a labour immigration regime. In the context of the EU, labour immigration policy generally refers to the regulation of immigration for purposes of work from outside of the EU. I use this defnition, however, in line with the broader regime concept used in this volume, rather than limit the analysis to laws and policy, I also discuss practices, for example, the practice of regularising undocumented migrants within annual quotas for the admission of migrants to Italy, in order to provide a more holistic understanding of the functioning of systems. Indeed, *de facto* labour migration regimes combine formal laws and policy with informal norms and practices, refecting socio-economic and administrative interests and logics. I also discuss policy on free movement, particularly after the Eastern enlargements (2004 and 2007), given that European policymakers view intra-EU East-West mobility as an alternative to non-EU labour immigration, particularly for low skilled occupations. Other types of migration for non-work purposes, including family, humanitarian and student migration, are also, to some extent, viewed as *de facto* labour immigration, as many of the migrants using these channels subsequently work in receiving countries. However, unlike East-West intra-EU mobility, these migratory forms are not always primarily workoriented. While I do not explore policy on these types of immigration, I discuss them with regards to their role as substitutes for labour immigrants.

The chapter will be organized as follows. First, I briefy outline some common Northern European labour immigration management tools and the channels through which foreigners have entered those labour markets over recent decades. Second, I describe the Italian labour immigration regime, including the management of EU mobility, prior to and after the Great Recession. These descriptions will include the policy and, as noted above, the practice of the regime(s), as the big "gap" with regards to Italian immigration policy has been a huge disjuncture between law/policy and practice. Third, I highlight differences and similarities between the Italian regime and Northern European regimes prior to and after 2008. The Chapter will end with a brief conclusion.

#### **10.2 Northern European Labour Immigration Regimes**

Over the past decade, Northern European labour migration regimes have shared key management tools, however, they also differ from each other in terms of the proportion of migrants gaining admittance through various labour and non-labour migration channels. There has been a general evolution towards a more selective approach

to labour immigration (de Haas et al., 2014), with, a notable facilitation of the entry of highly skilled non-EU workers. The latter play a crucial role in many European welfare states, particularly health systems (Bobek & Devitt, 2017), are viewed as harbingers of innovation and economic growth in competitive knowledge based economies and are also judged to be easier to integrate than low skilled migrants. The admission of the highly skilled is facilitated by reducing work permit processing times and providing them with stronger welfare and residency rights (Paul, 2012b; Ruhs, 2013; Devitt, 2014; Bonizzoni, 2018). While many European countries introduced points based systems in the frst decade of the twenty-frst century for the admission of highly skilled non-EU workers on the basis of their skills and experience, these channels have never been as signifcant, in terms of numbers of immigrants admitted, as demand based systems, where workers gain admission on the basis of a job offer (Chaloff & Lemaitre, 2009; de Haas et al., 2014).

However, despite a policy emphasis on attracting highly skilled immigrants, the migration of highly-skilled non-EU migrants to Northern Europe was limited, apart from to the UK and Ireland, during the frst decade of the twenty-frst century (OECD, 2009c). Subsequently, data on Blue Card approvals in continental Europe show the small number of non-EU highly skilled migrants availing of this channel into Europe over the past decade. The Blue Card is an EU wide work permit for highly qualifed employment, introduced in 2009, and recognised in 25 EU Member States. In order to apply for a Blue Card, the migrant must have a binding job offer or valid employment contract for at least a year, along with qualifcation certifcates and verifable work experience. In 2016, there were 20,979 blue card approvals across the 25 Member States, with the highest number issued in Germany (17,630) and France (750) (Burmann et al., 2018).

With regards to managing the admission of non-EU workers with specifc job offers, governments frst generally seek to ensure that resident or EU workers are not available for the jobs. This is done through labour market tests, where the jobs need to be advertised locally and in the EU for a specifc period of time to make sure that no suitable candidates are found within the EU. The other approach is the use of shortage occupation lists; governments produce lists of occupations, which are deemed in shortage on the basis of labour market analysis, for which work permits for non-EU workers can be requested by employers, without the necessity of passing a labour market test (Chaloff & Lemaitre, 2009).

While entry rights for low skilled migrants have been expanded since the mid-1990s, following circa 20 years of restriction, they tend to gain entry on time limited visas e.g., seasonal permits and working holiday visas (de Haas et al., 2014). Indeed, one of the main immigration trends worldwide over recent years has been an increase in temporary migration (Boucher & Gest, 2018). In the European context, this is due to concern regarding the integration of low-skilled non-EU migrants, based on perceived failures in successfully incorporating some non-EU origin communities, settled in Northern Europe following a period of large-scale low-skilled labour immigration in the post-war era (Castles, 2006).

Apart from skills and occupations, national origin plays an important role in non-EU labour immigration management, beyond the preference given to nationals of EU Member States. Indeed, various European countries manage non-EU immigration within the framework of bilateral agreements, which give preferential treatment to labour migrants coming from particular countries (de Haas et al., 2014; Paul, 2015). General caps or numerical limits on labour immigration have traditionally not been used in Northern Europe (with some exceptions such as in Switzerland, Alfonso, 2004), though they have been used in particular schemes for low skilled and high skilled non-EU migrants. Regarding the former, most OECD countries apply caps, quotas or targets for the admission of low skilled migrant workers (OECD, 2009a, c).

Since the EU enlargements of 2004 and 2007, some European countries have managed to fll low skilled labour gaps with mobile Eastern Europeans, thereby allowing for continuing/new restrictions on low skilled non-EU labour immigration (Paul, 2012a). Indeed, countries with large levels of free movement such as Norway, Austria, Germany and Netherlands tend to have low levels of non-EU permanent economic immigration (de Haas et al., 2014; Boucher & Gest, 2018). The labour market participation of family, humanitarian and student migrants also partially substitutes non-EU low-skilled labour immigrants (Finotelli, 2009; Devitt, 2014).

There have, nonetheless, been signifcant differences between Northern European countries with regards to the proportion of temporary versus permanent non-EU immigration and levels of free movement and other non-economic immigration. For example, taking the cases of Germany and France, prior to the 2004 enlargement Germany received a particularly high proportion of temporary non-EU migration, as opposed to France, where permanent non-EU labour immigration has been more signifcant proportionally. More recently, Germany has a much higher proportion of EU free movement in terms of overall permanent immigration than France. Finally, while asylum has been a signifcant channel of migration to Germany, family reunifcation has been more prominent in France, since the stop to labour immigration imposed in both countries in the early 1970s (Finotelli, 2009; Pastore, 2010; Boucher & Gest, 2018; OECD, 2020).

With regards to undocumented migrants, various Northern European countries, in particular the UK and Germany, are estimated to have large numbers (over a minimum estimate of 196,000 in 2008), which are comparable with the estimates for Italy and Spain (Clandestino, 2009). Time limited regularisations programmes have been recurrently used to legalise their position in various Northern European countries, including France, Germany and the Netherlands, with substantial numbers regularised in France and Germany. These regularisations are primarily granted on humanitarian grounds and are often associated with rejected asylum seekers (de Haas et al., 2014; Kraler, 2019). While governments carry out these humanitarian regularisations, such as the *Duldung* in Germany, because of diffculties returning the people to their origin countries, the regularized migrants are, in fact, functional equivalents to labour migrants.

#### **10.3 The Italian Labour Immigration Regime**

#### *10.3.1 Italian Labour Immigration Regime Mid 1990s – 2008*

#### **10.3.1.1 Entry Mechanisms: Annual Quotas and Recruitment from Abroad**

Since 1995, the admission of non-EU workers to Italy is largely subject to annual quantitative ceilings or quotas. Based on the Turco-Napolitano law of 1998, the quota is issued by the Presidency of the Council of Ministers following an examination of labour demand and supply data and the country's integration capacity, as well as consultations with relevant ministries, representatives from local authorities and social partners (Salis, 2012; Bonizzoni, 2018). However, despite comparatively high levels of labour immigration in the 2000s in the European context, with quotas reaching over 150,000 slots from 2005 (Sciortino, 2009; Devitt, 2011), the system was generally argued to have provided inadequate numbers of places to meet employers' demand for migrants, amongst other failings.

One of the main sources of data on potential labour demand has been the Excelsior survey of private sector frms, carried out by the Union of Chambers of Commerce since 1997. This survey has underestimated demand, as it does not cover agricultural frms or private households, which are signifcant employers of migrants. Moreover, scholars have highlighted the diffculties Italian businesses have in forecasting their needs (Sciortino, 2009; Salis, 2012; Colombo & Dalla-Zuanna, 2019). The setting of the quota is above all a political choice, largely determined by public opinion on immigration, which is negative compared to the rest of the EU (Geddes & Pettrachin, 2020). This explains why, until the middle of the frst decade of the twenty-frst century, quotas were set below Excelsior estimates and 50% lower than the number of work permit applications made (apart from in 2006). Nonetheless, the number of applications for work permits is not a valid indicator of labour demand in Italy. Indeed, the quota system has also been used as a channel for family reunifcation and other forms of chain migration and, as I explain below, as a *de facto* mass regularisation of undocumented migrant workers already working in Italy (Salis, 2012). There have furthermore often been delays in issuing the quota decrees (they have been released a few days before the end of the year they were supposed to regulate) and in issuing work permits (it can take more than a year for a residence permit to be issued) (Finotelli & Sciortino, 2009; Salis, 2012).

On the other hand, the quota system has been quite liberal with regards to conditions surrounding the issuing of permits and the rights to gain permanency in Italy. While labour market tests are conducted, the applicant employer can confrm the request for a non-EU worker even when European workers are available (Salis, 2012). Non-seasonal labour migrants can change employers and sectors and extend their permits, achieving the possibility of long-term residency status after 5 years (Bonizzoni, 2018). The Italian quota system thus facilitated permanent low-medium skilled labour immigration in those years, as did the recurrent regularisations of undocumented workers, which I discuss below.

The quotas can include sub-quotas for particular occupations/sectors, as well as for specifc national groups and the self-employed. These shortage occupations/sectors have generally not required a high level of education, and have included domestic and care work, fsheries, construction and truck drivers (Sciortino, 2009). From 2005, domestic/care workers increased their share of the general quotas for nonseasonal workers, taking up the entire quota for non-seasonal work of 2008 (Salis, 2012; Bonizzoni, 2018). The nationality-based quotas are based on international agreements concluded with sending countries, which aim to provide countries with labour immigration quotas in exchange for active collaboration on border control and readmission.

The quota system has not prioritised highly skilled workers, apart from in 2001, when 2000 entry slots were reserved for professional nurses and between 2003 and 2007, when between 500–1000 slots were reserved annually for IT professionals (Sciortino, 2009). There is also a quota channel for those with specifc pre-departure training in their countries of origin, however, permits provided on this basis have been quantitatively negligible. Moreover, in this period, specifc categories of skilled workers (including intra-company transferees, academic researchers and professional nurses) could gain admission outside the quotas due to the specialized nature of their work or the limited time to be spent in Italy e.g., posted workers or seconded executives. Most had time limited permits and were restricted with regards to changing employer and sector (Salis, 2012; Bonizzoni, 2018). However, while the quota system generally does not give preferential treatment to skilled migrants, it became increasing selective from 2000 onwards (with minor fuctuations and excluding 2006, when a large unselective quota for non-seasonal workers was issued), allocating most slots on the basis of nationality, seasonal work and occupation/sector (Bonizzoni, 2018).

Apart from the problems inherent in forecasting demand and the gap between forecasted demand and the size of the annual quotas, another much maligned issue with the Italian labour immigration system is the requirement of nominal recruitment from abroad. This means that employers must request authorisation to hire a foreign worker living abroad before s/he has come to Italy. In reality, as is well understood by Italian policymakers (Salis, 2012), immigrants arrived irregularly or more commonly overstayed their tourist visas, found employment and then sought regularization via the quota system- going home temporarily in order to do so – or a mass regularization.

#### **10.3.1.2 Regularisations**

While the quotas have partly functioned as *de facto* mass regularisations, the Italian state also carried out 5 explicit regularization programmes for undocumented migrants between 1986 and 2002. Indeed, more than a third of foreigners legally resident in Italy received their last residence permit in a regularization (Colombo & Dalla-Zuanna, 2019). Moreover, both left and right wing coalition governments have carried out these programmes; indeed, the largest ever regularization was carried out by a centre right government in 2002. A common EU visa policy, recurring mass regularisations, the use of the quota system as a regularization of existing immigrants, the lack of internal controls, a large informal economy and the diffculty of expulsion are all argued to incentivize overstaying and irregular entry (Finotelli & Sciortino, 2009; Colombo & Dalla-Zuanna, 2019). It is, however, notable that the majority of migrants gaining residency within a regularization scheme maintain their legal status thereafter, which demonstrates their capacity to retain employment status (Finotelli, 2009). Indeed, the employment rate amongst the foreign born has been higher than the rate amongst the Italian born, as opposed to the situation in Northern Europe (OECD, 2018), arguably demonstrating the effcacy of the system of *de facto* free movement and *ex-post* regularisation. The unemployment rate among foreigners has, however, been higher than the rate among natives since the Great Recession (Perna, 2019).

#### **10.3.1.3 Mobile EU Citizens: Functional Equivalents of Non-EU Labour Immigrants**

During the 2000s, prior to the Great Recession, Eastern European migration to Italy grew signifcantly. The European enlargements to 10 new Member States in Eastern Europe and the Mediterranean in 2004 and to Romania and Bulgaria in 2007 were not, however, equally signifcant with regards to migration to Italy. Transitional arrangements were introduced after both enlargements restricting the access of nationals from these countries (excluding Malta and Cyprus) to the Italian labour market. However, the level of movement from the A8 (eight eastern European accession countries of 2004) was low; indeed, the quota slots available to nationals from these countries between 2004–6 were not all used and the share of A8 nationals in the total foreign resident population increased by just 0.5% during those years (Salis, 2012). On the other hand, Romanian migration to Italy started to grow before the enlargement. In fact, while Romanians migrated in large numbers to Germany in the asylum channel up to the early 1990s, democratization in Romania and a more restrictive asylum policy in Germany channeled Romanian would-be-migrants towards the Italian informal economy (Finotelli, 2009). Very limited restrictions were imposed on their access to the labour market between 2007 and 2012. There were no quotas or restrictions on their employment in highly skilled professional activities or in the main sectors where they were already employed: agriculture, tourism, construction, domestic or care services and the metal industry. By the end of the decade Romanians were the frst foreign community in Italy (Salis, 2012, Fig. 10.1).

**Fig. 10.1** Growth of principal foreign communities in Italy 1991–2014. (Source: Author's elaboration based on ISTAT data on residents by citizenship)

#### *10.3.2 Italian Labour Immigration Regime 2008–2020*

#### **10.3.2.1 Drivers of Policy Change: Economic and Humanitarian Crises and Alternative Sources of Labour**

Unlike in Spain, which had a very similar system to the Italian one until 2004, the legislative framework for non-EU labour immigration to Italy has not changed since 2002 and there has been very little public debate regarding a reform of the system (Salis, 2012; Finotelli & Echeverrıa, 2017). However, the regime has evolved in a restrictive orientation within its current parameters over the past decade, in response to the economic and refugee crises, developments with regards to various migrant infows and European directives.

The Great Recession and subsequent, ongoing economic crisis, which began in 2011, had a signifcant and long-lasting negative impact on the Italian economy. Indeed, between 2008–2017, Italian GDP grew less or declined more than average EU levels (Colombo & Dalla-Zuanna, 2019). Labour demand fell, leading to a rise in migrant unemployment, higher than that among natives (while the overall unemployment rate reached 12% in 2013, the rate was 17.9% for foreigners). Along with the refugee crisis, which started in 2011 and peaked in 2016, the economic crises made a restrictive approach to immigration economically and politically expedient (Einaudi, 2018; Colombo & Dalla-Zuanna, 2019; Perna, 2019).

Furthermore, in 2011 senior offcials from the Ministry of Labour and Social Policies maintained that, based on a more in-depth analysis of labour demand and supply, there would be less need for non-EU labour migrants than before. As a senior offcial maintained in July 2011, "We will no longer need great numbers: family reunifcations, intra-EU mobility and second generations will naturally compensate the labour mismatch that is currently tackled through labour migration from

third countries" (Salis, 2012, p. 34). Indeed, as we saw above, Romanian migration to Italy has been signifcant over the past 20 years. Also, family migration became more substantial in the twenty-frst century, representing the second largest component of stocks and infows in the frst decade of the century and the most signifcant infow since 2011 (Salis, 2012; Geddes & Pettrachin, 2020). In the context of the European refugee crisis, asylum seekers have also represented another functional alternative to labour immigration. In 2007 more than half of new residence permits were issued for working reasons, compared to 32.3% for family reunifcation and 4.3% for humanitarian/asylum protection. By 2017, however, residence permits for working reasons represented only 4.6% of new permits issued, while family reunifcation represented 43.2% and humanitarian/asylum protection 38.4% of new permits (Perna, 2019).

#### **10.3.2.2 Quantitative and Qualitative Changes in Annual Quotas**

Between 2008 and 2009, the quota total dropped from 230,000 to 80,000 slots and the 2009 quota was only for seasonal workers. In 2010, the quota was more generous (184,080 slots), based on a belief that the economy was coming back on track (Pastore, 2016). However, between 2011 and 2020, less than 60,000 places have been available with around 30,000 available since 2015, under technical, centre-left, populist and populist centre-left coalition governments. Since 2009, most entries have been for seasonal migrants (which have come to dominate entries), people with stay permits in Italy who apply to convert them to work permits and applicants for self-employment permits. Indeed, while the level of overall migration to Italy in 2008 (around half a million a year) was comparable with infows into other large EU migrant destinations (Germany, Spain and the UK), between 2008 and 2018 infows declined by nearly 50% into Italy, while they increased substantially into Germany and stayed more or less the same in the UK and Spain (OECD, 2020) Figs. 10.2 and 10.3.

**Fig. 10.2** Non-EU labour migrant quota totals 1996–2020. (Source: Author's elaboration on basis of data from the *Ministero del Lavoro e delle Politiche Sociali*)


**Fig. 10.3** Non-EU labour migrant quota composition 1996–2020. (Source: Author's elaboration based on data from the *Ministero del Lavoro e delle Politiche Sociali,* cross-checked with Callia et al. (2012) and Perna (2019))

#### **10.3.2.3 Regularisations**

Nevertheless, the number of irregular migrants in Italy has remained high, partly due to continuing informal work opportunities in sectors such as domestic work and agriculture and partly as a result of the abolition of humanitarian protection in 2018 (which was subsequently overturned in 2020). Two regularisation programmes were carried out in 2009 and 2012, the frst solely aimed at domestic and care workers (Bonizzoni, 2018; Geddes & Pettrachin, 2020). Italian governments desisted from carrying out regularisations for 8 years, which represents the longest gap between regularisations since the frst one in 1986. However, the number of undocumented migrants was estimated at 533,000 thousand in early 2018 (Colombo & Dalla-Zuanna, 2019; Geddes & Pettrachin, 2020) and in the context of the Covid-19 pandemic, the centre left-populist government agreed to carry out a rather contentious regularization. If an undocumented migrant could prove that s/he was working in agriculture or domestic and care work, they could apply for 1–2 year permit and if they had could prove that they had been working in those sectors recently they could apply for a 6-month permit to look for a job. This was largely motivated by the need to provide workers for the agricultural sector in the context of pandemic related travel restrictions and to control the spread of Covid-19 and, unlike previous regularizations, was not opposed by the EU and Northern European countries (Human Rights Watch, 2020). In sum, over the past decade, regularisations have been fewer and more selective in terms of occupational sector.

#### **10.3.2.4 The Facilitation of Highly Skilled Non-EU Labour Immigrants**

Finally, the admission of highly skilled non-EU individuals for the purposes of self−/employment has been facilitated by the Italian state since 2007 due to European policy transfer and the development of a competition state logic, where states base their economic development strategy on skill-based innovation (Bonizzoni, 2018). However, the number of highly skilled migrants arriving in Italy remains small.

The self-employment channel within the general quota has become more selective with regards to skills and admits the following profles: entrepreneurs intending to invest at least 500,000 euros and expected to hire at least 3 workers; wellrecognized artists with "high professional qualifcations"; corporate offcers of frms registered in Italy; acknowledged professionals; and, since 2013, foreign citizens establishing (or taking part in) "innovative start-ups". Furthermore, like in other EU states, a start-up visa programme was launched by the Ministry of Economic Development and Ministry of Foreign Affairs in 2014, which aims to attract innovative non-EU entrepreneurs to Italy by providing a simplifed online procedure in English (http://italiastartupvisa.mise.gov.it/). However, between 2014 and the start of 2019, only 230 out of 434 applications had been successful and the number of applications fell over the 5 years (Nesheim, 2019).

Highly skilled non-EU labour immigration has also been facilitated following the implementation of four European Directives between 2007 and 2016, which introduced an EU visa for scientifc researchers, intra-company transfers and trainees, as well as highly qualifed workers (the so-called "Blue-Card workers"). Blue Card workers who have completed a post-secondary training course of at least 3 years can work in specifc sectors as long as they are earning a salary of at least 24,789 euros per year (Salis, 2012; Bonizzoni, 2018). Between 2013–5, the number of blue card visas issued nearly trebled but the numbers remained small – 348 in 2015 – and subsequently fell; Italy issued 254 in 2016 (Bonizzoni, 2018; Burmann et al., 2018).

#### **10.4 Differences and Similarities Between the Italian and Northern European Regimes**

With regards to the differences and similarities between the Italian regime, prior to and after 2008, and Northern European systems, it is worth reiterating that there is quite a lot of divergence in immigration systems in the North, for example with regards to the emphasis placed on permanent as opposed to temporary migration and the use of different functional equivalents of labour immigration (e.g. asylum, student, family migration) in different states. However, they share basic labour immigration management tools and strategies in response to demand for foreign labour. Furthermore, they all face the challenge of responding to employers demands for migrant labour, while giving the impression of control over the admission of foreign workers, in the context of high levels of politicisation and public concerns regarding immigration.

#### *10.4.1 Differences*

#### **10.4.1.1 Admission of Low and Medium Skilled Non-EU Migrant Workers**

A signifcant difference between the Italian labour immigration system between the late 1990s and the Great Recession and those in Northern Europe was the emphasis of the former on admitting immigrants for low and medium skilled work and at times providing most of them with a route to permanent residency. Northern European countries also require foreign labour for low and medium skilled work. However, they rely on free movement, non-economic migration channels and temporary labour programmes for non-EU nationals, as, for historic reasons, they have a negative view of permanent low-skilled non-EU labour immigration and they have long received signifcant non-economic infows, which can partially substitute labour immigration.

The Italian system was consequently more explicitly receptive to non-EU workers in low and medium skilled jobs during this period as workers were provided with work permits, many of which provided a route to permanent residency, rather than other types of permits. The higher demand for low-skilled foreign labour in Italy, compared to Northern Europe, in the context of low levels of economic growth and low levels of employment during the period of interest is explained by an ageing population and rising education levels, but also welfare and labour market institutions which have the effect of constraining the labour supply and producing low quality jobs. For example, with regards to the latter, the welfare state provides insuffcient elder care and childcare, resulting in demand for foreign care workers in the informal economy, while low levels of labour market control enable a large informal economy (Devitt, 2018). Despite low levels of employment in Italy, throughout its history of immigration, and higher unemployment amongst the low skilled (EUROSTAT, 2020), there was a pervasive view among policymakers until the Great Recession, at least, that Italians no longer wanted to do the jobs migrants do (Einaudi, 2007; Devitt, 2010; Einaudi, 2018).

#### **10.4.1.2 Infexibility**

Like the Spanish system, the Italian one has often been distinguished from Northern European systems based on its use of quotas. As noted above, while quotas have been used in the North, they generally have not dominated systems. The Italian quota system is very rigid as not only does it put a numerical limit on immigration, but it does not allow applications all year round, as the need arises, and instead specifes a time when applications for the next year must be submitted. This infexibility, while politically (to some extent) and administratively functional, is impractical for employers and immigrants. Moreover, with regards to political benefts, while the Italian quota system provides a clear element of public control in its

annual cap on numbers of "incoming" immigrants, this annual numerical limit makes it easier for the public to gauge the openness of each government to immigration than in Northern European systems, which, without an overall cap and with a plethora of channels, are not so easy to evaluate with regards to openness.

The development of the rigid quota system and its maintenance is in part accounted for by the politicisation of immigration within the fragmented party system in Italy in the 1990s. Notably, the *Lega Nord* populist party associated immigration with irregularity and crime and infuenced centre-right coalition governments, in which it was a member, in the frst decade of the twenty-frst century (Einaudi, 2007; Finotelli & Sciortino, 2009; Salis, 2012). The preferential treatment accorded to domestic and care workers in Italian quota decrees and regularisation processes is not only explained by the fact that there is a signifcant reliance on foreign workers in this sector. These foreign workers are also more socially and politically acceptable due to the fact that many of them are middle-aged women and most of them come from Eastern Europe or Latin America (and are therefore Christians) (Salis, 2012). More recently, opposition to immigration was the main factor explaining the rise of the revitalised *Lega* under Matteo Salvini to become the dominant political party in Italy in 2019 (Geddes & Pettrachin, 2020). Indeed, public opinion is on average more negative about immigration in Italy than in the rest of the EU (Geddes & Pettrachin, 2020), which may be partly due to the infuence of xenophobic rhetoric from politicians and partly due to the fact that infows of foreigners grew signifcantly in the context of economic stagnation from the 1970s onwards (Devitt, 2018; Geddes & Pettrachin, 2020).

#### **10.4.1.3 Implementation Gaps**

The Italian labour immigration policy is seen to have failed by scholars as there is a signifcant gap between law and practice. While it ostensibly aims to manage the infow from abroad of legal migrant workers with employment offers in Italy, the system has largely functioned as an *ex-post* regularisation of migrants, who come to Italy on their own initiative, without a job offer or permit, in a *de facto* free movement regime. While there is commonly a gap between explicit policy goals and outcomes in the area of immigration, the gap between law and practice in the Italian immigration system is particularly large. Most systems in Northern Europe fail to stem irregular migration and many fnd ways of regularising this population, however, the Italian system appears to have largely involved *ex-post* regularisation. Indeed, more mass regularisations of irregular migrant workers have taken place in Southern Europe than in Northern Europe.

The Italian system displays an extremely low level of *ex-ante* management. International recruitment is comparatively insignifcant, also in comparison with Spain, which has attempted to develop this since 2005. The bureaucratic issuing of permits is so delayed that they can only work as a regularisation as opposed to managing urgent business needs. For example, in 2006 an average time of 484 days was needed to remit a work permit. Moreover, as a result of the duration of the

administrative procedures, a signifcant share of the available quota has usually remained unused (Salis, 2012). These delays are partly due to a comparatively ineffcient bureaucracy (Della Sala, 2004; OECD, 2009b). Indeed, weak bureaucracies can have a negative effect on the enforcement of even well-designed labour immigration policies (Pastore et al., 2013; Finotelli & Echeverrıa, 2017).

Furthermore, while governments in Northern Europe turn a blind eye to a certain level of rule circumvention and irregular migration, the level of subterfuge – an open secret amongst Italian offcials and scholars – in the Italian immigration system appears (though in-depth comparative empirical analysis would be required to ascertain the veracity of this) particularly high. There is evidence that sub-quotas for domestic workers and recent regularisations limited to domestic workers – who are the most socially accepted migrant workers – have been used to legalise people outside of domestic work as well. Indeed, migration businesses often provide immigrants with fake contracts so they can apply for a residence permit (Salis, 2012; Finotelli & Echeverrıa, 2017). Moreover, the acceptance of illegal practices and use of amnesties is not exclusive to migration in Italy; the latter are commonly used in other areas such as tax evasion and illegal building (Finotelli, 2009).

#### *10.4.2 Similarities*

#### **10.4.2.1 Entry Mechanisms and Skills**

Italy has long shared labour immigration entry mechanisms with Northern Europe including labour market tests, admission from abroad on the basis of a job offer, preferential access into shortage occupations and on the basis of national origin and the use of regularisations (which can be described as *de facto* regularisations in Northern Europe, such as the *Duldung* in Germany).

Furthermore, while Northern European regimes prioritise highly skilled non-EU immigration, infows of these migrants remain limited, though higher in some cases than levels to Italy. The fact that the quota system is not very targeted with regards to skills does not mean that medium skilled and highly skilled migrants cannot and do not arrive via the quota system. The level of tertiary education amongst the foreign born in Italy is, however, comparatively low (as it is amongst working age Italians) (Eurostat, 2019), which is partly explained by the fact that most of the demand for foreign workers is in low and medium skilled work. It is, however, notable that, while the literature tends to emphasise demand for migrants in low skilled jobs in the Italian labour market, like in Northern Europe (apart from the UK), about half of the foreign-born in Italy are employed in medium skilled jobs (OECD, 2015).

#### **10.4.2.2 Infexibility and Implementation Gaps**

While the Italian quota system is particularly infexible, resulting in signifcant numbers of undocumented migrant workers, this relationship between infexible, restrictive immigration policies and irregularity (i.e. implementation gaps) can, as noted above, be seen across Northern Europe. Indeed, I contend that the difference is one of degrees. Northern European governments' concerns with demonstrating control over immigration often result in immigration policies, which are more restrictive than labour market needs, with the resulting necessity of turning a blind eye to or legalising the status of immigrants who have circumvented the rules in order to provide labour for employers (Cornelius et al., 1994; Joppke, 1998; Duvell, 2006; Messina, 2007). This is exemplifed in Germany, where permanent low skilled labour immigration is strictly curtailed and domestic care work is often carried out by Eastern European women who overstay tourist visas (Finotelli, 2009). Indeed, the quota system was initially established and evolved in response to a burgeoning politicization of immigration and, signifcantly, European pressures to protect the common European border (Finotelli & Sciortino, 2009; Paoli, 2018).

#### **10.4.2.3 Free Movement and Functional Equivalents to Non-EU Labour Migrants**

It could, furthermore, be argued that Italian policy on non-EU labour migration has been rather like a free movement approach where people move on their own initiative without a job offer and then register. Free movement and non-EU immigration channels, such as those for students and family migrants, all provide labour in Northern Europe. Indeed, offshore recruiting for low skilled labour is not just a challenge in Italy. Small businesses with low skilled labour needs, from cafes to building frms, across Europe, would be hard tasked to fnd workers abroad, beyond utilizing employees' networks. Indeed, during the period of mass labour immigration in post-war Northern Europe, when migrant workers were generally in demand in low skilled jobs, while there was some recruitment of foreign workers abroad by public employment services, many migrants came over on tourist visas, found work and then were regularised on case to case basis (OECD, 2009c).

Public concerns regarding immigration and a pervasive belief among policymakers that Europeans are easier to integrate than non-Europeans has led to a preference for intra-EU mobility as a form of labour immigration in most Northern European states following the Eastern enlargements. Intra-EU mobility is preferred by Italian policymakers as well. However, they have only been able to source substantial numbers of mobile European workers, willing to work in low-medium skilled jobs, since the enlargement to Romania (and Bulgaria) in 2007, as levels of mobility to Italy from the A8 was low. The growth in migration from Romania has contributed to allowing the Italian government to reduce infows from outside of the EU after the Great Recession, similarly to Northern European countries, such as Ireland, which curtailed low-skilled non-EU labour immigration in 2003, due to expected infows of mobile Eastern Europeans following the enlargement of 2004 (Devitt, 2016). Indeed, the Italian system has moved from a strong emphasis on permanent low skilled labour immigration to using the functional equivalents of family reunifcation, EU mobility and more recently humanitarian migrants**,** with some facilitation of highly skilled immigration as well. The quotas have become more selective over time and governments demonstrate a preference for seasonal low skilled labour migrants, much like in Northern Europe.

#### **10.5 Conclusion**

Overall, the Italian labour immigration regime in vigour between the late 1990s and 2008 shared much with Northern European systems. The management tools, beyond the use of all-encompassing numerical caps, were the same and the differences in policy and implementation were largely a question of semantics and degrees. The Italian system was more open about admitting permanent low and medium skilled labour immigration from outside the EU, as opposed to Northern European systems which have tended to use free movement and non-economic forms of non-EU migration for labour needs. The Italian system also had a more signifcant gap between its *façade* (laws) and practice than its neighbours in the North. While the Italian system largely involved *ex-post* regularisation, this approach, while used, does not dominate the Northern systems of managing labour immigration due to their longer standing access to functional equivalents to low skilled labour migration, which do not require recruitment from abroad.

The Italian system post 2008 has moved in the direction of the more selective labour immigration systems of Northern Europe, emphasising seasonal and occupational/sectoral permits, along with a stronger reliance on free movement and noneconomic forms of immigration, primarily family and humanitarian migration. This more restrictive approach, in the context of a large informal economy, will continue to result in irregular immigration and the need to intermittently regularise the undocumented population. This situation will remain in the absence of employer sanctions, strong, sustained and costly efforts to tackle the informal economy, more developed public elder and child care services, supports for new business ventures in place of unproftable ones and concerted efforts to recruit overseas in sectors with predictable needs such as agriculture.

Furthermore, the ability of Western European economies to rely on free movement for low skilled labour needs is time limited, due to economic development in Eastern European Member States. Indeed, there are already signs that this source of labour is reducing and European countries are starting to source labour from outside the EU again (OECD, 2020). Italian governments will soon once again fnd themselves opening up to more signifcant levels of non-EU labour immigration. It remains unclear whether policymakers will maintain the longstanding approach of *ex-post* regularisation within the quota system or attempt an institutional reform, but with the current fnancial constraints, a move towards a new, more costly system is unlikely.

#### **References**


 **Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 11 Seasonal Workers in Agriculture: The Cases of Spain and The Netherlands in Times of Covid-19**

**Jeroen Doomernik, Blanca Garcés-Mascareñas, and Berta Güell**

#### **11.1 Introduction**

A key debate in migration studies has been around the conditions that account for the emergence of different immigration and integration policy models as well as the factors that explain recent trends of convergence in discourses, policies and practices. Since the early 2000s, part of this discussion has focused on the South-North divide. As countries such as Greece, Italy and Spain went from being emigration countries to becoming (all of a sudden and intensively) immigration countries, a distinction started to be made between old immigration countries in Northern and Western Europe and new immigration countries in Southern Europe (Bruquetas & Doomernik, 2014). The former had received guestworkers and – in some cases such as the Netherlands and France – migrants from the former colonies in the 1950s and 1960s, and family migrants and refugees from the 1970s onwards. The latter were seen mostly as labour immigration countries, with high labour demands in low productivity sectors eminently covered by "spontaneous migrants" that legalised their situation in one of the numerous regularisation campaigns of the early 2000s. As mentioned in the introduction of the book, over the last 25 years, the image of Southern European countries as weak guardian of borders with precarious admission systems has contributed to forge the "negative exceptionality" of the Southern European model in comparison with the rest of Europe.

J. Doomernik

All fgures from Statistics Netherlands (CBS) StatLine: https://opendata.cbs.nl/#/CBS/nl/

University of Amsterdam, Amsterdam, The Netherlands e-mail: J.M.J.Doomernik@uva.nl

B. Garcés-Mascareñas (\*) · B. Güell CIDOB, Barcelona, Spain e-mail: bgarces@cidob.org; bguell@cidob.org

In the last decade, nonetheless, research on Southern European countries has shown that the reality is much more complex than that. For instance, on the basis of the Italian case, Finotelli and Sciortino (2009) challenged this bipolar perspective and what they called "the Southern regime stereotype" by arguing that the unsatisfactory outcomes of Italian mechanisms of immigration controls are not necessarily the epiphany of a weak policy apparatus but the result from a much more complex mix of factors, including unrealistic policy goals, contradictory international pressures, structural internal constraints and fragile party coalitions. Regarding the Spanish case, Garcés-Mascareñas (2012, 2022) similarly concluded that what was presented as a policy gap (non-working entry policies followed by periodic regularisation campaigns) was instead a policy in itself that made possible to reconcile contradictory imperatives, frst and foremost between a restrictive entry policy designed at the EU level and expansive domestic labour demands. More recently, research on the effects of the Great Recession in immigrants integration in Southern Europe has shown that countries such as Greece, Italy and Spain were indeed vulnerable to economic shocks, which affected negatively immigrants' socioeconomic integration, but somewhat resilient to "integration crises", with no relevant consequences or even positive results in migrants' legal status and public perceptions about immigration and immigrants (e.g. Finotelli & Ponzo, 2018; Ponzo, 2019).

In this chapter, we come back to this debate on the South-North divide by comparing the cases of Spain and the Netherlands with regard to migrant seasonal workers in agriculture and with a particular attention to their situation before and during the Covid-19 pandemic. The relevance of this research is fourfold. First, migrants' work in agriculture – though thoroughly studied – has hardly been included in the debate on the South-North divide. Second, agriculture is an essential economic sector and yet labour conditions tend to be rather poor and precarious. This cannot be disentangled from the processes of agri-food globalisation and productive fexibility, where farmers (especially on the small and medium scales) have little room to manoeuvre when it comes to adjusting prices in line with high production costs. Third, given the seasonal nature of a large part of the jobs associated with agriculture, this sector is the example per excellence of many circular migration programmes (within Europe but also including third countries) implemented at the national level but also promoted directly by the European Commission. Though it remains a national policy, it is one of the most europeanised (at least in discourse and stated goals) in the feld of labour migration. Fourth and fnally, agriculture has been particularly hit by the Covid-19 pandemic: because of the closedown of the hotel/restaurant sector and because of the introduction of mobility restrictions, which directly affected the supply of (mobile) workers.

With regard to the comparison, the cases of Spain and the Netherlands allow us to revisit some of the features attributed to different immigration and integration models. In his seminal article on "early starters and latecomers", Arango (2012) refers to the role of different factors such as the "age effect", meaning the stage of the migration cycle in which each country fnds itself; the "generation effect", that is the historical context in which the initial and formative phases of immigration take place; "historical precedence", which means the infuence that forerunners exert on those that follow; and structural characteristics, with different types of economy and social systems. Coming back to Arango's factors, the Netherlands and Spain seem to differ in everything: while the Netherlands is a "senior" immigration country and therefore a forerunner with a rather formal labour market and (originally at least) robust welfare state, Spain is a "junior" immigration country, thus among the late runners and with a labour market much more dependent on lowskilled and informal labour and with a much limited but more inclusive (particularly for those undocumented) welfare regime. The question is whether, despite all this, the Netherlands and Spain are as different as expected in their response: frst, to the labour needs of seasonal agricultural workers and, second and more specifcally, to the "stress" generated by having to cope with seasonal labour demands (and ensure proper work and housing conditions) in times of Covid-19.

This chapter makes use of the qualitative research conducted in the framework of the EC H2020 project ADMIGOV. Apart from reviewing relevant reports from key stakeholders, newspaper articles, offcial documentation, statistics and academic publications, 30 in-depth interviews with key informants were conducted in Spain (mostly in the regions of Huelva and Lleida) and 13 in the Netherlands. Following the theoretical framework discussed in this book, we understand that "migration regimes" are a mix of factors playing in complex constellation of actors. Therefore, for each case, we will not only pay attention to migration policies but also to a range of factors such as the history of migration, the immigration policy phase, the functioning of the labour market in this specifc economic sector and, fnally, the impact of Covid-19, which we understand as a "stress test" to reveal the inherent and previous contradictions and shortcomings but maybe opportunities as well. Before we delve into each of the two cases, in the next section we will briefy discuss the concept of "migration regime" and explain why, in this case it has to be complemented by what Pastore (2014) has called the "governance of migrant labour supply".

#### **11.2 Migration Regimes in the Light of the Governance of MLS**

If we understand "migration regimes" as a constellation of political principles, norms and practices, thus a world full of gaps, ambiguities and strains resulting from this "work through practices", the next question is to what this work (and the public and private actors involved in it) responds to. To answer this, in this chapter we make use of Pastore's (2014) concept of the governance of Migrant Labour Supply (MLS), which goes beyond migration policies *stricto sensu* and captures the way through which states mobilise alternative tools in order to match and reconcile the different and often conficting sets of interests. According to Pastore, common to this "complex and constantly evolving policy mix" is the state's need to fnd the "paths of least resistance", meaning ways that allow to maximise fulflment of labour immigration demands while minimising resistances to it.

In fnding these "paths of least resistance", Pastore argues that states draw on labour migration policies *stricto sensu* and the management of intra-EU labour mobility. In parallel, beyond migration regulation policies, states may also choose to respond to labour demands in two ways: frst, by giving access to domestic labour markets to immigrants originally admitted for reasons other than work (humanitarian, family, study, etc), what Pastore calls "functional equivalents"; and two, by reducing the dependency on immigrant labour by increasing the presence of native workers in given employment sectors, what he terms "functional alternatives". While Pastore puts the state (or, more broadly, complex and dynamic structures of multilevel governance still centred on states) at the heart of his analysis, basically by recognising its role in determining channels of entry and access to the domestic labour market, this chapter will address the same questions but shifting the focus from the state to the regime, thus from government decisions to actors and practices.

Finally, these questions will be analysed in the context of the Covid-19 pandemic, which in the agricultural sector affected both the demand side (with the closedown of the hotel/restaurant sector) and the supply side (with severe mobility restrictions, particularly in 2020). Again, following Pastore (2014, p. 386), our assumption is that all crises can be analysed as a "revealer" or "lens" to understand reality, in this case the working of the above mentioned "policy mix" in the specifc sector of seasonal agricultural work. What did the constraints imposed by the pandemic reveal us regarding these different policy options and contradictions? Given similar constraints, were responses in Spain and the Netherlands similar? What do these similarities and/or differences tell us about the governance of MLS in both countries? Though it may be too early to respond, were changes induced by the Covid-19 pandemic superfcial and transient or did they come to remain? Did they reinforce former trends or open unexpected new arrangements?

#### **11.3 Spain**

#### *11.3.1 Becoming an Immigration Country*

In the mid-1980s, Spain unexpectedly went from being a country of emigration to having a positive balance in migratory fow (Izquierdo, 1996). The main factor triggering immigration to Spain was the growing imbalance between an autochthonous labour force that had slowly been raising its "job acceptability level" and the persistent demand for workers in low productivity sectors with high levels of informal economic activities requiring a cheap and fexible labour force. After the 1990s, this imbalance on the labour market was further aggravated by two other factors: economic growth from 1995 to 2001 with an increased demand for labour involving almost 670,000 new jobs per year; and the decline by approximately two million people of the new native workers entering into the job market after 1992 due to the demographic decline of the cohorts of those born after 1976 (Garcés-Mascareñas, 2012, p. 112).

As a result of these transformations, the number of legal foreign residents in Spain went from 250,000 (0.75% of the total population) in 1985 to almost 900,000 (2.18%) in 2000, 1.9 million (4.48%) in 2004, three million (6.7%) in 2006, 3.9 million (almost 10%) in 2007 and 5.5 million (12%) in 2008. This phase of growth was interrupted in 2008 due to the economic crisis, which led to rising unemployment reaching 25% in 2012. As a consequence, net migration became negative in 2011 and so did population growth from 2012 onwards, with Spain losing more than 200,000 inhabitants in 2013 (Oliveau et al., 2019). This trend changed in 2015, when immigration started to increase again. As of 2020, there were 7.2 million foreign-born residents (more than 15% of the total population), including fve million (10.57%) born in a non-European country and 5.4 million (11.45%) without Spanish citizenship. In 2019 the main countries of origin were Morocco, Rumania, UK, Italy, Colombia, China, Venezuela, Germany, Ecuador and Bulgaria (INE, 2021).

#### *11.3.2 Labour Migration Policies*

Though Spanish economic growth depended on migrant labour, entry policies were rather restrictive: the frst Foreigners Law (1985) was promulgated to block entry to immigrants en route to Western European countries via Spain and, even more important, entry policies were restrictive in actual practice, with long and very complex procedures. This restrictive entry policy did not mean, as we have seen, that low numbers of immigrants were entering Spain. Most arrived with a tourist visa, found work and subsequently legalised their situation. This mismatch between legality and reality – between a particularly restrictive policy and a reality notable for large numbers of people entering the country – made it possible to comply with contradictory demands: on the one hand, demands for closure by the EU but also by the trade unions who did not look kindly on the entry of new workers into a job market characterised by high unemployment fgures; and on the other, demands for openness by employers but also by an increasing middle-class who rapidly became dependent on female migrants care work.

While regularisations have frequently been interpreted as the best illustration of the "failure" of immigration policies and, more generally, the state's loss of control, regularisations in the Spanish case should be understood primarily as a de facto entry policy. Basically, because the end result was deferred "entry" – deferred since the condition for every regularisation is a period of illegal status – of however many immigrant workers were required by the employers. As González-Enríquez (2009) noted, this is nothing more than a cheap recruitment model in the place of destination. Cheap because the costs and risks of the migratory process were shouldered by the immigrant and because in political terms it was possible to have a high-numbers policy without putting it in writing and thus without needing to justify it.

If we look at recent years, it is unavoidable to mention the increase in the number of asylum applications: from 2588 requests in 2012 and 5947 in 2014 to 14,881 in 2015, 31,120 in 2017, 55,668 in 2018 and 118,264 in 2019 (CEAR, 2020). After being at the bottom of the list of numbers of asylum seekers in EU Member States, Spain has now become one of the frst destination countries. Beyond the global context, two major changes in Spain explain this shift: frst, with the new Asylum Law of 2009 the number of inadmissibility applications decreased extraordinarily; second, the job offer, which was the key entry door through regularisation before the economic crisis, became harder to get. Those who previously entered the country via the job offer from 2009 onwards had the option of doing so by means of requesting asylum. However, as most asylum seekers (between 60 and 80% depending on the year) do not receive any kind of protection, in the mid- and long-term their chances to be recognised as legal residents depend again on regularisation, thus again on the availability of the job offer.

#### *11.3.3 Agricultural Seasonal Work: Past and Present*

Spain is the frst European Union's market garden, representing 23% of the total production. Many of the tomatoes, peppers, cucumbers, and also strawberries and other red fruits consumed in the EU come from Spain. Unsurprisingly, then, Spain is the EU's leading exporter of fruits and vegetables and one of the world's top three, alongside China and the United States. Fruit and vegetable production represents a quarter of all the value produced by the agricultural sector. Spain is also the largest producer of olive oil, accounting for 50% of world production.

As the agricultural sector (especially in its exports dimension) grew, it became more and more diffcult to fnd seasonal workers, also due to the increase in the job acceptability level of the autochthonous labour force. As a result, the demand for seasonal agricultural workers was mostly covered by migrants. The frst arrived in Spain in the 1980s, especially from Morocco and West Africa (Senegal and Gambia). At the end of the 1990s, programmes of recruitment in origin began, seeking labourers in Morocco and, subsequently, in countries of Eastern Europe (like Poland, Rumania, and Bulgaria) and in Latin America, as well. Nevertheless, in the 1990s, recruitment in the agricultural sector remained mainly at destination, with a considerable amount of contracting that was irregular or formalized *ad hoc* by means of the many processes of regularization. It was not until the 2000s that recruitment programmes in origin began to mobilize a certain volume of workers.

Indeed, after few pilots, the frst circular migration programmes (with a national framework but managed in practice by each Autonomous Community) were launched in 2001. The leading regions were Huelva and Lleida, which is no coincidence as their crops (above all red and seed fruits respectively) are highly seasonal, meaning that employers need high numbers of workers for limited periods. This programme (called *Gestión Colectiva de Contrataciones en Origen*, GECCO) functioned thus as a way to guarantee the availability of workers from a particular origin and in a particular moment in time. In the case of Lleida, the whole system was led by the *Fundación Pagesos Solidaris*, which centralised the recruitment process and even reception and training at destination. In the campaign of 2007–2008, which is when the highest number of workers came through GECCO, employers in Huelva recruited about 40,000 workers and in Lleida about 7000 (Díaz et al., 2014).

It is important to note that employers were key in the well-functioning of these programmes. Not only because they led part of the process but also because without their collaboration these programmes amounted to virtually nothing. This was clear with the bilateral agreements signed with Gambia (2006), Guinea Conakry (2006), Cape Verde (2006), Mali (2006) and Senegal (2007). In exchange for these countries' collaboration in the struggle against illegal immigration and, above all, with the condition that they agree to the repatriation of their citizens, the Spanish government promised to facilitate legal immigration from these countries. However, employers were not so keen to comply. They argued that they preferred workers from Latin America. This meant that legal ways of entering remained barred in practice.

The economic crisis of 2008 represented a turning point. With unemployment rising quickly, the Spanish government decided to freeze the programme in order to promote employment of those workers (national or not) already in the country (López-Sala, 2016). It was a freeze rather than a ban, as workers (particularly those repeating and working in the felds rather than in the packaging) continued to come. According to Gualda (2012, p. 635), this measure sought to keep the "machine greased" so as to be able to reactivate it in case labour demands increased or labour supply (of those already in the country) turned not to be enough. Interestingly, the economic crisis did not reduce jobs in the agricultural sector and migrant workers continued to be dominant (slightly more than 50% in Huelva and between 80–90% in Lleida).

Following Molinero Gerbeau (2018), the dominance of migrant workers in agriculture, even after the freeze of GECCO, should be explained by two parallel processes. First, as shown by previous research, migrant workers – already resident in Spain and who the economic crisis left unemployed – turned (back in many cases) to agriculture (López-Sala, 2013; Gadea et al., 2015). In contrast to national workers, who are mostly reluctant to work in this sector even when they are unemployed, they probably did not have many other alternatives (e.g., support from the family). Second, employers continued to recruit in the countries of origin but without going through the GECCO programme. Two factors facilitated this "individual", "spontaneous" or "private" (meaning not state driven) recruitment in origin: Eastern European countries (of special importance here, Rumania and Bulgaria) had entered the EU, thus recruitment in these countries was possible outside GECCO and without having to go through the national employment agency; and also in these countries previous networks of former workers and their respective friends, family and acquaintances fed the system with concrete workers to be employed every year.

Molinero Gerbeau (*ibid*) wonders whether the end of the 2008 economic crisis led these migrant workers residing in Spain back to other (better paid) economic sectors and therefore meant the reactivation of the GECCO programme or the rise of this "individual", "spontaneous" or "private" recruitment system in Eastern Europe. The answer is probably both. Both the GECCO programme and recruitment of Romanian and Bulgarian workers in their countries of origin raised considerably in the last years. The next question, addressed as well in this chapter, is what happened in 2020 in a context of pandemic and therefore of high mobility restrictions.

In terms of labour conditions, foreign workers recruited in origin countries through the GECCO programme have a full-time job during their whole stay in Spain. A clause in their contract includes some exceptions (e.g., bad weather) that allow to cancel working days up to 25%. In practice, this means that seasonal workers under GECCO come to Spain assuming that they will earn a salary corresponding to at least 75% of the whole duration of the contract. Workers recruited directly by employers (either in origin or destination) have their labour conditions regulated by the collective bargaining agreements, which may differ from region to region. In general, these collective agreements set up around 40 working hours per week and a maximum number of extra hours. In 2020, with the increase of the minimum wage (*Salario Mínimo Interprofesional*, SMI), salaries in agriculture raised up to 7.41 euros/hour.

With the 2008 economic crisis and the 2012 labour reform, temping agencies became key in the recruitment of agricultural workers. According to our interviewees, this was due to their capacity to respond to several employers' needs: frst, by funding campaigns, which became more diffcult after 2008 as banks put more restrictions to fnancially support employers; second, by complying with all the paperwork and payments to the Social Security and the Tax Offce; third, by selecting workers and thus facilitating the whole recruitment process; and fourth, by introducing more fexibility according to the sudden and changing production needs. Not all are advantages though. Temping agencies have also been highly criticised, specially by trade unions and social organisations that argued that some temping agencies do not respect the collective bargaining agreements neither in terms of salaries nor regarding workers' accommodation. Moreover, it is also claimed that the subcontracting of workers may foster employers' lack of involvement in the wellbeing of workers and that the lack of coordination between temping agencies and employers may end up leaving workers unprotected.

#### *11.3.4 Working in Times of Covid-19*

In the early months of the Covid-19 pandemic employers' associations in Lleida announced in the press that they were in need of around 35,000 workers and that they feared not being able to cover these demands due to mobility restrictions. In parallel, partly as a result of the announcement of a regularisation campaign in Italy and the parliament petition led by many immigrant and activist groups for a major regularisation in Spain (the so-called #regularizaciónYA! campaign), rumours started to circulate about a possible regularisation through a job contract in

agriculture. All these led many people from different parts of Spain and even from outside the country to move to Lleida: some were nationals or legal residents who had been expelled from other economic sectors, mostly from the catering and hospitality sector which were severely affected by the lockdown; others were immigrants working in the informal economy (e.g. street vending), who saw a chance to work and regularise their situation. The early arrival of so many people in Lleida created important problems in terms of accommodation and many of them were not fnally employed.

Based on our interviews, labour demands during the frst year of the Covid-19 pandemic were covered in different ways. First, many foreign workers from abroad (though not all of those expected) did come anyway. This was the case of 7000 female Moroccan workers who arrived under GECCO in the early months of the campaign (between December 2019 and March 2020), thus just before mobility restrictions were introduced. While 7000 arrived, another 10,000 expected female Moroccan workers did not. In Lleida, employers' associations put pressure on the government to organise a cordon sanitaire to facilitate the arrival of seasonal workers from Eastern Europe (mainly Rumania). While their arrival by plane was denied, it was fnally allowed by land. According to the sub delegation government's offce in Lleida, half of the expected Romanian workers (around 2500 out of the usual 5000 or 6000) did fnally manage to come. Employers' association organised several buses, for instance AFRUCAT brought in this way around 600 people.

In 2020 seasonal labour demands in agriculture were also covered by facilitating the mobility of workers within Spain despite severe mobility restrictions. Again, due to the pressure put by employers' organisations, the government issued a "certifcate of mobility" to allow individuals with a work contract (previously sent by email) to move across provinces. In this way, for instance, ASAJA was able to employ 1660 temporary workers that usually concatenate campaigns across Spain. The state offered as well the possibility to transfer Moroccan female workers (those that had arrived before the lockdown) from Huelva to other regions, eminently Lleida. The argument was that it was a win-win solution as in this way new labour demands would be fulflled while Moroccan female workers already in Spain (and who could not return to Morocco due to the closure of borders) could be employed. In practice, however, most employers rejected this possibility by arguing that they needed "their" workforce (with specifc knowledge and skills) from previous years (especially from Romania). Another argument was that they were obliged to employ previous workers frst as they were holders of a permanent seasonal employment contract.

In general terms, employers were rather successful in ensuring the mobility of those workers (either from Eastern Europe or the rest of Spain) required by the sector. Their argument, almost a slogan, was that "the fruit needed to be harvested at any price". The priority given to employers' demands posed tensions not only between employers and public health representatives but also, in a later stage, between employers in agriculture and those in other sectors that were substantially affected by the severe lockdowns following the frst outbreaks of the summer, particularly in Lleida. As one of our interviewees stated: "This lack of control of how they came, how they were and how they circulated later is one of the reasons for the outbreak of the summer in Lleida. (…) The fruit had to be harvested at any cost. (…) I think that the right to public health has been violated. And here, the administrations were all warned" (ST-LL8a).

In the initial scenario of perceived labour scarcity of early 2020, the Spanish government approved a decree-law (*Real Decreto* 13/2020) to attract other potential groups to work in agriculture. These included: young migrant aged 18–21 that used to be under protection as unaccompanied minors and who did not have a work contract; migrant workers whose residence permit run out of period between March and June 2020; asylum seekers; and unemployed people that with this decree could combine unemployment benefts with a job contract in agriculture. This strategy, however, did not work very well. In the case of young migrant, there were several documentary restrictions that made it diffcult in practice. For national workers coming from other economic sectors, work and living conditions in agriculture led many of them to drop out even before or immediately after taking the job.

Finally, particularly during the peaks of the campaign, some employers also drew on irregular workers, mostly males from Maghreb and Sub-Saharan countries. Although employers and public administrations systematically deny the employment of irregular workers, there is an implicit consensus that informal arrangements may happen sometimes. One of our interviewees stated it as follows: "In general, we can say that undocumented people do not work. It doesn't mean that a man doesn't tell someone, 'listen, brush it off and I'll give you 25 or 40 euros'. I'm not telling you this doesn't happen" (ST-LL2).

#### **11.4 The Netherlands**

#### *11.4.1 Three Origins of Immigration*

As a result of post-World-War II immigration, the Netherlands has become an ethnically highly diverse society. Immigration had broadly three origins: (post) colonial, labour and asylum. Indonesia was a Dutch colony until 1949 and Surinam a part of the Dutch Kingdom until 1975. These countries' independence caused many of its inhabitants to resettle in the Netherlands. On January frst, 2020, 356,000 residents were of Indonesian descent and the same number of Surinamese origin (defned as being born in that country of origin or having at least one parent for whom this was the case).1 The Dutch Antillean islands and Aruba are still part of the Kingdom. The number of people whose roots lie in these islands stood at 166,000.

From the 1960s onward migrants arrived from Turkey and Morocco. Initially these came as "guest workers" to fulfl labour demand few Dutch workers were

<sup>1</sup> https://www.cbs.nl/nl-nl/nieuws/2020/19/landbouw-droeg-in-2019-evenveel-bij-aaneconomie-als-tien-jaar-eerder

interested in and in industries that were not viable without cheap and undemanding personnel. Even though formally a regime of labour and residence permits applied, implementation was pragmatic: spontaneous migrants arriving as tourists could easily regularize upon fnding employment. When the 1973-oil crises sparked a large recession, these industries could no longer survive, at least not based in the Netherlands, which made many of these "guest workers" redundant. Though both these migrants and the Dutch government may have believed that their presence was temporary, the fact was that many of them stayed and brought over their spouses and children. In 1973 Moroccan and Turkish nationals accounted for 28 and 46 thousand inhabitants respectively (Penninx et al., 1994: 12, Table 1.1). Presently, the Netherlands is home to 409,000 people of Moroccan and 417,000 of Turkish origin. The resulting communities have been the subject of the Dutch government's integration policies as defciencies have long been and still are in evidence in terms of educational attainments, labour market participation and earnings, and housing. The government, additionally, has pursued restrictive immigration policies towards labour migrants since 1973. From the early 2000's it also put in place measures aiming to reduce family-based migration from less developed nations (including Morocco and Turkey) (Doomernik, 2017).

From the mid-1980s refugee migration to the Netherlands gained considerably in importance. Notably the end of the Cold War caused considerable displacement within Europe. The Netherlands saw the arrival of asylum seekers from war-torn former Yugoslavia as well as from Iran, Somalia, Eritrea, Afghanistan, Iraq and Syria. People from these countries of origin taken together number 330,000. In political debates, refugees are considered a burden rather than an addition to the (migrant) labour supply. In this regard, a government sponsored cohort survey of Syrian refugees from 2014 to the present shows their welfare dependency to be very high (70% in 2018) and their employment to be precarious (Dagevos et al., 2020).

#### *11.4.2 Labour Immigration Policies*

In 2013 the Dutch government introduced its Law on a Modern Migration Policy with the aim to simplify rules for regular (i.e., non-asylum based) admissions and it also was a paradigmatic change. Whereas earlier labour migration had been considered as something that was anomalous and only exceptionally allowed, it now had been made part and parcel of this Modern Policy. The aim, however, was not to return to the low- and unskilled labour immigration from before 1973. Instead, the policy is geared towards highly skilled (college level) workers and entrepreneurs. The so-called Knowledge Migrant scheme allows recently graduated persons to remain in the country for a year whilst looking for employment, including the option to start a company. This option is also available to someone who has graduated from a selected foreign university (e.g. one of the top 200 as published by the Times Higher Education World University Rankings).

Mid-skilled and lower or unskilled workers had meanwhile freely become available from the Central and Eastern European countries that had joined the European Union in 2004. Its citizens did not immediately have the freedom to work in the Netherlands (this had to wait until January 2007, and for Romanian and Bulgarian workers even 7 years longer, i.e. until 2014). In this respect the Dutch government was more hesitant than those of Sweden, Ireland and the United Kingdom, which had immediately and unreservedly opened up their labour markets for workers from these new Member States. As Kremer (2016) observes, the Netherlands was suffering from a "guest worker" trauma whereas at the same time the demand on the Dutch labour market was no longer for the type of workers who were recruited in the 1960s. Presently the largest numbers of recent labour migrants in the Netherlands stem from Central and Eastern Europe. In 2020 close to 200,000 Poles were registered as well as 40 thousand Bulgarians and about the same number of Romanians (39,000). Next to those there are many more workers from these countries who do not register with a municipality because they do not experience the need to do so or because there is no obligation. The latter is the case when their stay does not exceed 4 months, which typically applies to seasonal workers.

Meanwhile, labour immigration from third countries for other than highly skilled workers remains severely restricted. All such admissions are regulated by the *Wet Arbeid Vreemdelingen* (WAV) (Law on Aliens' Employment). The general rule is that an admission serves the needs of an employer (i.e. is demand driven). Admissions from abroad must fll a vacancy which cannot otherwise be fulflled from the labour force already present within the European Economic Area (EEA). Depending on the precise nature, employers may or may not be required to demonstrate they have undertaken recruitment efforts within the EEA. Normally the permit to work and reside is valid for an initial 1 year period. After 5 years, the migrant is completely free to remain and take on any job. Until then the permit limits them to a particular type of employment and employer.

#### *11.4.3 Seasonal Work: Past and Present*

The importance of seasonal work for the Dutch economy is diffcult to gauge. However, statistics show the overall importance of the agricultural sector to be modest and declining. In 1995 it contributed 2.8% to the national GDP. In 2019 it had dropped to 1.4% (Afrian et al., 2020). Bakker et al. (2004) surveyed seasonal work in the Westland, an expansive area of green houses near the city of The Hague. They found that the "guest workers" which were attracted to Dutch industries became important for Dutch horticulture as well. For decades Turks and Moroccans were the most signifcant category of non-native workers. They also facilitated further network-based irregular labour migration of younger cohorts into the Westland. Staring (2001) shows how Turkish "tourists" as they are referred to by their established compatriots, are incorporated in an extensive and largely closed ethnic

network which provides access to work, for instance in the Westland, by Turkish intermediaries.

In the 1980s Poland became a signifcant additional source of workers for the Westland's horticulture. Among them were German Poles (people living in a part of Poland which was part of the German State before 1945 and in the possession of both a Polish and a German passport), irregulars and workers with a work permit (Bakker et al., 2004, p. 92). It was well known throughout the 1990s that the Westland employed large numbers of irregular workers through the mediation of abusive agents (Siegmann & Williams, 2020). In response a coalition of labour inspectorate, tax inspectors, alien's police and others joined forces in the Westland Intervention Team (WIT) which was created in 1999. This was also inspired by the de-regularization of the temp agency branch in 1998 which took away virtually all thresholds for such agents. The employers interviewed by Bakker et al. (2004) put the blame for abusive and underpaid irregular employment on these agents but admitted that sometimes peaks in demand were such that they could see no alternative, even at the risk of serious fnes (*ibidem*, pp. 94–07).

The WIT was disbanded in 2012 because irregular employment ceased to be a serious problem, not least because Polish workers, using their freedom to move as EU-citizens, have become dominant among the seasonal workers in the Netherlands. According to Statistics Netherlands (data for 2017) there are 178,600 Polish, 23,400 Romanians and 12,100 Bulgarians employed in the Netherlands.

In terms of current working conditions, the fact that irregular employment has lost its salience does not mean abusive employment relationships have also disappeared. The liberalization of the temporary employment market of 1998 appears to have created much scope for abusive practices. In 2020, the Dutch government asked an investigative committee – the so called *Aanjaagteam* (loosely translated as boost team) – to critically evaluate the employment conditions of Central and Eastern European workers in the Netherlands. Its fndings cover all sectors of employment, many types of which are more or less permanent instead of seasonal in nature.

In the Netherlands most temporary migrant workers are hired through a temp agency (of which there are about 14,000). Legally speaking these workers are this agency's employees who are seconded to the actual employer. This makes the actual employer often oblivious of the living conditions of their workers and unaware of perhaps long travel distances between the accommodation and the place of work. Normal practice is that hirings are on a so-called "zero-hour contract". Even though such contracts do create an employment relationship, they do not guarantee actual work. This means workers are entirely reliant on the demand for their labour in order to receive a wage. Obviously, this means that in case of any disruption in a business operation the worker immediately loses their income. Once temp workers have been employed for a certain amount of time, their position should become more secure and rights are accrued. In order to avoid this from happening, the worker is re-hired through another agency which legally brings them back to square one. The worker remains in phase A, which is the term used for a contract in which one has no income when sick or in the absence of work (Aanjaagteam, 2020, p. 21).

A second issue that emerged from the committee's research was that temporary employment can become a business in and of itself. The agency is obliged to offer housing to migrant workers and is permitted to deduct the expenses for this service from the wages for up to 25%. If sub-standard conditions are offered, which frequently appears to be the case, the agency makes a proft. This is also the case if the employer charges the costs of health insurance for the worker while not having actually paid for it. Also reported are instances of unpaid over-hours and underpayment of workers who do not understand the contract they have signed.

#### *11.4.4 Working in Times of Covid-19*

Unlike some neighbouring countries, the Netherlands has not imposed restrictions on the arrival of seasonal workers who were able to arrive. It seems shortages in labour supply did not take on serious forms and, in fact, many Poles and Rumanians were already present before the pandemic hit in March 2020.

However, some abuses were reported. The daily *De Volkskrant* (Dirks, 2020) reported how the local government of the city of Rotterdam (located close to the *Westland*, as mentioned a region with a high concentration of greenhouses) assisted Polish workers and their families who became homeless because of lacking income and abusively high rents. It was also reported that many migrants had never registered with the municipality. The informal nature of a rental agreement may be one of the reasons why people are not recorded in the population register. The same *Volkskrant* report mentioned working conditions in the greenhouses to be in violation of Covid-19 rules: maintaining a minimum distance of 1.5 meters between workers is largely being ignored. Reports about the working conditions in slaughterhouses, meatpacking and distribution centers showed cramped working conditions resulting in high infection rates. Lack of adequate housing conditions appeared to be one of the biggest problems.

#### **11.5 Conclusion**

The cases of Spain and the Netherlands seem to represent completely different models of immigration. Following the features identifed by Arango (2012) in his article on "early starters and latecomers", the two countries are in different "stages of the migration cycle", being Spain in a much earlier phase with most of their immigrant population still without access to Spanish citizenship; present different "generation effects" and "historical precedence", having Spain a clearly labour driven immigration and the Netherlands an initial guestworker phase combined with postcolonial migration and later on with family migration and refugees; and show different structural characteristics, having Spain a persistent demand for workers in low productivity sectors with high levels of informality and the Netherlands an immigration policy exclusively geared towards highly skilled workers and entrepreneurs.

However, if we focus on seasonal workers in agriculture, differences are not as relevant as expected. In both cases, seasonal labour demands were covered initially by recently arrived immigrants, in a later stage by immigrants (and their acquaintances) already in the country and in the last years also by Eastern European workers that can go back and forth without the constraints imposed by international borders. Interestingly, the Spanish state – together with employers' organisations – was much more active in organising recruitment in origin (through the so-called GECCO programme). However, since the 2007 economic crisis this has become a very limited option in terms of numbers. The privatisation of recruitment, by employing workers directly either *in situ* or abroad (mostly in Eastern European countries), seems to be the rule both in Spain and the Netherlands. The use of temp agencies (and the de-regulation that follows) is also a common feature of both cases, though in the Netherlands it seems to be a much more generalised practice. De-regulation (and therefore deprotection) seems also to go a step further in the case of the Netherlands (at least on paper) with the "zero-hour contract", which means that business disruption lies entirely on the shoulders of workers.

Despite these minor differences, in both cases we see convergence towards a decreasing public intermediation at the admission stage (which is facilitated by free movement within the EU) and a growing private intermediation through temporary agencies. Both these trends could be framed as fallouts of a liberalising season in the political and economic history of Europe, which has the EU as framework and grandmaster.

Coming back to Pastore's concept of governance of Migrant Labour Supply (MLS), it is not clear whether the way seasonal labour demands in agriculture have been covered in practice in Spain and the Netherlands responds to the mobilisation of alternative tools by the state or rather by employers themselves in different given policy, economic and social contexts. This is not just a nuance. As explained, in the early 2000s the Spanish state organised recruitment in origin in several African countries. While this responded to the state's need to open legal migration channels in exchange to migration control and readmission agreements, in practice these programmes did not work as employers' preferences (in terms of migrants' origin and recruitment procedures) did not match. In the Netherlands, this mismatch between employers' and state's interests is illustrated by the fact that the Dutch government decided to postpone opening the labour market to Eastern European workers despite being these countries a possible key source of seasonal workers.

In addition to this mismatch between employers' and state's interests, the governance of MLS is also shaped in practice by, on the one hand, the global agricultural chains, where farmers (especially on the small and medium scales) have little room to manoeuvre when it comes to adjusting prices in line with high production costs; and, on the other hand, the labour preferences of the autochthonous workers, increasingly away from the harsh working conditions in agriculture. The way for employers to reconcile these conficting demands is by turning to the cheapest and most vulnerable labour force: either those recently arrived in the country and whose desirability level has not reached (yet) that of national workers or those go move back and forth from origin and destination countries either across borders (and there the state has more to say) or without borders within the EU (and thus with a more direct role by employers). In this regard, we can conclude that this seems to be the "paths of least resistance" in both Spain and the Netherlands.

Finally, this chapter aimed as well to consider whether the Covid-19 pandemic, which in agriculture affected both the demand and supply sides, induced structural changes in the governance of MLS. The answer is probably no. In Spain structural changes had already been introduced with the economic crisis of 2008. As shown by Molinero Gerbeau (2018), since then labour demands have mostly been covered by migrant workers already resident in Spain (and who the economic crisis left unemployed) and by Eastern European workers recruited outside the GECCO programme, thus through more private (meaning non state driven) channels. In this regard, the Covid-19 pandemic seems to have intensifed a tendency that was already there. In the case of the Netherlands, shortages in labour supply did not take on serious forms, as many Poles and Rumanians were already in the country before the pandemic hit in March 2020 and afterwards the government did not impose restrictions on their arrival. What seems to have changed due to the Covid-19 pandemic is the public awareness on the working and living conditions of agricultural seasonal workers. Indeed, in both cases, the media covered several cases of abuses and there were intensive political and public debates. The question, still to be seen, is whether these debates will lead to more protection or will only result on few superfcial changes particularly oriented to specifc and temporary anti Covid-19 measures. The answer, again, will have to do with these different "paths of least resistance", defned not only by the state but also by the different actors involved, i.e., workers, trade unions, activists and above all by employers themselves.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Part V Intra-EU Mobility**

# **Chapter 12 Migration Policy and Welfare Chauvinism in the United Kingdom: European Divergence or Trend-Setting?**

**Alessio D'Angelo**

#### **12.1 Introduction: The European Way to Brexit**

The Brexit referendum of 2016 marked the departure of the United Kingdom from the European Union, and so from its mechanisms of free movement. A shock to the rest of the continent, this seismic change owes much to the growing hostility towards European migration following the EU enlargement of 2004 and, later, to the increase in South-North fows due to the global fnancial crisis of 2008. For a long time at the periphery of intra-European migration trends, at the start of the twenty-frst century the UK had become one of the main nodes of this regional system – a role which the country decided to repudiate. Such hostility was not framed so much in terms of generic xenophobia but, rather, as welfare chauvinism and, in turn, connected to a rejection of EU migrants – and migrants in general – as fellow citizens with full welfare entitlements (D'Angelo & Kofman, 2018). This chapter examines the last few decades of political and policy developments around migration and intra-European migration in the United Kingdom, the key trends that led to the (not so) unpredictable Brexit referendum, and the scenarios which have been set in motion with the UK-EU Agreement of 2020, following 4 years of wearying negotiations. In doing this, it contributes to a (re)assessment of Britain's recent history and the extent to which this has been diverging from – or indeed infuential for – wider European processes.

As will be argued, despite the strong sense of British exceptionalism which has informed UK discussions, some of the fundamentals underpinning the reshaping of its policies have much in common with what is happening elsewhere in the continent, with the stratifcation of welfare rights for different categories of migrants being used as a mechanism to regulate entry and settlement. Thus, what at political and institutional level appears yet as the major rupture within the European

University of Nottingham, Nottingham, UK e-mail: Alessio.Dangelo@nottingham.ac.uk

229

A. D'Angelo (\*)

<sup>©</sup> The Author(s) 2023

C. Finotelli, I. Ponzo (eds.), *Migration Control Logics and Strategies in Europe*, IMISCOE Research Series, https://doi.org/10.1007/978-3-031-26002-5\_12

framework, may reveal itself as part of a wider trend among both Northern and Southern European countries: the reconfguration of the welfare-migration nexus in even more restrictive terms. Chauvinistic welfare policies "operate as bordering practices", with welfare provisions increasingly being "withdrawn from a group of people designated as undeserving" (Guentner et al., 2016:391). Within this race to the welfare bottom, we are witnessing not just a reduction of the entitlements for new migrants, but also an erosion of the rights of established foreign-born residents – both third-country nationals and EU citizens (D'Angelo & Gargiulo, 2021) – and a further racialisation of ideas of citizenship.

At the same time, the recent British history can be interpreted as the attempt of some political forces to disentangle the UK from what was perceived (or depicted) as an increasingly unmanageable EU migration regime. Here the term "regime" is used in the sense of "complex and multi-layered political regulations of migration that escape realist defnition of the state as an acting entity" (Cvajner et al., 2018). The combined efforts of chauvinism and Euroscepticism have strived to reject the very idea of such a diffuse regulatory complexity, and to reassert the primacy of the national government on both external and internal forces. The British doctrine of a "hostile environment" for migrants (Goodfellow, 2020) has entailed a top-down, far-reaching strategy which required a wide range of public and private sector actors to fall in line with the implementation of rigid practices of "internal bordering" (Yuval-Davis et al., 2018). Also in this case, however, the UK approach can be shown to be more quintessentially European than its own Governments would like to believe.

Bringing together these different perspectives, this chapter provides an original contribution to (re)assess not just Brexit, but the wider role of the UK in shaping the welfare-migration nexus across the European Union. This is achieved by critically examining the policy developments, political events and public debates from postwar Britain to the Boris Johnson Government of 2019–2022, drawing on a wide range of secondary sources and scholarly analyses. The following sections start by identifying welfare chauvinism as one of the founding principles of the post-war migration regime in the UK. This went hand in hand with ideas of marketisation and stratifcation of the labour market – both internally and internationally – which would be further developed by New Labour. The chapter then examines the emergence of the Austerity Agenda, in parallel with welfare restrictions for new EU arrivals and the consolidation of the "Hostile Environment" model in the early 2010s. Though also intended to pander Euroscepticism and the moral panic against European migrants, these developments had the effect of paving the way for the Brexit referendum which, in turn, created new possibilities for exclusionary welfare policies. In these respects, the role of the UK within the EU emerges as a crucial case study: not so much one of divergence, but of trend-setting; coherently with an ideological alignment and infuence (Antonucci & Varriale, 2020) which, though often forgotten, dates back to the 1970s.

#### **12.2 Post-war Britain: Between Welfare Chauvinism and European Marketisation**

Post-World War II economic migration to Britain was characterised mostly by arrivals from its former colonies, particularly the Caribbean, South Asia and Ireland (D'Angelo & Kofman, 2018). Conversely, the role of immigration from the rest of Europe remained quite marginal for several decades, both numerically and in public discourses. In spite of some sizeable components, such as the Italian workers who settled since the 1950s (D'Angelo, 2007), the UK remained pretty much outside of the South-North intra-European migration system. This pattern had been actively encouraged by the British Nationality Act of 1948, which gave the right of settlement to anyone born in a British colony. The idea of the Commonwealth as a space of relative free movement and diffuse rights, however, was bound to be short-lived. Since the 1960s, the country saw the progressive introduction of immigration and nationality legislation aiming to restrict the eligibility of certain groups to live in the country and to access welfare rights (Williams, 2020). Hence, from the start, migration and welfare policies were inextricably entangled, with both dominant political parties sharing one major trait: welfare chauvinism, at least meant as the belief that "welfare benefts should be restricted to citizens" (Balch & Balabanova, 2016, p. 20). The concept of welfare chauvinism is also related to defnitions of "who is considered part of the 'community' that produces/distributes welfare provisions and what such membership is based upon" (Keskinen et al., 2016, p.: 323).

As noted by Guentner et al., (2016, p. 396), Labour's "enthusiasm for controls" in the 1970s, was going to be exceeded only by Margaret Thatcher's Nationality Act in 1981, which created a highly stratifed system of citizenship and entitlements. Many people, including children born and raised in the UK, saw their status demoted to "overseas citizens", becoming effectively stateless (Runnymede Trust, 2019). As argued by Tyler (2009, 61) "the existence of populations of failed citizens within Britain is not an accident of fawed design", but the foundation of a regime of postimperial state racism which excludes or disqualifes specifc populations from welfare and other rights. In the 1990s, successive laws by Tory and, later, Labour governments, further restricted access to benefts and services for migrant people. One key element in this process was the introduction of the principle of "no recourse to public funds" (NRPF): the requirement for new migrants "to support themselves as a condition of entry" (Dwyer et al., 2019:139). First emerged in the 1970s, NRPF was explicitly adopted by the Immigration and Asylum Act of 1999, excluding anyone subject to immigration controls from a range of welfare provisions, such as income support and housing benefts (Alberti, 2017).

Meanwhile, in 1973, the UK had joined the European Economic Community, the precursor of the European Union, seemingly marking a new era of modernization and the transition from the Empire/Commonwealth model to "a new transnational federation sharing sovereignty" (Holmwood, 2021). In recent times, the British presence in the EU has been depicted – both internally and across the continent – as uneasy and contentious, attracting particular hostility from the more conservative sectors of the political spectrum. However, at that point in time, membership of the European project was seen as a strategic move in pursuing the British liberal political agenda and its ambition for "increased market competition and privatization" (*ibid*.). In fact, it was a minority within the Labour party which was most strongly averse to EU membership – perceived as an obstacle to socialist reforms – and which in 1975 called for, and lost, a referendum on the issue. Only later did Conservative opposition emerge: frstly, against the evolution of the European project into a political and constitutional union, and secondly against the development of its social pillar, as conceived by socialist French Commissioner Jacques Delors. This opposition was epitomized by the "No. No. No." speech of Margaret Thatcher, in 1990 (BBC, 2014). Over the following two decades, the UK opted out from many major European initiatives, such as the Schengen Area of borderless movement and, later, the Euro. Still, the UK has been much more than a reluctant partner; rather, for over 40 years there have been considerable reciprocal infuences with the EU in the area of Social Policy. As noted by Hantrais (2018, p. 269), from the very start "the Commission relied heavily on British social policy specialists to coordinate the work of many of the European networks and observatories, particularly in the areas of family policy, parental leave, poverty and social exclusion, ageing and older people, and health". With regard to welfare in particular, "the UK is also the country where neoliberal policies were frst adopted on European terrain", later infuencing conservative and "Third Way" politicians across the whole continent (Keskinen et al., 2016, p. 325).

In all of this, and for a very long time, intra-European migration really was not part of the complex debate about EU membership. Even after joining, the arrivals of migrants from the continent saw only moderate increases till the end of the 1990s. These were largely made up of young people coming for study or work experiences, as well as professionals employed at the upper-end of the labour market. European citizens – often labelled as "expats", rather than "migrants" – were not much of a target of anti-immigration concerns, which instead focused on Black, Muslim and other racialized communities, as well as on refugees. These groups were widely depicted as a source of social and cultural problems (Atkin & Chattoo, 2007) and as a potential burden on the State's resources: a clear example of the "ethnonationalist" and racializing undertones which are often associated to wider sentiments of welfare chauvinism (Keskinen et al., 2016).

#### **12.3 Global Britain at the Heart of the European Regime**

At the start of the twenty-frst century, the UK strived to present itself as one of the centres of a knowledge-driven, interconnected new world (D'Angelo & Kofman, 2018). The New Labour Britain of Tony Blair (1997–2007) often has been hailed as an era of openness to globalisation, diversity and international migration (Holden, 2001; L'Hôte, 2010). The reality is somewhat more complex. On the one hand, migration policy saw an acceleration of restrictive measures against migrants, with the Prime Minister even defending the tough stance of previous Tory governments as necessary to tackle racism (Guentner et al., 2016). On the other, New Labour discourses were marked by an emphasis on integration and racial equality and, at least rhetorically, were "strongly supportive of a policy of multiculturalism" (Sommerville, 2007). Such model of multiculturalism was characterised by a strong local dimension, supported through the funding of community projects and associations. This helped the burgeoning of ethnic minority and migrant-led organisations, which in many cases acted as local providers of targeted social support (D'Angelo, 2015), working complementarily – or in parallel – to public services. The role of migrant community organisations in providing state-approved legal, welfare and migration advice was an important mechanism to enable some of the most vulnerable to access their welfare rights and navigate a state bureaucracy otherwise exclusionary. From the mid-2000s, however, this approach was scaled down signifcantly in the name of "social cohesion", with minority organisations accused of reinforcing divisions and resentment (D'Angelo, 2015) – a trend which would later lead the Conservative party to declare the "failure of multiculturalism".

At macro-economic level, New Labour was committed to the management of migration for economic gain and within the context of a "Third Way" vision aspiring to combine "the fexibility of American labour markets with some Europeanstyle protections" (Gingrich & King, 2019:90). The economic expansion of those years helped to "mask more general class-based welfare retrenchment" (Guentner et al., 2016:396), counterbalanced by some measures to reduce in-work and child poverty. Within a globalising world, the UK began to reconsider its position between global and regional trends. The pursued model was one in which competitive global markets would source skilled professionals from all over the world to fll shortages in the UK – particularly in sectors such as Health and IT. From 2008, such mechanism was going to be managed through a restrictive Point Based System (BPS). Alongside this, the growing integration into the European system would allow the removal of migration routes for "less skilled" migrants, to be replaced by mobile workers from the poorest EU Member countries (D'Angelo & Kofman, 2018). At a time of relatively low unemployment, this was seen as an uncontentious strategy to generate further economic growth (Sobolewska & Ford, 2020:144). In this regard, Antonucci and Varriale (2020:47) argued that intra-EU mobility could be seen as an essential aspect of the UK's "hegemonic role in Europe", part of a wider European regime characterised by a differentiation between core (North) and peripheral (South) countries. Functional to this was also the "knowledge-based economy" paradigm, which started as a fagship New Labour policy to then become one of the pillars of the Lisbon Strategy (2000–2010) of the European Union. This implied the leading role of Northern European countries – such as the UK – to transform the EU into the "most competitive and dynamic knowledge-based economy in the world, capable of sustainable economic growth with more and better jobs and greater social cohesion" (European Council, 2020).

It is in this context that, following the EU enlargement of 2004, the Blair administration decided, unlike most other EU governments, not to impose temporary controls on migration from the new Member States of Central and Eastern Europe. As observed by Sobolewska and Ford (2020, p. 144), "the decision was arrived at with very little ministerial or cabinet discussion, and no public consultation": something quite remarkable, considering its prominence in the public debates of the subsequent years. In fact, the effects of this decision proved to be quite different from what was expected. A study commissioned by the Home Offce – and based on the assumption that other countries would also open-up – had estimated arrivals between 5,000 and 13,000 annually (Dustmann et al., 2003). Instead, between May 2004 and September 2008, the UK saw the arrival of over 900,000 people from the new Member countries, and particularly from Poland. This fed popular views of EU migration as unpredictable and unmanageable, also thanks to the relentless media campaigns of those years. Thus, when Bulgaria and Romania also joined the EU in 2007, the access to the labour market for citizens of those countries was restricted, amid concerns about their impact on public services and wages (D'Angelo & Kofman, 2018).

The global economic crisis of 2008–2009 was going to bring further discontent, which, in turn, exacerbated anti-immigration sentiments. In other words, as revealed by an analysis of the British press by Balch and Balabanova (2016, p. 35) "welfare chauvinist ideas became more prevalent in the public debate when times were harder economically". In 2010, the new Labour leader Gordon Brown promised to act on people's concerns on immigration – famously promising more "British jobs for British workers". His mixed messages on the topic, however, produced a "media disaster" (Goodfellow, 2020, p. 120), displeasing people at the opposite ends of the migration debates. This costed him, and the party, the 2010 election, opening the way to a new coalition between the Conservatives and the Liberal Democrats.

#### **12.4 Austerity and the "Hostile Environment"**

A coalition government was quite new in the UK political tradition and at frst it was hard to see how the very different stances of the two parties – also with regard to the European Union – were going to be brought together. The "Austerity Agenda", however, soon emerged as one of the defning features of that decade (Farnsworth & Irving, 2021). For the Conservatives in particular, the aftermath of the fnancial crisis was seen as an opportunity "to pursue long-term objectives of permanently shrinking the size of the welfare state" (Taylor-Gooby, 2016, p. 713), and doing that also by further limiting benefts to British citizens (Gingrich & King, 2019). Welfare chauvinism and, more generally, a redefnition of those who were "worthy" of support, gained further political credit, with Prime Minister David Cameron explicitly presenting migration and welfare policy as "two sides of the same coin" (Morris, 2018).

The immigration of EU nationals was increasingly seen as problematic, since their rights of movement and ability to access welfare derived directly from the EU Treaties, and so were much more diffcult to regulate. The most contentious element was identifed in "low-skilled" migrant workers, because of the assumption that

they would "'steal' the jobs of the British working class" (Antonucci & Varriale, 2020, p. 47). This class element was intertwined with a national one, with major distinctions being made in public narratives between citizens of the old EU15 Member States on the one hand, and those of the new accession countries, mostly Eastern Europeans, on the other (McCarthy, 2018, p.2). British tabloids ran relentless media campaigns against Polish workers frst, and against Bulgarians and Romanians later on. As observed by Fox et al. (2012, p. 691) cultural differences were explicitly invoked to justify a racialized exclusion which implied "degrees of whiteness" and further complicated the "legacy of institutionalised racism" which had characterised the migration policies of the past. By 2015, an Ipsos-Mori (2015) poll found the UK public had become increasingly concerned about EU membership. 60% of respondents wanted greater restrictions on free movement, with only 11% happy with the current arrangements. Among those who wanted more restrictions, over 70% mentioned pressures on public services and 60% believed EU freemovers came to the UK largely to claim welfare benefts.

Much of this debate was completely disconnected from the evidence. In fact, all available data and research showed the positive net contribution of these migrants to the economy and welfare of the UK (D'Angelo, 2019). Even offcial ministerial reports noted "little evidence of any statistically signifcant labour market displacement caused by EU migrants" (DBIS, 2014, p.41). The Government's own independent Migration Advisory Committee suggested there was little indication of the UK welfare system being a pull factor for EU migration (MAC, 2014). Nonetheless, the narrative of EU migrants "stealing British jobs" and, at the same time, somewhat paradoxically, exploiting the welfare system, became stronger and stronger among the electorate and found very little political opposition across all parties, including Labour and the "Europhile" Liberal Democrats. As for the Tory party, this became increasingly divided between those who continued to see EU-membership as a political and, above all, economic necessity and those who wanted to ride the Eurosceptic and welfare-chauvinistic wave. This chasm was made all the more serious by the seemingly unstoppable rise of UKIP (UK Independence Party), the antiimmigration, anti-EU party lead by Nigel Farage.

The strategy of David Cameron to appease the electorate and its own MPs was twofold. Firstly, in January 2013, he promised that, should the Tory party secure an overall majority in the 2015 elections – an outcome at the time considered highly unlikely – the UK would hold an 'in or out' referendum on its EU membership. Secondly, he supported the introduction of a set of measures to restrict the access to welfare benefts for EU migrants (EEA), and particularly targeting job-seekers and their families (D'Angelo & Kofman, 2018). Newly arrived EU-nationals were prevented from accessing Housing Benefts and Jobseeker's Allowance (JSA), with the introduction of a minimum period of residence of 3 months. The implementation of a stricter Habitual Residence Test (HRT) for key means-tested benefts added a further level of discretion and bureaucratic burden, reducing welfare access and discouraging applications (D'Angelo & Gargiulo, 2021). Finally, the introduction of the Immigration Act 2014 saw the extension of "internal bordering" processes into all aspects of everyday life (Yuval-Davis et al., 2018). A wide range of public and private actors – including employers, landlords, banks and the National Health Services (NHS) – "found themselves legally obliged to undertake migration status checks on all those seeking to use their services or work with them" (Sobolewska & Ford, 2020, p. 169).

Overall, these measures led to the creation of what has been defned "The Hostile Environment" (Goodfellow, 2020). The term was initially coined by the then Home Secretary Theresa May, specifcally to refer to expanded checks for "illegal" immigrants. The impact of this shift, however, was much more wide-ranging, gradually affecting most categories of migrants as well as Black and minority ethnic British citizens. As noted by Sobolewska and Ford (2020, p. 169) "when combined with Britain's unusually complex citizenship and migration regimes […], and the lack of any universally held or accepted form of identifcation documents in Britain, this was a recipe for disaster". In 2018, disaster took the shape of the so-called "Windrush scandal", with hundreds of Caribbean-born residents being wrongly detained, deported and denied legal and welfare rights (Gentleman, 2019). An independent review on the scandal identifed "the organisational factors in the Home Offce which created the operating environment in which these mistakes could be made, including a culture of disbelief and carelessness when dealing with applications" as well as "an institutional ignorance and thoughtlessness towards the issue of race […] which are consistent with some elements of the defnition of institutional racism" (Williams, 2020, p. 7).

In parallel to all this, intra-European migration patterns had started to change quite signifcantly, with a shift in their regional trajectories. The number of migrants from Eastern Europe had been declining since the end of the decade, due to the relative improvement of the economies in those countries (D'Angelo & Kofman, 2017). Instead, migration from Southern Europe, and especially from Italy and Spain, saw a signifcant growth after the 2008 economic crisis, with the diffculties encountered by those welfare and labour market systems persisting even after the global recovery. For its part, the United Kingdom, although also hit by the "global crunch", continued to register an unemployment rate well below the European average. Its extremely fexible (indeed highly casualised) labour market made it relatively easy for newcomers to enter. This was particularly the case for those EU workers who could simply relocate to the country without having to fnd a job beforehand. The annual National Insurance Number (NINo) registrations of Spanish and Italian citizens, which in the early 2000s were both around 10,000, rose respectively to over 50,000 and over 40,000 by 2013 (D'Angelo & Kofman, 2017) and stayed high for the rest of the decade. Although Southern Europeans migrants remained off the radar of xenophobic and welfare-chauvinistic discourses, their contribution to the overall growth of EU-migration statistics helped sustaining anti-EU concerns. Overall, the number of EU-born migrants estimated to be living in the UK increased from 1.5 million in 2004 to 3.2 million in 2015.

#### **12.5 Brexit: Reconfguring the Migration-Welfare Nexus**

After the surprise electoral victory of 2015, David Cameron had to follow through on his promise of a referendum on the exit of the UK from the European Union; what by that point had become widely known as Brexit. In spite of his Euroscepticism, the Prime Minister was not in favour of the country's departure from the EU. His plan was to negotiate some concessions from the Brussels authorities and, on that basis, to secure a vote in support of a continued European membership. The "New Settlement" fnally secured in February 2016 – and which would have automatically come into force if the UK had voted to "remain" – focused on welfare issues. A key part of this was the so-called Emergency Break, which would have allowed the British government to extend the waiting period before EU migrants could access non-contributory in-work benefts from 3 months to 4 years. Additionally, child benefts for children living abroad would be paid at a rate refecting the standard of living in each country of origin, rather than at the UK rate (Kennedy, 2017): another aspect which had no fnancial relevance, but that had become contentious.

At that point, however, no amount of welfare reform would have appeased the anti-EU sentiments which had been unleashed and which had taken increasingly identarian tones. The Leave campaign led by Nigel Farage and by Conservative mavericks Michael Gove and Boris Johnson urged the electorate to "take back control", with more than a hint to the country's Imperial past, but also with scare stories about the potential entry of Turkey into the EU and the risk of the Mediterranean refugee crisis soon having a major impact on the British shores too. The latter was a typical example of Southern Europe being depicted as the "soft underbelly" of the regional migration system, with spurious suggestions that European "free movement" was somehow connected to refugee arrivals (as in the infamous "breaking point" poster campaign run by UKIP). Still, after years in which pundits and opinion polls (Curtice, 2016) had explained Brexit was an extreme scenario which was never going to materialise, the results of the referendum of June 23rd 2016 were received as a shocking surprise, even among those who had been campaigning for it (D'Angelo, 2019): 52% of UK voters came out in support of Brexit. According to a Lord Ashcroft (2016) poll, 80% of Brexiters saw immigrants as a "force for ill", and exiting the EU was expected to bring better immigration regulations, improved border controls and a fairer welfare system.

Immediately after the vote, David Cameron resigned, to be succeeded by Theresa May. Her premiership was going to be entirely dominated by the negotiation of a Brexit Deal with the European Union. This was signed in Autumn 2018, but voted down a number of times by Parliament, eventually leading to May's departure. It was going to be Boris Johnson – the third Conservative Prime Minister in a matter of 3 years – to renegotiate some aspects of the Deal, secure a new electoral victory (with the slogan "Get Brexit Done") and fnally get the Deal through Parliament, certifying the exit of the United Kingdom from the European Union on 31st January 2020. During the "transition year", the UK and EU also negotiated a Trade and Cooperation Agreement (TCA), signed in December 2020. The new relationship

saw the end of free movement between the two parties, and the withdrawal of the UK from the Single Market, the Customs Union, and most EU programmes. However, many political and economic issues were left unresolved, opening a long phase of complex disentanglement from decades of regional integration.

In terms of migration and welfare rights, newly arrived EU migrants were going to be subject to the same regulations previously applied to third country nationals. In the aftermath of Brexit, the development of a new migration system ended up coinciding with the Covid-19 pandemic and with the substantial halt of all international migration resulting from it (Sumption, 2021). Quite soon, this showed the reliance of the UK labour market on migrant and transnational workers, particularly in sectors such as health and personal care, but also agriculture, retail and logistics. Paradoxically, after years of campaigns against EU free movement, Britain had to rush into organizing emergency charter fights to bring in Romanian workers, notwithstanding the Covid-19 restrictions (D'Angelo, 2020). Meanwhile, opinion polls had started to indicate a reduction in the concerns about immigration and much more positive views about its impact (Runge, 2019). The new consensus seemed to shift from the previous obsessions with limiting the number of arrivals, to one focusing more on their selection for economic beneft. In January 2021, Home Secretary Preti Patel brought forward a new "Point Based Immigration System". This was for the most part an update of the previous 2008 PBS, with the major difference being the fact that this now applied equally to EU and non-EU nationals. Citizens from the 27 countries of the European Union, including skilled workers, "now require a visa to live or work in the UK, and must pay substantial costs, such as the NHS surcharge of £624 per person per year, which must be paid up-front when applying for the visa" (Walsh, 2021). On the other hand, for non-EU migrants, the new system represented a relative relaxation of the entry criteria, with lower salary and skill thresholds and no cap on numbers (Portes, 2021).

Even before any new measure was put in place, the symbolic effect of the Brexit referendum produced quite a signifcant impact on migration trends. After a peak ahead of the vote, net migration statistics showed a decline between 2016 and 2018, to then level off, standing around 58,000 in the year ending March 2020 (ONS, 2020). Data from the Offce for National Statistics also reveal a reduction in the proportion of EU citizens coming to the UK for work-related reasons and, among those, an increase in those who arrived for a defnite job as opposed to looking for work (Lomax, 2019). This trend has been consistent since straight after the referendum, suggesting that – well before the UK actually left the Union – it was the very idea of intra-European mobility that got affected. A signifcant number of EU nationals were put off by the unwelcoming post-Brexit climate and, later, by the uncertainty of the pandemic and the relative weakness of the pound, which made other European countries more attractive (Bounds, 2021). By summer 2021, British employers reported the worst staff shortages in over 20 years (Partington, 2021a), with a study by a major employment website suggesting the number of EU citizens seeking work in the UK had fallen by 36% since Brexit (Partington, 2021b). Meanwhile, since the mid-2010s, non-EU net migration had gradually increased to some of the highest levels since record began in 1975.

As for those EU nationals who had entered the UK before Brexit, already during the referendum campaign the Westminster government had guaranteed that they would have been able to remain without a substantial change in their rights. For this purpose, in 2019, the Home Offce launched the so-called "EU settlement scheme" (EUSS). This allowed EU citizens to apply for a status equivalent to an "indefnite leave to remain" if they could demonstrate their residence prior to 31st December 2020 and having exercised EU Treaty rights for a continuous period of 5 years. Those who entered the UK before that date but did not have 5 years of residence yet, could receive a "pre-settled status" of a temporary nature, pending the acquisition of the requirements for permanency. For all, the deadline for submitting an application was June 30th, 2021. Three months ahead of that (March 31st), 5.3 million people had applied to the scheme (Home Offce, 2021), with the main nationalities including Polish (975,180), Romanian (918,270), Italian (500,550), Portuguese (376,440) and Spanish (320,850). Of all the applications reviewed at that point (just under fve million), 53% were granted settled status, 44% were granted pre-settled status and 3% had other outcomes, including refusal or withdrawal. As of 31st May – 1 month before the deadline – the number of applications received had reached 5.6 million (Morris & Reuben, 2021).

Since the EUSS scheme was announced, many advocacy organisations expressed concerns that this was going to leave out a substantial number of European citizens, who in many cases remained unaware of the procedures to be followed or whose applications could be rejected for defects of form or inability to produce adequate documentation. For these EU citizens – who had entered the UK freely under EU law, with entitlements nearly identical to UK nationals (Dunin-Wasowicz, 2019) – the risk was to become unlawfully resident overnight. Although the Government claimed applications were processed in about 5 days, data obtained through "freedom of information" requests to the Home Offce found thousands of delayed applications, many involving children, and in some cases sitting in the system for over a year (see e.g. Bulman, 2021). As highlighted in an IPPR report, those excluded from the scheme could face a range of "hostile environment" measures, including "barriers to starting a new job, renting a new property, accessing free secondary healthcare, making a beneft claim, opening a bank account, and obtaining a driving licence" (Morris, 2021). This would amount to a dramatic repetition of the "Windrush scandal". Like before, within the wider British welfare regime, signifcant decision-making and discretionary powers are left with employers, landlords, doctors and a range of private actors with no knowledge of migration laws and procedures (McKinney, 2021); however, these are actively encouraged to be overzealous in order to avoid facing fnes or prosecution. Also in this respect, the British regime was meant to be diffuse but also highly consistent in its bordering endeavours.

This new climate of hostility towards EU "free movers" was also epitomised by the number of EU nationals which were prevented from entering the UK during the Covid-19 pandemic. According to fgures published by the Guardian newspaper (Tremlett & O'Carrol, 2021), a total of 3294 people were stopped and sent back between March and April 2021 because they did not have a job in the UK or any other valid reason for entry (this compares with 493 in the frst quarter of 2020,

when air traffc was 20 times higher). Several citizens of Italy, France, Bulgaria, Greece and other nations were forced to spend the night in the airport or locked-up in detention centres, waiting to be placed on a return fight. The media reports of their stories often echoed those of non-European migrants, sending the clear message that EU-nationals were now as unprivileged as any other migrant.

In parallel to this, it is interesting to notice how, now that EU migrants could not be constructed as much of an internal problem anymore, British public discourses shifted once again against asylum seekers and refugees. A clear example of this was the media hype around the increased – but still pretty limited – sea crossings from the French to the English shores. Even when, on 24th November 2021, 27 asylum seekers drowned in the Channel, the UK Government doubled down on its plans to further militarise its costal borders, whilst many media outlets deployed the narrative of the "illegal migrants" seeking easy access to an overly generous welfare system. Overall, these constant attempts at re-shifting the target of anti-immigration concerns have been interpreted as the "active remaking of colonial modes of rule through the ongoing logics of authorised and unauthorised mobilities" (Davies et al., 2021: 2322), keeping alive the permanent threat of racialised undeserving others: from the Empire to Europe, and back again, with chauvinism as a constant thread.

#### **12.6 Conclusions: A Very British European Regime?**

As examined in the previous sections, Brexit was not a sudden departure but, rather, the "awakening" (Sobolewska & Ford, 2020) of tensions which had been bubbling under the surface of the political mainstream for years. Some of these major trends include ideas of Austerity and one of its key corollaries: welfare chauvinism. As noted by Farnsworth and Irving (2021), Austerity – the systematic cut of public expenditure – is not an economic project but a "political project aimed at transforming the welfare state". This has been most clearly articulated in the "Anglo-neoliberal" world and has some of its roots in Thatcherism (Irving, 2021), but it ended up characterized a long era of social policy throughout Europe and well beyond. As pointed out by Guentner et al., (2016, p.: 392), across the EU, after "a neoliberal roll-back, and in an ongoing "dual crisis" of welfare states and national identities […] the very foundations of welfare provisions are put in question".

Austerity and welfare chauvinism are connected to an ongoing narrative of the deserving and underserving, which changes its target groups over time, but which is also constant in its function: stratifying the access to welfare and services in terms of migration origin, race, gender and class, making welfare and civil rights conditional, exclusionary, and the primary mechanism of internal bordering. Keskinen et al., (2016, p. 322) argued that "it has become more legitimate than ever to claim that welfare benefts should be reserved for certain groups alone, notably those considered 'natives' and bearing a self-evident right to belong to the nation, and to develop policies on such bases". This process needs to be constantly fueled by popular hostility and the sense of an impending crisis (Menjívar et al., 2018). The extent to which these are "real" or "manufactured" – as in the various examples presented in relation to the UK – remains up for discussion and, to an extent, is beside the point.

Mechanisms of stratifcation are not at all new and, although the way in which they are put in place varies across countries, depending on legislative frameworks and socio-economic differences, they emerge among the defning traits of migration regimes throughout Europe, as exemplifed by the work of Morris (2003) on the UK, Germany and Italy. In all these cases – and more – welfare stratifcation, underpinned by ideas of "deservedness", rather than universalism, and by pushing the minimum common denominator to the bottom, produce restrictive effects not just for migrants, but for everyone. The basic requirement to implement such stratifed systems is not the ability of the state to control entry into the country, but that of regulating access to welfare and public services. In turn, this requires a "migration regime" (Cvajner et al., 2018) characterized by the concerted efforts of a wide range of public and private sector actors under the centralizing direction of the government.

In the UK of the mid-2000s, EU nationals were framed as easily available unskilled workforce; however, their citizenship rights were seen as incompatible with the principles of welfare-chauvinism which had defned decades of British policies across the political spectrum. Thus, it is possible to interpret Brexit not so much as a repudiation of the whole European migration regime, but simply as an attempt to reject the status of these transnational citizens and to place them back into their natural role of "mobile workers" (D'Angelo & Kofman, 2018).

As examined in the previous sections, in spite of the myth of the UK as a reluctant member of the EU, constantly trying to push back and diverge from trends of Europeanisation, another reading is possible. Indeed, the history of the past four decades has been marked by major British infuences on the European regime. Ideas of "Third Way" welfare models, the vision of a geographically differentiated "knowledge economy", and the diffusion of practices of "internal bordering", all originated – or frst reached the mainstream – in Britain, to then become highly infuential into policy discourses and processes across the whole of Europe.

As discussed elsewhere (D'Angelo & Gargiulo, 2021), even the idea of regulating welfare access through the recognition – or rejection – of the status of "residence", in spite of extremely different administrative systems, is something we can fnd in all corners of Europe, from the United Kingdom (with its "Habitual Residence Test") to Italy (Gargiulo, 2021). What these different European regimes have in common is the delegation of "bordering" powers to local actors, with signifcant degrees of discretionality, but all within a coherent framework of chauvinistic "hostility", which is informed more by ideological than pragmatic consideration (D'Angelo & Gargiulo, 2021).

In all these respects, the UK emerges as an "extreme" case of these general trends (Taylor-Gooby, 2016, p. 717), with Brexit as a means to accelerate in this direction. In fact, Brexit can be seen as the paradoxical result of taking some of the essential features of European regimes to their extreme consequences: the exit from the European Union as EU trend-setting. In the long run, it is unlikely that these dynamics will lead to further departures from the EU; rather, they may push to a further redefnition of the European Migration-Welfare nexus in exclusionary, chauvinistic terms and into a restriction of the actual remit of EU citizenship. Very different outcomes are of course also possible, and with Britain now outside of the European institutions, a new political consensus is all to be built.

#### **References**


Williams, W. (2020). *Windrush. Lessons learned review. Independent review*. House of Commons.

Yuval-Davis, N., Wemyss, G., & Cassidy, K. (2018). Everyday bordering, belonging and the reorientation of British immigration legislation. *Sociology, 52*(2), 228–244.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 13 Turning the Welfare-Migration Nexus Upside-Down: The Case of European Retirees in Spain**

**Claudia Finotelli**

#### **13.1 Introduction**

European citizenship guarantees freedom of movement within the EU and equal access to social rights for EU citizens.1 Yet, post-recession intra-EU fows from Southern to Northern European countries have challenged the overall acceptance of the freedom of movement as inherent to EU citizenship and equal access to welfare. The mobility of young Southern (and Eastern) European workers started to be perceived as a burden by destination countries. In 2013, German, Austrian, Dutch and UK representatives complained to the European Commission that intra-EU migration "burdens the host societies with considerable additional costs, in particular caused by the provision of schooling, health care and adequate accommodation" (Mikl-Leitner et al., 2013). Since then, scholars have addressed the challenges of post-crisis intra-EU mobility to alert against the limitations of freedom of movement as a constitutive element of EU citizenship (Bruzelius et al., 2017; Engbersen et al., 2017; Lafeur & Stanek, 2017) or to nuance both theoretically and empirically the instrumental uses of EU citizenship (Finotelli et al., 2018; Harpaz & Mateos, 2019). Legal restrictions to social rights went along with exclusionary practices towards "unwelcome" EU citizens from Southern and Eastern European Member States (Perna, 2018). Social rights retrenchments within the EU citizenship regime are currently a reality in several EU countries contributing to the growing awareness of EU citizenship as a citizenship with shrinking social rights (Barbulescu & Favell, 2020).

C. Finotelli (\*)

Complutense University, Madrid, Spain e-mail: cfnotel@cps.ucm.es

© The Author(s) 2023

<sup>1</sup>This work was supported by the Spanish Ministry of Science and Innovation (Grant Nr. PID2019-104706GB-100).

C. Finotelli, I. Ponzo (eds.), *Migration Control Logics and Strategies in Europe*, IMISCOE Research Series, https://doi.org/10.1007/978-3-031-26002-5\_13

Interestingly, the debate on European citizenship as a source of equal access to welfare has mainly revolved around labour mobility among Southern and Eastern European workers in Northern Europe (Trenz & Triandafyllidou, 2017; Barbulescu, 2017; King et al., 2017). After having been addressed as transit countries for asylum seekers and hotbeds of irregular migration, Southern European countries were rapidly stigmatised as sources of unwanted EU-citizens. By contrast, non-labour motivated mobility of (mostly elderly) EU migrants and its welfare impact have largely been side-lined by the scholarly debate with few exceptions (Thym, 2015; Bruzelius et al., 2017; Finotelli, 2021). In other words, scholars have mainly addressed the question of how the retrenchment of intra-EU mobility rights has affected Southern Europeans *en route* to Northern Europe rather than the impact of Northern European intra-EU migrants in Southern Europe. But what about the non-labour-motivated mobility from Northern European countries in Southern Europe? To what extent is it perceived as a welfare burden for Southern European countries?

To answer these questions, this chapter analyses the relationship between intra-EU mobility and welfare chauvinism by focusing attention on the mobility of Northern EU citizens in Southern Europe, who are motivated by factors other than labour, and who have traditionally been welcomed in their new countries of residence. In line with the concept of regime proposed in the introduction of this volume, the aim of the chapter is to show how the perception of intra-EU migrants as "welfare scroungers" has contributed to modify the mobility regime of intra-EU citizens through restrictionist legislation changes and local practices, and hence control the freedom of movement and its consequences as a constitutive element of EU citizenship.

To do so, the following pages address the provision of healthcare assistance to intra-EU retirees in Spain, as one of the main categories of non-labour-motivated movement in the EU. The choice of Spain as a reference case is due to its relevance as a destination country for intra-EU retirees from Northern Europe (King et al., 2000; Casado-Díaz, 2006; Gustafson, 2008; Huete et al., 2013). As Finotelli (2021: 609) noted, "The EU is not only a space of freedom of movement, but also of freedom of retirement" in which Spain has the lion's share with the largest population of non-native Europeans over 65 in Southern Europe (Eurostat, 2016). In addition, the Spanish healthcare sector is a tax-funded healthcare system which provides healthcare on a universal basis. Against this backdrop, analysis of the Spanish case will show that intra-EU retirees as healthcare users in Southern Europe can be perceived as a welfare burdens who challenge the equity of the welfare regime and therefore legitimate restrictions in the access to social rights.

The majority of the analysis is based on a review of the local newspaper articles, offcial documents and parliamentary memos. The chapter is divided into four parts. The frst section provides an overview on retirement migration in Spain as it has been addressed in the literature. The second part addresses the evolution of the public perception on the presence of intra-EU retirees in Spain in particular in regard to their use of healthcare services. The third part analyses the modifcations of the legal framework currently regulating access to healthcare provision for EU citizens in Spain while some considerations about the relevance of intra-EU mobility to better understand the welfare-migration nexus are provided in the fourth and concluding part of the chapter.

#### **13.2 Intra-EU Retirement Migration in Europe and in Spain: An Overview**

International retirement migration does not represent a type of mobility aimed at improving individuals' opportunities by inclusion into welfare (Bommes, 1999; Sciortino & Finotelli, 2015), but rather a type of mobility aimed at improving individual well-being (King et al., 2017). In contrast to other forms of migration, international retirement migration can be seen as the consequence of an excess of resources, rather than of a shortfall of them (King et al., 2000; Warnes et al., 2004). Moreover, international retirees are seen by destination countries as a group of migrants with great economic potential. In the last few years, Southern European countries have implemented measures to attract wealthy Northern European retirees through tax reductions or the construction of senior housing in attractive environments. This, of course, has caused certain discomfort in countries of origin and destination. For instance, the favourable Portuguese taxation regime for nonresidents was sharply criticised by the Portuguese public, while the Swedish and the Italian governments scolded their own retirees for searching for tax benefts in countries such as Portugal (Cottone, 2017; Smith, 2020). Nevertheless, the fact that international retirees are intra-EU migrants who take advantage of a favourable institutional environment has certainly diminished the research interest for the welfare challenges represented by this group of migrants. Indeed, international retirement migration has been traditionally explained from the perspective of the tourism-migration nexus. The motivations of residential tourists are mainly related to the search for a warmer climate and a desire to improve their own health conditions (King et al., 2000; Rodríguez, 2001); this has turned international retirement migration into a particularly extended form of lifestyle migration (Benson & O'Reilly, 2009; Janoschka & Haas, 2013). International retirees are considered "active migrants" who search for gratifying experiences abroad, rather than vulnerable migrants at risk of becoming dependent on welfare services such as healthcare and long-term care in their country of residence. In these studies that analyse international retirement migration, tourism rather than the aging process plays the crucial role.

In contrast to tourism, the welfare-migration nexus is a rather under-researched angle in the feld of international retirement migration. Scholars have started to analyse how elderly migrants deal with their vulnerability when they enter their "fourth" age (Cela & Fokkema, 2017; Ciobanu et al., 2017). However, very few researchers have studied how this applies to international retirees. Some scholars have analysed how poverty or illness encourage a return to the country of origin (Dwyer & Papadimitriou, 2007), while others have rather focused on transnational practices to solve the contradiction between the obligation to have only one residence for fscal and legal purposes and fexible forms of individual mobility (Ackers & Dwyer, 2002; Horn et al., 2013). Analyses of the welfare-migration nexus in the case of international retirement migration have mainly used Spain as a reference case due to its long history as destination of intra-EU migrants and the progressive aging of its population of international retirees. As a matter of fact, with respect to the age structure, the share of retirees older than 70 years have increased between 2008 and 2019. This means that more and more of those retirees that were formerly considered residential tourists are bound to become people with an increasing need of healthcare and care services (Finotelli, 2021). Most of the works on the welfare migration nexus focus on British retirees in Spain since they represent the largest and less concentrated group of retirees in Spain. Some researchers have already started to focus on the physical vulnerability of British retirees and possible return strategies (Huete et al., 2013; Giner-Monfort et al., 2016), while a smaller group of researchers has looked at what happens to those who decide to stay in Spain despite worsening health conditions (Ahmed & Hall, 2016; Hall & Hardill, 2016). Yet, most of the currently available works on the welfare-migration nexus focus on the use that retirees make of welfare services and their satisfaction with the Spanish healthcare system (Betty & Cahill, 1999; Legido-Quingley & La Parra, 2007; Coldron & Ackers, 2009; Legido-Quigley et al., 2012; Calzada, 2018). Others have analysed how international retirees rely on affordable private welfare providers such as migrant care workers (Gavanas & Calzada, 2016), charity organisations (Haas, 2013) or transnational practices to obtain social protection (Schriewer & Rodes, 2008; Gehring, 2015). However, there is a persisting research gap about the social and institutional impact of international retirement migration and retirees' growing welfare dependency in the countries of destination.

#### *13.2.1 The Healthcare "Scrounger" Stereotype*

The Spanish healthcare sector provides healthcare on a universal basis and is mainly fnanced through tax allocations and allocations from general state budgets. No arrangements are made for co-payments for certain services in the Spanish healthcare sector, while the quality of the services provided is still perceived as satisfactory by a large part of the Spanish population (OECD, 2016). That is why the spending cuts enacted during the Great Recession did not affect positive attitudes regarding the public healthcare system among the population (Petmesidou et al., 2014). Yet, the Spanish population's positive attitude did not prevent growing discomfort among the public towards intra-EU retirees as benefciaries of the Spanish welfare regime, and particularly its healthcare regime. Healthcare access of international retirees in the previous years had been an object of relatively little attention by scholars and the general public. At the end of 1990s, research conducted in the Province of Alicante, one of the most attractive regions to EU retirees showed that at least the half of the interviewed immigrants, including EU retirees, visited a

doctor no more than once a year, while this was not the case of natives who visited their doctor more often (Ferrer Cascales et al., 1997). Such fndings were in line with the fact that most EU retirees at the time had not reached their "fourth" age, where healthcare and more general care needs start to appear. However, in the frst decade of the new century, new types of healthcare patterns among intra-EU retirees started to emerge. It was, for instance, observed that intra-EU retirees residing in Spain often require more costly treatments in comparison to other categories of migrants as they are more exposed to chronic diseases due to the progressive aging of this group (Fernández Molina, 2005; Pi, 2007). But it is only in concomitance with the Great Recession that the perception of intra-EU retirees as a welfare burden enters into the Spanish public debate. In 2009, a report by the General Council of Nurses questions, for instance, the large number of hip-surgeries and by-pass surgeries in the hospitals of the Costa del Sol received by German and Dutch citizens (Nexotur, 2009). The same report refers to alleged "trickeries" implemented by many EU-retirees to circumvent EU legislation in order to be treated in the Spanish healthcare system, such as for instance a sudden attack related to a pre-existent illness to be attended by a Spanish hospital. According to the President of the Trade Union of Medical Doctors of the Autonomous Community of Valencia, "picaresque" is widespread and "many citizens, foremost Europeans, arrive in Spain not only attracted by sun and beach but also by the 'good reputation' of its healthcare professionals" (Martín Plaza, 2009). The stereotype of Spain as "Europe's retirement home" and as "incomparable surgery paradise" for European tourists arriving in Spain to enjoy sun, sand and high-quality medical treatment turns into a frequent topic in the Spanish public debate. Intra-EU retirees start to be represented as a burden for the Spanish healthcare regime fuelling the perception that healthcare provision for natives could be at stake. The President of the Nurse Council argues that a foreigner older than 65 is a healthcare consumer by default for whom the Spanish healthcare regime is not prepared (Nexotur, 2009). Likewise, the President of the Autonomous Community of Extremadura claims for a healthcare regime "for Spaniards and only for Spaniards" (Martín Plaza, 2009). The debate also involves criticism of Member States "unfairness". So, it is suggested that the UK Embassy provides information to UK citizens about how get access to the Spanish system giving "practical advice on the procedures to follow to be able to enjoy the Spanish healthcare system as any other taxpayer" (Martínez, 2012) while some media point to the existence of residential developments aimed at host intra-EU retirees during few months while they are "unfairly" treated by a Spanish hospital (García, 2012). Dissonant voices were rare. It is for instance the case of the Economic Head of the Healthcare Service of the Autonomous Community of Murcia who declared that the Government of Community of Murcia analysed the hospital admissions of European patients and the outpatient surgery performed on these patients in recent years and has not detected this problem. In his words: "Despite the myth generated about the supposed mass trips of European retirees to Spanish hospitals to undergo hip surgery, the so-called health tourism has never existed. This is what all administrations and most experts defend. It's not real; It simply does not exist" (Parra, 2010).

The general discontent about healthcare access of non-labour motivated migrants from other EU countries seemed to confrm the argument that tax-funded healthcare regimes, such as the Spanish one, are bound to produce "spill over" costs, which could easily turn into a source of potential political confict over freedom of movement as a contributor to equal access to welfare (Ruhs & Palme, 2018). That is also why EU regulation on healthcare access became a matter of discussion at EU and national level especially after the Great Recession of 2009.

#### **13.3 Cross-Border Healthcare Provision for EU Citizens**

#### *13.3.1 The European Framework*

Healthcare policies are the exclusive competence of Member States. However, a set of norms has been approved in the last decade, with the objective of contributing to "the protection of human health" as a basic right of EU citizens. This has led to a two-track system in the provision of healthcare services for EU citizens. The frst track of healthcare assistance for intra-EU migrants is represented by Regulation (EC) No 883/2004 of 29 April 2004 and Regulation (EC) No 987/2009 of 16 September 2009 (hereafter the Regulations) on the coordination of social security systems, while Directive 2011/24/EU (implemented in Spain by Royal Decree 81/2014 of 9 March 2014) on the application of patients' rights in cross-border healthcare (hereafter the Directive) represents the second track. The subject and the object of the Regulations and the Directive are similar, but their objectives are different: the purpose of the Regulations is to guarantee freedom of movement, while the Directive regulates free provision of services within the EU (Álvarez González, 2018).

To receive unplanned healthcare assistance under the Regulations, EU patients need to be insured in a public insurance system in their country of origin (Finotelli, 2021). Reimbursement under the Regulations is inter-institutional, and patients are not allowed the option to choose treatment from a private healthcare provider. The Regulations not only address access to healthcare in the case of residency in another EU country: in the case of stays shorter than 3 months, EU citizens are required to present a European Healthcare Card and can receive medical treatment only in the case of an accident or sudden illness following travel to the country of stay.

In contrast, unplanned healthcare assistance under the Directive can be provided upon explicit request by the patient without any previous formality. Nevertheless, treatment benefciaries must pay for their treatment in advance and apply for reimbursement afterwards. In such a case, reimbursement is processed by their national institutions according to the rates established for the same treatment in the country of origin (2011/24/EU, Article 7(4)). However, and in clear contradiction with the Regulations, patients can decide between public and private healthcare providers in the country of destination. Since reimbursement procedures under the Regulations

are more advantageous to patients, the Regulations are used as the reference norm for access to unplanned healthcare assistance, except in the case that the patient explicitly requests to be treated under the Directive (Álvarez González, 2018). Finally, neither the Regulations nor the Directive include long-term care and palliative care in the authorised treatments.

As can be seen in Table 13.1, in the case of unplanned healthcare assistance major differences between the Directive and the Regulations regard reimbursement procedures and the possibility to choose private healthcare providers. More similarity can be found for planned assistance since both the Regulations and the Directive formally require previous authorisation from citizens' corresponding national institutions. In both cases, authorisation can be issued if the treatment is admitted by the national healthcare authority and if the corresponding treatment cannot be provided in a realistic timeframe in the patient's country of origin.

The approval of the Directive was characterised by a deep tension between the principle to guarantee free healthcare access to EU citizens and the will of Member State to keep control on patients' freedom of movement. The preparation document of the Spanish Presidency in 2010 declared that the main Presidency objective was "to strike the right balance between the rights of the patients in cross-border healthcare and the responsibilities of the Member States for the organisation and delivery of health services and medical care" (Council of the EU, 2010, p. 2). The declaration was related to the fact that reimbursement and prior authorisation represented a major contentious issue during the approval procedure of the Directive. Against this backdrop, the Spanish Presidency also reminded that the European Court of Justice "has held on to a number of occasions that it is possible for the risk of seriously undermining the fnancial balance of a social security system to constitute *per se* an overriding reason in the general interest capable of justifying an obstacle to the


**Table 13.1** EU citizens' access to healthcare

Source: Finotelli (2021)

freedom to provide services". This allowed the Presidency to introduce relevant modifcations into the second version of the Directive to be presented to the Commission. According to the new version of the document, prior authorisation can be denied for reasons of general interest, such as the general need to maintain oversight over costs, or to guarantee a certain degree of equity in access to high-quality healthcare treatments (Boggero, 2018). In addition, according to the general rule, the Member State competent to grant a prior authorisation according to Regulation 883/2004 (i.e. the Member State of residence) is also the Member State who should reimburse the cost of cross-border healthcare for pensioners. However, if a pensioner is treated in his country of origin (it means where he/she was initially insured), the new Directive proposal establishes that this country would have to provide healthcare at its own expense. In this vein, the new document contradicts the previous version of the Directive which established that the Member State of affliation was responsible for reimbursing the cost of hospital care provided in another Member State in the case that its social security system would have assumed the costs if the healthcare had been directly provided in its territory (Art. 8.3a). The objective of the modifcation was to avoid that pensioners with their residence in Spain could go back to their country of origin or original country of affliation to be treated there while the new country of residence, Spain for instance, had to reimburse the treatment costs. This allowed the Spanish government to block the Directive initiative promoted by Sweden and the United Kingdom who was particularly benefcial for these countries due to their condition of sending countries of pensioners (Martínez de Rituerto & Prats, 2010). At the same time, the approval of the new Directive, and in particular the long-standing debate on the defnition of the Member State of affliation, also allowed the Spanish government to modify the regulation of healthcare provision for EU citizens in Spain.

#### *13.3.2 The Provision of Cross-Border Healthcare in Spain*

The new Regulation of healthcare access of EU citizens in Spain, and in particular of intra-EU retirees, was boosted by the new government ruled by the Partido Popular (2011–2016), elected during the peak of the economic crisis in Spain. The 2012 annual report published by the Spanish Court of Auditors (*Tribunal de Cuentas*) denounced the "repeated misuse" of the European Healthcare Card. As the Court reported, Spain provided healthcare assistance to EU citizens insured in their home country (and not resident in Spain) who decided to move to Spain with the aim of receiving medical treatment under the Healthcare Card, which was only intended for temporary stays. Instead, these new residents were supposed to use the E-112 form, which, in turn, required the specifc approval of the corresponding Autonomous Healthcare Service (as requested by Art. 20, Reg. 833/2004) (*Tribunal de Cuentas*, p. 143). As the report stated, "the misuse of the Healthcare Card allows them [retirees] to keep privileges in their country without giving up complete and cost-free healthcare in Spain" (*ibidem*, 142). The publication of the report was

supported by the declarations of the then Minister of Health, Ana Mato, who argued that the budget of the Spanish National Health System was assuming the healthcare of people who already have it covered within their country. In particular, according to the ministry, almost 700,000 foreigners have accessed the health card without the right to do so, which has caused an expense of 917 million euros (García, 2012).

As a response to this and other perceived challenges, such as irregular migrants' access to universal care in Spain, the Spanish government approved Royal Decree n. 16/2012 (hereafter the Decree) on urgent measures regarding healthcare provision in Spain. The Decree did not only exclude irregular migrants from universal healthcare but also introduced other types of restrictions that specifcally addressed EU citizens with the aim of preventing potential misuses of the universal healthcare provided in Spain. As the Decree's preamble stated:

The Court of Auditors has shown that the National Healthcare System, fnanced through the National Budget Plan, has taken over the healthcare of people who already enjoy comprehensive sickness coverage either through the institutions of their country of origin or through private insurance systems, which is currently eroding its fnancial capacity and preventing the improvement of services by their managers.

The objective of the Decree was to prevent the erosion of the National Healthcare System's fnancial capacity. For the frst time, the Decree described how inactive EU citizens had to prove their insurance situation, and specifed which rights and duties were associated with it. Its main objective was to guarantee that a Social Security Card would only be issued to intra-EU retirees who were formally registered as residents of Spain in order "to revise a situation that was described as being 'like paradise' for inactive EU residents" (Finotelli, 2021, p. 614). According to the new procedure introduced by the Decree, inactive EU residents have to explicitly apply for their Spanish Social Security Card at the corresponding offce of the Spanish Social Security System. To do this, they have to prove that they are insured in their home country by presenting an S1 certifcate, issued by the Social Security Offce in their home country. In addition, they have to demonstrate that they are registered with the Spanish municipal registry (*Padrón Municipal*), and are in possession of a Foreign Residence Card (NIE). The latter is issued by the local Foreigners' Offce, where applicants for the Social Insurance Card have to provide documentary evidence that there is no risk of them becoming "a burden on the Spanish state during their period of residence" (Art. 7.1.). Once registered in the National Security System, healthcare assistance to EU citizens is provided on the basis of the principle of non-discrimination with Spanish citizens. Clearly, the new procedure represented a substantial change compared to the past.

The goal of the new national legislation (including the procedure for obtaining an NIE) was to prevent EU citizens without full residency in Spain from repeatedly receiving medical assistance from the Spanish healthcare system beyond the assistance provided in the case of emergency situations with the European Healthcare Card. By introducing the requirement of municipal registration, the reform was also intended to correct the practice among most intra-EU retirees, who did not enrol in the municipal registry. This had made it diffcult for local authorities to plan the service provision of their population. Before this, there were reports that some municipal registry offces had tried to tackle this problem on their own. For instance, for a certain period of time, municipal offcials in a well-known town on the Costa del Sol refused to register EU residents in the municipal registry if they had not presented their Foreign Residence Card to prove that they had been residing in Spain for at least 3 months (Finotelli, 2021). Strikingly, this was not applied to non-EU citizens, who could register with their passport alone. This "bad" practice, which has been now corrected by the National Statistical Offce, only provides anecdotal evidence about administration routines at the local level. Nevertheless, it cannot be excluded that the intention of the offcials involved was to reduce the number of people eligible for healthcare insurance in the region (Finotelli, 2021). The above-mentioned episode also suggests that traditionally "welcome" migrants, such as economically inactive, supposedly wealthy European retirees, might not be exempt from discretionary implementation practices at the local level to prevent any alleged misuse of Spanish healthcare services.

Overall, analysis shows that the set of "norms and practices" (Jenson, 2007) constituting the EU citizenship regime also experienced changes in Spain after 2009. According to the general public, the Decree put an end to the era of welcoming "legions of European pensioners eager for sun, sea … and hospital beds" (García, 2012). In more general terms, the Spanish case showed that interaction between intra-EU residents and universal healthcare regimes could trigger a restrictive turn in the European citizenship regime, where the alleged welfare burden represented by EU citizens, and their observation as potential welfare scroungers by the state, could be contrasted by legislation changes and possibly negative implementation practices. Yet, the main objective of limiting fnancial erosion was only partially achieved as the next section will show.

#### **13.4 Intra-EU Mobility: Between Welfare Restrictions and New Market Opportunities**

The decision to more accurately defne the category of people who have access to the Social Security Card was closely linked to concerns about the fnancial capacity of the Spanish Healthcare system. Restrictions helped to reduce negative reporting on this category of migrants in the media (Finotelli & Sebastián Izquierdo, 2019). However, it does not seem that these restrictions, regardless of their harshness, have done anything to contribute to reducing the amount of Spain's bill. The reasons for this lie partly in the unchanged mechanism of the intra-European reimbursement regime. According to EU regulation, the reimbursement amount for treatment provided during occasional stays (such as tourism) is always based on actual expenditure, while reimbursement for treatment provided to residents can also be calculated on the basis of a fxed amount per calendar year (Spanish Government, 2016). Spain, together with the United Kingdom, Portugal, Sweden, Ireland, Cyprus,

Norway, Finland, and the Netherlands, implements reimbursement on the basis of fxed amounts, which in the case of Spain is 250 euros per month (Congreso de los Diputados, 2016). To assess the amount of what other Member States owe to Spain, the fxed amount (*cuota global*) is multiplied by the number of foreign inactive residents (mostly pensioners) residing in the country during a given year (*Tribunal* de Cuentas, 2012, 177).

The report of the Court of Auditors highlighted the existence of a remarkable gap between Spain's expenditure for the healthcare assistance of EU citizens and the amounts that Member States reimbursed Spain for this purpose (*Tribunal* de Cuentas, 2012). In fact, other EU countries, led by the UK, owed Spain a total sum of about 450 million euros. The Court of Auditors argued that cumbersome reimbursement mechanisms and timeframes as well as a lack of transparency in the available accounting data were the reasons for such a large amount of money. In addition, the Court claimed that the fxed amount of 250 euros calculated for Spain is below the European average and is not suffcient to cover the actual expenses of the Spanish Social Security System for EU residents, particularly British nationals. Such an unfavourable balance between invoices based on fxed amounts and actual expenditures did not change after the reform of 2012. As information from the Spanish government shows, in 2015 Spain billed around 467 million euros for fxed-amount charges as well as 168 million euros of actual expenses for medical treatment to citizens of other EU Member States, half of them British citizens (see Table 13.2). In turn, Spain owed only 46 million euros to other European countries for their healthcare assistance to Spanish citizens abroad (Spanish Government, 2016).

Taking into account that fxed amounts are always related to *inactive* residents in Spain, the data described above reveal that the majority of Spain's credit with other countries encompass healthcare expenses for inactive EU citizens who reside in Spain but do not work there, such as EU retirees. One political party, the centreright *Ciudadanos*, drafted a (non-legislative) motion in 2016 to discuss this fnancial issue in the Spanish parliament. However, the Spanish government has shown little interest to date in tackling this issue. As one deputy from *Ciudadanos* argued in an interview, the Spanish government might not be interested in publicly raising the debate of compensation funds "because the income provided by tourism activities


**Table 13.2** Billing amounts for healthcare assistance provided by Spain according to standardised rates and real amount of expenses (2015)

Source: Spanish Government, 2016, published in Finotelli (2021)

de facto offsets the costs generated by EU residents using the Spanish public healthcare system" (Finotelli, 2021, p. 616). In fact, in 2018 more than 82 million tourists visited Spain; most of them originating from other European countries (INE, 2018), while 14.6% of the Spanish GDP and 14.7% of employment came from tourism (WTTC, 2019). Moreover, the relevance of residential tourism for the construction and service sector has been widely acknowledged by scholars (Analistas Económicos, 1997; Llopis Vañó, 2017).

Interestingly, Spain's attractiveness as a tourist destination may not only represent a reasonable explanation for the government's inertia in discussing the compensation system but also promoted strategies aimed at attracting patient mobility in the Spanish private healthcare sector (Finotelli, 2021). In fact, the approval of the European Directive in 2011 and its ratifcation in Spain in 2014 has opened a window of opportunity to the private healthcare sector to attract European patients into Spanish private hospitals. Since its approval, the Directive was never intended to be understood as an instrument to encourage EU citizens to receive treatment outside their Member State of affliation. This notwithstanding, the General Secretary of the Union of European Private Hospitals (UEHP) welcomed the Directive as an important step to foster intra-EU mobility in the private healthcare sector (*iSanidad,* 2018). Likewise, the Spanish private healthcare sector immediately saw the Directive as a great chance to boost the travel of EU private patients to Spain by turning the country into nothing less than the worldwide standard for the international healthcare market. To achieve such an objective, the Spanish Federation of Private Clinics created in 2013 *Spaincares*, a private-public collaboration with major Spanish hotel chains and with the support of the Spanish Ministry of Tourism, which became operational in 2014. As the organisation's managers noted, the initiative did not focus on European patients already living in Spain, since this category of patients usually chooses the national (public) healthcare system for treatment (Finotelli, 2021). The Spanish private healthcare sector was instead interested in those patients who would explicitly move to Spain to receive the same treatment that they could not receive within a reasonable period in their home country. The "recruitment" of cross-border patients had nothing to do with the alleged misuse of the public healthcare system by EU temporary residents or tourists. As the general director of the Offce of the High Commissioner for La Marca España, Francisco Rabena, explained, "These are not people who come to be treated in the Social Security system, to increase the queues, to get the emergency infrastructure overcrowded, or to increase the trouble in fnding a bed. They are foreigners who come to Spain with their private health insurance, to private hospitals, with private professionals" (Márquez, 2018). Shifting the system of reference from the public to the private healthcare sector had clearly changed the perception of European citizenship and the therewith attached social rights. In something less than a decade, the public perception of intra-EU retirees as "healthcare scroungers" that burdened the public healthcare regime had been quickly substituted by the representation of the welcome European cross-border patient with benefcial effects for the private healthcare sector.

#### **13.5 Conclusion**

The debate on intra-EU mobility and welfare in Europe has shown how concepts of "undeservingness", "scrounging", "unfairness" or "welfare-transit" were not only used to describe bogus asylum seekers or irregular migrants *en route* to Europe but could be easily transposed to describe "unwelcome" intra-EU migrants from Southern and Eastern Europe. Yet, analysis of the Spanish case in this article shows that the restrictionist turn was not limited to South-North labour-motivated mobility but can be easily extended to the intra-EU non-labour motivated mobility in Southern Europe. This has been for instance the case of the restrictions to public healthcare for EU citizens which were introduced in Spain after the Great Recession of 2009. As Finotelli (2021, p. 615) argued in a previous study "Paraphrasing Maurizio Ferrera (2016), it can be argued that the Spanish central government has tried to reintroduce a 'modicum' of state autonomy on intra-EU mobility by applying 'more severely' the principle that EU citizens may reside in another Member State for a period longer than 3 months if they have suffcient resources for themselves and their families." Yet, such a restrictive turn was followed by the decision of the Spanish government together with the private healthcare sector to promote Spain as a healthcare tourism destination under the umbrella offered by the Directive on cross-border patients' mobility. Against this backdrop, the Spanish case shows that different approaches can be taken when the logics of the EU citizenship regime intersect with logics other than the "protection" of the public healthcare sector such as accumulation logics embedded in the economic and the tourist sector. So, the importance of Spain as a tourist destination has prevented to revise the reimbursement procedure for healthcare treatment provided in Spain to intra-EU retirees. Likewise, the approval of the Directive has promoted widespread euphoria to attract cross-border patients possibly keen to receive medical treatment in the Spanish private sector. Even though the objective to turn Spain into a medical destination par excellence has not been achieved to date (Finotelli, 2021), analysis of the Spanish case shows that the European citizenship regime as embedded in a complex interplay of regulations and practices where important imperatives other than that of guaranteeing equal access to healthcare play a role. The purpose to guarantee fair access to the public healthcare sector to EU citizens and natives was balanced by the will to promote the access of EU citizens to the private sector showing how the context of reference, in this case the public or the private healthcare system, can make a difference on how EU citizens are observed (and acted upon) within the same EU citizenship regime.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Part VI Asylum Policy**

# **Chapter 14 Welcome Culture and Bureaucratic Ambiguity: Germany's Complex Asylum Regime**

**Dietrich Thränhardt**

#### **14.1 Welcome Culture in the Crisis of 2015 and Afterwards**

During the European asylum crisis of 2015, Germany opened its doors. The country accommodated 890,000 asylum seekers in 2015 alone, and 706,852 people received approval for asylum in 2016 and 2017, out of 1,154,995 in the entire EU. Chancellor Merkel responded to the welcoming attitudes of an engaging majority of the German population and the media, after the agonising reports about smuggled refugees suffocated in an abandoned lorry in Austria, and the iconic picture of a little boy washed up on the Turkish coast. Merkel identifed with the opening, and became TIME magazine's Person of the Year. When she said, "*Wir haben so vieles geschafft – wir schaffen das*! (We have done so much, we can do that)," she evoked memories of earlier immigration waves, and the country's economic strength and social cohesion. Support came from all walks of life: church communities, student groups, schools, retirees, business groups, and trade unions. Volunteers spontaneously collected and provided food, blankets, and children's toys, as well as practical and emotional support. Surveys showed that a record of 46% of the population

Due to the contentious nature of asylum policies, I have relied on offcial government documents, as well as on many NGO publications and press reports, often based on interviews with refugees, volunteers, or offcials, both open and confdential. The most informative sources have been parliamentary inquiries, with detailed questions; these require the government to gather and present statistical information that sheds light on administrative processes. One member of the opposition in the parliament (*Bundestag*) who specialises in these issues has extracted hundreds of pages of information, following the events year by year. My role has been to check and double-check all this information, systematise it, and defne the characteristics that form a "regime". Since the activation of temporary protection for displaced Ukrainians, we can speak of two regimes, one for Ukrainians, and another for all other asylum seekers and refugees.

D. Thränhardt (\*)

University of Münster, Münster, Germany e-mail: thranha@uni-muenster.de

contributed in some way, and only 18% answered that they would not want to contribute at all (Ahrend, 2017). Elderly women remembered the harsh times after being expelled from their homes in the territories Germany lost in 1945, and wanted to help out of their own experience. Against the ever-present memory of the Nazi past, this was a kind of positive redemption (Freedland, 2015). Despite setbacks, disappointments, and terrorist attacks, the welcoming spirit is still alive among many volunteers. The Russian aggression in Ukraine triggered a new wave of aid in 2022, with hundreds of thousands of Ukrainians fnding shelter in private homes.

The arrival of refugees overshadowed all other public issues from August 2015 until 2020 (graph in Fachkommission, 2021, p. 87). This "welcoming spirit" was omnipresent in 2015, but the public discourse soon became polarised, and the xenophobic *Alternative für Deutschland* (AfD) party has been voted into all state diets (*Landtage*) as well as into the federal parliament (*Bundestag*) since 2015. Even though the general public has continually come out in favour of helping refugees to fee from war and oppression, doubts about security – and particularly young male refugees – became more widespread. The Christmas market attack in Berlin in December 2016 and other murderous events, as well as right-wing arson attacks, stoked fears. After the Cologne rape events on New Year's Eve 2015–2016, the media began to speak about a change in the public climate. Open conficts between Chancellor Merkel and Horst Seehofer (Leader of the CSU – the CDU's sister party – and Bavarian Minister-President) about limits on the number of refugees and about border controls affected the government's reputation from 2016 to 2018. Since Seehofer's warning about a "reign of injustice" (*Herrschaft des Unrechts*), a "myth of illegality" was in the air with respect to the decisions Merkel made regarding refugees in 2015 (Detjen & Steinbeis, 2019). Despite this open confict, Seehofer served as Federal Minister of the Interior from 2018 to 2021, which meant that he was responsible for asylum matters. In March 2020, public attention shifted entirely to the Coronavirus pandemic. Since February 2022, the Ukraine crisis has led to a new nationwide "welcoming spirit," but this time it has been far less controversial, encompassing all political strata.

#### **14.2 Quality Problems in the German Asylum Decision System**

When Chancellor Merkel concluded in March 2018 that, all in all, the exceptional humanitarian situation of 2015–2016 had been well handled (*gut gemeistert*), she was right with respect to the work of local governments, state governments, and volunteers. Together, they were able to accommodate the refugees, cooperating effectively with volunteers and charitable organisations. With the exception of a few cases in Berlin, nobody had to live on the streets, and no informal settlements sprang up.

Contrary to public expectations, the Federal Offce for Migration and Refugees (BAMF) emerged as the weakest actor in the crisis. It was hampered by a lack of resources and poor organisation, and could not even register half a million applicants in 2015. The failure to decide about asylum applications in time affected all other integration and welcoming activities. "All day waiting" (Christ et al., 2017) frustrated the ability of refugees, volunteers, and employers to be more productive, led to conficts in refugee centres, and delayed integration. The government reacted by opening language courses and allowing work before an asylum decision for Syrians, Iraqis, Iranians, and Eritreans, and later for Somalis as well (Table 14.1).

The BAMF President, Manfred Schmidt, resigned in September 2015. The new BAMF leadership accelerated the decision-making process, and tripled the agency's personnel between 2015 and 2020. The BAMF became the largest asylum agency in the world, and was lauded as the "asylum world capital" by Gerald Knaus, the mastermind of the EU-Turkey Agreement of 2016 (Knaus, 2020, p. 156). Its bureaucratic coherence has been re-established since 2018. However, the quality problems did not disappear, despite lower numbers of incoming refugees.

Judges became critical about the ineffciency of the asylum decision process. In a parliamentary hearing in 2019, Stefanie Killinger, the president of an administrative court in Lower Saxony, complained about the "inferior quality" of asylum decisions in Germany. Sending "unclear signals of an asylum lottery", German authorities "play with the hopes and expectancies and in extreme cases even with the life of people who take part in the lottery," she testifed. Together with the good chances to remain in the country even after a negative asylum decision, this creates, she argued, a "pull factor" – "everybody who comes here sees himself as a potential winner in the lottery." Backlogs at the BAMF and at the courts lead to long years of waiting, resulting in good chances to stay in the country even if all legal means are exhausted. Thus, Killinger summarised, "we are situated in the worst of all worlds" (Innenausschuss, 2019, p. 10).

In the same hearing, a renowned lawyer with 40 years of experience told the Bundestag that in more than half of his cases, the applicant had not been presented with the objections that had led to the negative verdict, so that they could have been clarifed. This had been a systemic fault for decades, he argued. Improvement was not in sight, and the agency should not hand down such poor decisions (Innenausschuss, 2019, p. 19). He added that BAMF decision-makers often apologised to him for not being suffciently prepared and not having read the fles.


**Table 14.1** Average duration of asylum procedure in months, 2015–2020

a In 2015, about 500,000 asylum seekers had not even been registered. This was rectifed in 2016 and 2017. Source: BAMF yearly reports

Hearings were scheduled shortly after asylum applications, without appropriate preparation. As a consequence, lawyers could not adequately represent the applicants. Asylum offcials often had to decide on applications from many countries of origin, rendering them unable to obtain country-specifc expertise.

Statistical evidence supports the criticism concerning the malfunctioning of the German asylum system. In 2020, judges invalidated 31.2% of administrative asylum decisions. This was up from 26.4% in 2019, 31.4% in 2018, and 8.0% in 2017 (BT-Drs. 19/28109). On average, asylum decisions took 8.3 months in 2020, and court decisions 24.3 months (BT-Drs. 19/28109). At the end of 2020, 129,320 people were still waiting for a fnal court decision. In the end, most asylum seekers stay, and the number of deportations to non-European countries is low. In 2020, only 755 people were deported to African countries and another 675 to Asian countries (BT-Drs. 27,007, pp. 3–4). The long administrative and judicial processing times hampered integration, and the country spent €4.2 billion on accommodation and welfare benefts for people with an undecided status in 2020.

The problematic quality of asylum decisions was also obvious regarding the profound discrepancies between local BAMF organisation units. The results differed astoundingly for people with the same nationalities (see Table 14.2).

These discrepancies have been brought to public attention by parliamentary scrutiny in the last few years. The BAMF claimed that they had introduced oversight and quality control, but the discrepancies did not disappear. Understandably, three-quarters of asylum seekers who were denied asylum appealed to the courts (see Table 14.3). Even those with poor chances had reason to appeal, since they could remain in the country as long as court proceedings continued.


**Table 14.2** Recognition rates in local BAMF organisation units in 2020 (%)

Source: Deutscher Bundestag, Drucksache 19/28109, Question 3

**Table 14.3** Asylum seekers appealing against rejection (%)


Source: Deutscher Bundestag, Drs. 19/8701, Drs. 19/28109

#### **14.3 British and Italian Parallels and the Common Tension Between Asylum Principles and the Political Will to Reject Asylum Claims**

Looking into media reports about procedures in the United Kingdom, we fnd similar or worse problems. Again, the term "lottery" has been used to characterise the outcomes. And again, the reason given is the low quality of asylum decisions, due to the pressure to decide cases under time constraints. One BBC reporter at the evaluation units was reminded of the "atmosphere of a call centre" (Brewer, 2018). Offcials relied on "copy-paste" methods, and decided on cases having only limited knowledge of the specifc circumstances (Brewer, 2018; Lyons & Brewer, 2018). The Home Offce itself spoke of the need "to fx the broken asylum system so that it is frm and fair" (The Guardian, 12 November 2020). Similarly, limited administrative capacities have led to a backlog. Since the UK has had to deal with a rather low number of asylum requests, the shortage of personnel cannot be attributed to overburdening, but is a result of political decisions. British authorities decided 28,460 cases in 2019, while 60,548 applicants were still waiting for a decision in 2020. Thus, the waiting time is more than 2 years on average (Eurostat, 2021; Taylor, 2020).

Insofar as the UK and Germany both rank high in the World Bank Government Effectiveness Index, systemic defcits over decades cannot be explained by an inability to organise proper governance. Instead, the problem lies in the ambivalence between the legal principles of asylum and the political desire to contain infows. This is widely expressed in the political discourse in most European countries, and is a source of motivation for many governments (Thränhardt, 2019). Asylum agencies then come under pressure to produce a low percentage of positive asylum decisions. This pressure cannot be formulated in offcial texts, because this would be illegal and invalidated by the courts. The ambivalence is then present within the agencies, as it would be illegal to order agents to break the law. As a result, politicians like Austria's former Chancellor, Sebastian Kurz, can praise the Australian model of incarcerating and deterring refugees, but they cannot implement it. Contradictory signals about acting according to the law and simultaneously reducing numbers leave asylum offcials in limbo, and these offcials react with a broad range of different decisions. The authorities then turn a blind eye to the quality of the decisions. In the end, it is the courts that have to deal with the mess.

While the deciding bodies or individuals are theoretically free in their decision, bound only to the law, in practice we fnd pressure from the interior ministries of these countries. In Germany, we can follow this policy with respect to refugees from Afghanistan. In 2015, Minister of the Interior Thomas de Maizière was concerned with the high numbers of asylum seekers from Afghanistan. He wanted to "send a signal": "Stay put. We will repatriate you" (Lehnert, 2021). He added that only a few Afghans had the right to asylum, in contrast to the fact that 86.1% had actually received asylum in the third quarter of 2015 (BT-Drs. 18/6860, Question 1b). In turn, the coalition parties concluded that they wanted to create and improve intra-Afghan possibilities of refuge, and revise and adjust the BAMF's decision principles to follow this line. BAMF guidelines were adapted, and young men were expected to look for internal protection in Afghan cities. The courts were divided concerning the cases, but they granted asylum more and more frequently (Lehnert, 2021). In 2020, the BAMF lost 60% of the Afghani cases decided by the courts – a record percentage (BT-Drs. 19/28109, Question 22e). The BAMF, however, stuck to its line, even though it had lost a majority of court cases. As a law-abiding institution, it should be expected to follow the arguments laid down by the courts, and change its practices. However, the confict continued, at the expense of the taxpayer. In the end, most Afghans got asylum through the courts, including via judgements of the German Constitutional Court and the European Court of Justice.

The Territorial Asylum Commissions in Italy serve as a parallel example. The European Council on Refugees and Exiles (ECRE) has described the relationship as follows: "These bodies should be independent in taking individual decisions on asylum applications but, due to their belonging to the Department of Civil Liberties and Immigration of the Ministry of Interior, in more cases, they received instructions from the Ministry of Interior. Some examples are the instructions given for the grounds of inadmissibility, manifestly unfoundedness, border procedure" (AIDA, 2020, p. 32). In 2017, Italy was the last country to discontinue the tradition of having a UNHCR representative prepare asylum flings for the asylum commission. Now, this task has been assigned to an offcer of the Ministry of the Interior.

Such institutional ambiguity explains how the decision lottery exists not only in countries with weak government effectiveness, but also in those with strong ones. We can demonstrate these effects using Germany's BAMF. Between 2015 and 2019, the BAMF was technologically upgraded and personnel was tripled. Asylum decisions, however, were still ill-organised, and personnel was shifted to reviewing procedures against accepted refugees. All cases were checked again after 3 years, with very few negative decisions that again are challenged in court.

In this contentious atmosphere, ministries and asylum agencies often operate in a secretive style, and it is then up to journalists, parliamentarians, and researchers to shed light on the black box. In Germany, regular detailed parliamentary inquiries force the government to present information, often over several hundred pages long (for Austria, Diakonie, 2019; for Britain, Grierson, 2020; for Spain, Janker, 2020). All in all, these patterns and attitudes create a common regime in most Western European countries concerning the ambiguity of asylum decisions. In the UK, the problems were more intense because of a radical reduction of administrative capacities. In the last Eurostat statistical summaries that included the UK (in 2019), the country had a record success rate of 70% of court cases won by asylum applicants.

#### **14.4 The Policies of Backlog and Encampment**

Backlogs in asylum processes are as ubiquitous as government proclamations about speedy trials. At the height of the asylum crisis in 2016, the German government made a systematic and illuminating argument for speedy decisions:

The asylum procedures are too long, lasting six months on average. Therefore, persons affected live in insecurity about their further fate for too long. Those whose applications are decided positively in the end and are allowed to stay in Germany are offered integration activities relatively late. They then need a fairly long time until they can join the labour market. But even for those who have to wait long for a negative decision, the duration of the procedures hinders a return to their countries of origin. Particularly children, who integrate faster, participating in school education, can then be cut off from an environment with which they just have become familiar. Not least because of this, after a negative decision, the likelihood of exceptional leave to remain (*Duldung*) increases with the length of stay. This then requires resources that are needed for accepted people in need of protection. (BT-Drs. 18/7203).

The EU has set the same tone for years: in 2010, the Commissioner for Home Affairs, Cecilia Malmström, argued that "the best way to ensure effciency and fght against so-called abuses is to invest in an asylum process which provides robust and qualitative decisions in a rapid manner" (European Commission, 2010). Three years later, Article 31(2) of the Directive on Common Procedures for granting and withdrawing international protection (2013/32/EU) stated: "Member States shall ensure that the examination procedure is concluded as soon as possible, without prejudice to an adequate and complete examination." In the introduction to the European Commission's "New Pact on Migration and Asylum" in 2020, the Commission declared that "The frst pillar of the Commission's approach to building confdence consists of more effcient and faster procedures" (European Commission, 2020).

Thus, procedures should have been processed more quickly when the number of asylum applications declined and personnel tripled in Germany. However, procedures became even slower. As explained above, the average processing time in Germany was 8.3 months in 2020 (BT-Drs. 19/28109). Governments of wealthy countries with well-functioning bureaucracies still allow drawn-out, poorly handled procedures to occur, while at the same time arguing for faster processing.

Over the last few years, the character of many asylum centres has changed, from an admission site to a long-time encampment, and an environment of repatriation pressure. EU rules for the length of holding people in centres have been loosened. In Germany, the length has been extended sixfold, from 3 months to 18 months, and even to 24 months in some states. In the end, however, most occupants stay in the country, and are transferred to normal local accommodation. Some are accepted via court decision after years of waiting; others are unable or unwilling to return to their home countries and receive exceptional leave to remain (*Duldung*) (González Méndez de Vigo et al., 2020).

Waiting, uncertainty, backlogs, and overcrowding of admission facilities are not limited to Germany, but are a frequent feature of asylum reality in Europe – not only in countries and situations of sudden infows, but also in environments where planning and organising is possible. At the end of 2020, Eurostat counted 765,700 unresolved asylum applications in the EU, with 257,200 or 33.6% of them in Germany (Eurostat, 2021). Whether such inadequacies are the result of deliberate inaction, neglect, or incompetence is often not easy to investigate. In Germany, the 2018 coalition agreement envisaged large "AnkER" centres (arrival, decision, repatriation – *Ankunft, Entscheidung, Rückführung*) with the aim of speedy decision-making and direct departure of those who were not granted asylum. In the end, however, the decisions in the new centres took even longer than those in traditional settings (BT-Drs. 19/25435, p. 56), and people had to stay in the AnkER centres for increasingly long periods. Most of the centres are former military barracks, fenced in and monitored. The new 2021 coalition agreement stated that the AnkER concept would be discontinued.

These long delays in crowded camps are harmful for the physical and mental health of refugees as well as for their trust in the system, and for their integration. This is particularly true for children who are held there; they do not live in an appropriate environment and are denied a standard school education (González Méndez de Vigo et al., 2020). Employment bans linked to encampment also delay the economic integration of refugees. One empirical study found that "longer employment bans considerably slowed down the economic integration of refugees and reduced their motivation to integrate early on after arrival".

The media and activists in Germany, however, were much more concerned with the deplorable conditions on the Greek islands. The EU-Turkey Agreement did not work, largely due to the failure of the Greek government to carry out asylum procedures (Knaus, 2020, 190–199). Centres developed into long-term detention camps, which were overcrowded and degrading (see the contribution on Greece in this volume).

All in all, there is a broad consensus that the current asylum regime is defcient. It is expensive, but delivers poor results; it spoils the good will of the population and saps the energy of millions of volunteers who want to assist refugees. Most importantly, it holds millions of refugees in limbo and forces them to concentrate their activities not on integration and success in the host societies, but on ways to get past the hurdles that states build up to keep them out. Not only are the camps on the Greek islands expensive for taxpayers, but the long duration of camp life in Germany and other European countries is as well; it is demotivating and sometimes dehumanising for refugees, it delays the integration process, and it is ultimately futile for the asylum procedure. Large, permanent camps are public displays of crisis, incompetence, and unwillingness.

#### **14.5 Best Practices in Europe and the Reluctance to Optimise**

In 2015, while Germany was discussing the shortcomings of the asylum system, EU Vice-President Frans Timmermans told a German television station on 9 September that asylum procedures took 8 days in the Netherlands and 8 months in Germany, and this explained why so many Balkan refugees– a group with slim chances of getting asylum – came to Germany rather than to the Netherlands (Thränhardt, 2016a, b). In a search for solutions, the Bertelsmann Stiftung then published reports about the Swedish, Dutch, and Swiss asylum procedures (Parusel, 2016; Thränhardt, 2016a, b). Bertelsmann wanted to explore if Germany could learn from effcient practices in other countries. This next section takes up this question as well, and discusses the results.

In 2010, the Netherlands had found a way to organise fast, high-quality asylum decisions: the "improved procedure". A strict timetable is set, combined with statefunded legal representation, so that the applicant's asylum plea can be fully presented by a legal expert. Lawyers apply for each client. The assigned lawyer meets the asylum seeker before the hearing, prepares the client, and advises him or her, takes part in the hearing, comments if necessary on the draft opinion before the asylum decision, and can appeal to the courts in case of a negative verdict (Thränhardt, 2016a, b). The system worked speedily, and experts presented the system as a successful model to the outside world.

Later, however, the Dutch government cut administration personnel, so asylum seekers now face long waiting times before the "improved procedure" begins. The 2017 Dutch coalition agreement called for the provision of legal assistance only "after notifcation of intent to deny an asylum application has been issued." The coalition wanted to "eliminate gold-plating" and reduce the level of asylum standards. It is, however, doubtful that eliminating the "improved procedure" would "reliev[e] pressure on the justice system," as the coalition agreement (p. 52) asserts. Clearly, the 2017 Dutch coalition government mistrusted their effcient asylum system, and chose to focus on "detention" – a catchword in the coalition agreement.

Switzerland adopted the Dutch combination of speedy procedures and quality decision-making as well. Their "structured model" was discussed intensely among all levels of government and the public, and tested in Zürich from 2014 on. In a referendum on 5 June 2016, 66.8% of Swiss voters accepted the new system, and it was introduced across the country on 1 March 2019. After decades of strife, the new asylum procedure has been accepted by the public and has moved out of the headlines. Switzerland invests in the quality system. Charitable organisations provide legal representation, and receive a fxed remuneration for each client, with contracts over 5 years. Switzerland provides 5000 places in six federal asylum centres, even when the number of asylum seekers is low.

The "*Taktphase*" is organised like Swiss clockwork. After the asylum seeker has gone through medical and security tests, and Switzerland has been identifed as the country responsible, the legal representative meets the client to discuss the case. This representative is present at the hearing, can assist the client, and can make representations. Next comes "triage": the decision either to follow the structured model, or to switch to the extended procedure if the case is complex and further evidence is required. In the structured model, the asylum offcial then produces a draft asylum decision, and the legal representative has 1 day to comment. The offcial then writes the defnitive verdict. In case of a negative decision, the representative has 7 days to appeal to the Federal Administrative Court. The Court announces its judgement in 20 days, or it returns the case to the asylum administration if it identifes shortcomings.

The legal representative can bring the client's individual story into legal arguments. Through the whole process, the presence of a representative challenges the asylum offcial to pursue the case carefully and provide a well-founded decision. Both the offcial and the legal representative have access to training, data systems, and exchanges with colleagues. Legal counselling offers asylum seekers information and orientation, instead of rumours and fear. The system is designed to investigate thoroughly, and to come to a well-founded decision.

After a year's experience with the new system, there was a broad consensus that it worked well. The average asylum procedure took 50 days; 35 for Dublin cases, 50 in the structured model, and one hundred days in the extended procedure (SEM, 2020). Charitable organisations wanted some more time and fexibility, particularly for vulnerable clients, and ten instead of 7 days to prepare the appeal. An external review by the Swiss Center of Expertise in Human Rights was positive (Evaluation PERU, 2021). The state secretary argued that the new system offers asylum speedily for those with good cause, and discourages those without. The acceptance rate is high, but the quick process discourages those who stand little chance of getting asylum. The Swiss example demonstrates that a fast and fair asylum procedure can be organised, and that growth in public trust corresponds with effectiveness and open communication.

In 2017, Germany's BAMF organised a pilot project, imitating the Swiss procedure, in cooperation with Caritas, Diakonie, and the German Red Cross. The BAMF's internal evaluation was extremely positive, with respect to "improved administrative effciency", "better clarifcation of the facts", "asylum applicants' improved understanding", and a "better quality of asylum decisions". The "effciency of the agency's procedure" had also been improved. Five applicants had withdrawn their applications when advisers explained to them at an early stage that they had no chance of getting asylum. Deciders reported that planning became easier. The clear advice helped asylum seekers to understand the asylum system, and made them less dependent on rumours and smugglers.

In a "political decision", however, the Federal Ministry of the Interior overruled the BAMF specialists, and ordered the BAMF to keep the report secret (BT-Drs. 19/19535, p. 12). Later, it was leaked and published by the Refugee Council of Lower Saxony (*Flüchtlingsrat Niedersachsen*), an NGO (Evaluation, 2018). The coalition of Germany's major charities commented: "When refugees are able to access independent and free asylum procedure counselling, this improves the due process, fairness, quality, and effciency of the asylum procedure. Counselling does not decelerate it and helps the BAMF to identify persons with special protection needs (such as traumatised, sick, and disabled persons). This is our experience from the pilot project" (BAGFW, 2019).

Instead of working with charity advisers, the BAMF organised an internal counselling system, with a number of asylum offcials put in a new BAMF division, rotating with their colleagues. Charities can also offer counselling in parallel if there is state funding for it. However, the counselling is not comprehensive, differs among states, and does not include legal assistance, the core of the Swiss improvement (Thränhardt, 2020). It is an expensive but ineffcient arrangement. The new coalition agreed in 2021 to reintroduce independent asylum counselling, but it has not yet been formalised in law and implemented.

In Austria, all NGO counselling has been terminated, and the government has established a special "Federal Agency for Assistance and Support Services" (*Bundesagentur für Betreuungs- und Unterstützungsleistungen*). The intention was to cut off any civil society infuence, and – in the words of the former Minister of the Interior – "to tell many people: that you have no chance" (Kickl OE 24). Shortly after the legislation passed, the minister and his party, the FPÖ, were ousted due to a scandal, and the implementation was modifed under the subsequent conservativegreen coalition. From 1 January 2021 on, all counselling has been provided by the new state agency, independent in its work and bound to confdentiality (Diakonie, 2019; Die neue Volkspartei, 2020; Gahleitner-Gertz, 2020). In Austria as well as in Germany, asylum procedures still take a very long time, and the decision quality is insuffcient (Lahner, 2020, Table 14.2). Clearly, both governments, like many others, were more concerned with reducing the number of asylum seekers than with optimising the quality of the procedure. However, as long as there is an independent judiciary, these measures will fail to bring down numbers, but will continue to generate enormous costs, not only fnancially, but also in terms of public trust and refugees' quality of life.

#### **14.6 Conclusion: Administrative Ambiguity in an Integrative Asylum Regime**

In Europe, we fnd stronger and weaker regulators, and small and large asylum bureaucracies. What is in common, however, is the organised ambiguity towards the acceptance of refugees. This can be as simple as in Greece, where the Mitsotakis government attempted to abolish the Migration Ministry in July 2019, but then reinstated it after 6 months; this slowed and then spurred asylum procedures, all the while holding people in inhuman environments (ANSA, 2020). In Germany, with the largest asylum agency in the world, bureaucratic arrangements are organised in such a way that people wait for a decision in centres, hearings are organised in a fuzzy way, counselling is disorganised, and the courts have to resolve and correct a vast number of cases. In both countries, and in many others, the procedures are characterised by ineffciency, deliberately holding asylum applicants in limbo. On the other hand, the EU itself – as well as many European governments, including Germany – fund the humanitarian organisations and activists who support refugees and often distrust these agencies. Both sides frequently argue that the system is broken.

Partially an element of Cold War hypocrisy (Martin, 1989), showing the moral superiority of Western democracies against the Communist foe, the principles of asylum and particularly *non-refoulement* became part of an established international treaty system, nominally accepted by most countries in the world. In the European Union, these principles were codifed in an elaborate system of EU directives, to create an "area of freedom, security, and justice" for refugees. In her

seminal work, Natascha Zaun (2018) has shown how "strong regulating states" were successful in transferring their level of asylum administration into EU directives, while "weak regulators" cared little. Strong regulators were careful to include a right to legal representation, but no provision to fund and organise it. Thus, many asylum seekers are unable to present their case convincingly, and states allow for a fawed asylum decision process.

However, the asylum regime is much more than its bureaucratic practice. Eurostat's data on "fnal decisions" show that the courts grant asylum for an impressive percentage of the claimants that had been previously turned down by the asylum administrations. In Germany and Austria, the absolute numbers of these "fnal decisions" are almost as high as frst instance decisions: 100,100 against 128,600 in Germany, and 10,000 against 10,500 in Austria. In the EU as a whole, it is 232,800 against 521,000 cases. Eurostat (2021) reported that in 2020, 61.7% of asylum appeal court cases in Austria and 35.5% in Germany were decided in favour of the refugees. A system of judicial expertise has also developed at national and transnational level. This judicial system is complemented by a network of civil society organisations that had become energised and expanded during the 2015 asylum crisis, and organised into NGO structures, with links throughout various sectors of society. Public opinion is divided, moving in this or that direction after revelations and scandals, but in principle supportive of asylum. All in all, we can conclude that Germany has a receptive asylum regime, even if the practice of the responsible ministerial and administrative agencies remains ambivalent. The welcoming parts of the regime, however, are largely reactive, against the restrictive tendencies of the Ministry of the Interior. This makes it protracted, confrontational, and expensive, and hinders a proactive and coherent asylum policy.

#### **14.7 Postscript: The New Regime for Displaced Ukrainians – A Blueprint?**

Reacting against the Russian aggression in Ukraine, the EU activated the Temporary Protection Directive for displaced persons. In doing so, they introduced a second refugee regime alongside the existing one. Ukrainians are free to choose the country of refuge, they can return to Ukraine and come back again, and they can work and study. Temporary protection functions smoothly and without the toxic conficts between EU governments and within countries that had erupted in 2015 (Thränhardt, 2022). Ukrainians do not go through the asylum bureaucracy, and most are housed by ordinary citizens, relieving the burden on the states. It is an open question if the experience of an alternative regime in the same area will lead to reforms of the asylum system, thus making it more effcient, more humane, and more productive. Reem Alabali-Radovan, Minister of State at the Chancellery and Federal Commissioner for Migration, Refugees and Integration, has called it a "blueprint for what is possible if everyone cooperates" (FAZ, 2022).

#### **References**


World Bank Government effectiveness index. (2010, update 2017). World Bank Government effectiveness index, 2010 — European Environment Agency (europa.eu).

Zaun, N. (2018). *EU asylum policies: The power of strong regulating states*. Palgrave Macmillan.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 15 Looking Into Policy Change: How the Italian Asylum Regime Came of Age**

**Irene Ponzo**

#### **15.1 Introduction**

In the feld of asylum, Italy has generally been depicted as a latecomer and recalcitrant recipient of the EU legislation (Trauner, 2016; Zaun, 2017). Indeed, for a long time Italy accepted requests to adapt its asylum policies to European standards with only the aim of being included in the Schengen Club without sharing the policy frames and goals underlying those requests.

In this contribution, I argue that the European refugee crisis represented a turning point in this regard, since in this contingency Italy began to share some fundamental traits with older European asylum countries. From the early 1990s to the "North African Emergency" declared in 2011, every new refugee infow triggered the dilemma of whether or not to provide protection and assistance, whereas the public debate currently revolves around the quality of protection and reception which are regarded as due. Moreover, secondary movements, which had been encouraged for a long time, have been drastically curbed by the adoption of the hotspot approach in 2015 and the consequent systematic fngerprinting of people who irregularly cross the borders. Finally, responsibility-sharing, the traditional concern of Northern European countries, is now at the core of Italian claims with regard to asylum.

In order to understand this policy change, I will analyse the development of policy frames and practices over the last three decades. Indeed, as illustrated in the volume Introduction (see Chap. 1), we can conceive of a migration regime as a mix

© The Author(s) 2023

I. Ponzo (\*)

FIERI (International and European Forum of Research on Immigration), Torino, Italy e-mail: ponzo@feri.it

C. Finotelli, I. Ponzo (eds.), *Migration Control Logics and Strategies in Europe*, IMISCOE Research Series, https://doi.org/10.1007/978-3-031-26002-5\_15

of ways of both observing and acting on migration (Cvajner et al., 2018), in other words as a mix of frames and practices concerning migration phenomena.1

Specifcally, I will argue that, after two decades of stability, practices and frames started to change in 2011 as a result of modifcations in the social phenomenon of migration and in the institutional settings where negotiation between Member States occurs. In contrast, the country's weak political-institutional capacity has slowed down this process, undermining the consolidation of the new practices. Moreover, this weakness appears as largely responsible for the gaps with older destination countries in the management of asylum whereas divergence in policy frames and goals has lost relevance.

Using a biological metaphor of human life, Sect. 15.2 reconstructs the "infancy" and "puberty" of the Italian asylum regime since the early 1990s; Sect. 15.3 illustrates its "adolescence" after the outbreak of the Arab Spring; and Sect. 15.4 explains its "coming of age", which occurred during the European refugee crisis. Section 15.5 focuses on the limited national political-institutional capacity as a key explanatory factor for some persisting weaknesses of the Italian asylum regime.

The reconstruction of the events that occurred in the 1990s and the frst decade of the 2000s (see Sect. 15.2) is based on secondary evidence, i.e., scholarly accounts supplemented by analysis of offcial documents. The analysis of the subsequent developments (see Sects. 15.3, 15.4 and 15.5) has been drawn from the research conducted within FIERI.2 Specifcally, it relies on the Horizon2020 projects CEASEVAL and TRAFIG where 53 semi-structured interviews with stakeholders and experts were conducted between 2018 and 2022.

#### **15.2 From Infancy to Puberty: The Emergence of the Italian Asylum Regime**

For a long time, in Italy the arrival of refugees was framed as a marginal phenomenon and gained little attention in policymaking (Marchetti, 2014; Campomori, 2016). This may sound odd, given that its coast represents a substantial part of the European Mediterranean border. Yet, the frst arrivals of Albanians by sea in 1991 took the country by surprise since "boat people" were perceived as a rather exotic phenomenon circumscribed to developing countries (Hein, 2010). Indeed, the maritime coast had been an irrelevant border for a long time given that Italy mainly played a key role as a channel for resettling refugees from the East to the West rather than a frontline state in the management of refugee infows. Following a rather institutionalised practice, from the end of World War II to the fall of the Berlin Wall,

<sup>1</sup>Drawing from the seminal work of Hall (1993) on policy change we could regard the change of regulation and practices as a modifcation of policy instruments (second-order change), and the change of policy frames as an alteration of the hierarchy of policy goals (third-order change), with the latter possibly leading to a shift in policy paradigm.

<sup>2</sup>For further info see https://trafg.eu/ and http://ceaseval.eu/

Italy was employed for two-step resettlements: around 220,000 individuals reached Italy from the Soviet Union and were resettled to other countries, mainly North America and Australia, while in the 1990s an agreement between Italy and Israel allowed thousands of Jews in transit to be transferred to the United States (*ibidem*).

This particular history of asylum probably contributed to framing the arrival of refugees as a transient phenomenon even when it was no longer the case. Against this backdrop, in the 1990s and 2000s European pressure and norms over asylum deeply affected Italian legislation without signifcantly impacting policy frames and actual practices. For those two decades, legislation changed deeply but Italy went on perceiving itself as a transit country and framing refugee fows as emergencies. Consistently, policy practices developed around three pillars: fostering secondary movements of asylum seekers towards other countries; keeping the asylum system (commissions for processing asylum claims, reception facilities, etc.) underdeveloped so that ad hoc emergency solutions and bottom-up initiatives became the ordinary ways to cope with refugee infows; opting for national form of temporary protection, convertible into a work, study or family permit in order to limit the number of benefciaries of international protection and the relative due assistance.

#### *15.2.1 Infancy, When You Think That Problems Will Fade Away: The 1990s Emergency Approach*

Section 10 of the Italian Constitution states that "a foreigner who, in his home country, is denied the actual exercise of the democratic freedoms guaranteed by the Italian Constitution shall be entitled to the right of asylum under the conditions established by law". Yet, this constitutional right has never been transposed into law: Laws 722/1954 and 95/1970, which enacted, respectively, the Geneva Convention and the Protocol of New York, were the only pieces of legislation on asylum passed by the Italian Parliament until 1990, when Law no. 39 was adopted.

Although the killing of the South African refugee Jerry Masslo in 1989 was a powerful trigger (Colucci, 2018), "the rationale behind the Act 39/1990 was part of the Schengen process, as a way to both implement some requirements for membership and to reassure 'old' immigration countries like Germany and the Netherlands that Italy was indeed able to prevent entry of unwanted migrants into the Schengen space" (Finotelli & Sciortino, 2009, p. 123). Since one of the requirements for entering the Schengen Club was repealing the so-called "geographical limitation" to the Geneva Convention which had allowed Italy to accept as refugees only citizens from European countries, this provision was included in Law 39/1990. In addition to that, Law 39/1990 introduced some procedures for recognising refugee status and set up the Central Commission for the Recognition of International Protection that included offcials from the central government who performed that task as an additional, non-exclusive assignment, and were complemented with experts from UNHCR with consultative functions.3 No specifc material support was foreseen except for small economic allowances (around 18 euros per day) provided during the asylum procedure for a maximum of 45 days. The underdevelopment of the asylum system, embodied in the nonprofessional staff of the Commission and the lack of material support, would remain a persistent trait of the Italian asylum regime for a long time.

The repealing of the geographical limitation together with the collapse of the Soviet Union soon generated a series of subsequent infows of refugees while the resettlement programmes of North America and Australia abruptly ceased, given that refugees from Eastern Europe lost the political relevance they had had during the Cold War (Hein, 2010).

As a consequence, the new Law found an immediate application when, in 1990, Albanian refugees brought to the country on Italian boats went through the asylum procedure (Bona & Marchetti, 2017). Yet, this was an exception rather than the rule. The refugee infows that followed throughout the 1990s were mainly managed by adopting an emergency approach and ad hoc measures rather than by applying Law 39/1990: people feeing conficts, when not immediately returned, were generally granted renewable temporary protection. This solution was partially aimed at managing those cases where it was assumed that the Geneva Convention, and then Law 39/1990, could not be applied because people were feeing general violence instead of individual persecution (Vincenzi, 2000; Campomori, 2016). At the same time, it allowed for speeding up the procedures and avoiding the provision of refugee status and the consequent assistance due (Bona & Marchetti, 2017).

Against the backdrop of recurrent exceptional measures, the solutions adopted to cope with the almost 80,000 displaced people who began arriving in 1992 from the former Yugoslavia produced long-term consequences for the Italian asylum regime. Since the collective centres set up by the state offered only 2000 places and maintained rather low-quality standards, civil society started to mobilise to host people in spaces made available by local authorities, parishes and private citizens under the coordination of local committees which were also responsible for fundraising. This mobilisation was then partially institutionalised: Law 390/1992 foresaw some form of coordination with the state through consultations with local authorities, NGOs and voluntary associations. Moreover, agreements were signed between Prefectures, i.e., the local branches of the Ministry of the Interior, and local administrations to cover at least part of the accommodation expenses (*ibidem*). This experience deeply shaped the subsequent development of the Italian reception system by determining a bottom-up, voluntary-based imprint almost incompatible with the mandatory distribution of asylum seekers which indeed has striven to emerge.

The other main piece of legislation concerning reception passed at that time was Law 563/1995 (the so-called Apulia Law) which, for the frst time, introduced reception centres to provide frst assistance to migrants in view of their

<sup>3</sup>Before 1990, the decisions on refugee status were made by the Cpe (*Commissione paritetica di eligibilità*) established in 1952 and made up of representatives of the Ministries of Interior and Foreign Affairs and of UNHCR (Hein, 2010).

identifcation and possible repatriation (CDA).4 Although the Law referred to a defned time span (1995–1997) and territorial area (namely Apulia where three centres were indeed set up) it has remained for a long time the main piece of legislation regulating frst reception centres, including the hotspots introduced in 2015 in the face of the European refugee crisis (Giannetto et al., 2019), revealing the longstanding underdevelopment of the Italian asylum policy.

#### *15.2.2 Puberty, When the Adults Ask You to Be Responsible: Towards the Setting Up of Ordinary Measures*

At the end of the 1990s, Italy started to develop some ordinary measures that represented the frst steps on the path out of a totally emergency-based approach relying on ad hoc solutions. Still, as with Law 39/1990, those changes were largely driven by European pressures (Paoli, 2018). Indeed, the abolishment of controls at sea and land borders between October 1997 and March 1998 as a result of enforcement of the Schengen Agreement was accompanied by increasing pressure on Italy for better management of infows, including those of asylum seekers, since Northern European countries such as Germany, the Netherlands and Sweden were afraid that their generous asylum and welfare systems might act as magnets for secondary movements (Boswell & Geddes, 2011).

A frst change in the government's approach emerged as early as 1998, when the government put the Kurds from Turkey and Iraq through the ordinary asylum procedure, with the aim of avoiding a pull effect of the easily accessible temporary ad hoc protection, and demonstrating to other Member States that Italy did not have excessively liberal immigration policies (Vincenzi, 2000).

In that same year, the centre-left government led by Romano Prodi passed the frst systematic and comprehensive immigration act, Law 40/1998 (called *Legge Turco-Napolitano*). Among the many provisions, this Law authorised the opening of centres for the temporary detention of irregular migrants to be returned, as a response to the Schengen requirements. At the same time, it introduced a permit for humanitarian reasons, defned in rather vague terms, lasting 2 years and convertible into a work, study or family permit. This nation-based protection put an end to the ad hoc pieces of legislation issued for managing specifc infows and was largely applied for the following two decades, until it was abolished by the Migration and Security Decree in 2018 (see Sect. 15.5).

At the same time, the arrival of people from Kosovo, together with the entry into force of the Dublin Convention in 1997, which put the responsibility for processing asylum claims on the frst-entry country, stimulated an effort to develop a more standardised and structured reception system that turned into the launch of Shared Action (*Azione Comune*) in 1999 (Marchetti, 2014; Bona & Marchetti, 2017). This

<sup>4</sup>They were renamed First Aid Centres and Assistance (CPSA) in 2006.

was a network of public and non-public actors stemming from the initiatives developed in the 1990s, co-funded by the European Commission, and based on the principle of decentralised provision of integration services and reception, with projects set up by local actors, including municipalities (CENSIS, 2005; OECD, 2019).

Shared Action developed into the National Asylum Programme (*Programma Nazionale Asilo*, PNA), which started in October 2000 with a Memorandum of Understanding signed by the Minister of the Interior, UNHCR and ANCI (National Association of Italian Municipality), and entered into force in July 2001. Municipalities became the only local actors eligible to submit reception projects to the Ministry of the Interior (Hein, 2010; Marchetti, 2014). The aim was to institutionalise the Italian reception system and meet the criteria of the European Fund for Refugees through which the PNA was funded (Bona & Marchetti, 2017). Indeed, funding schemes can be regarded as alternative means of pressure: as "soft policies" to pursue Europeanisation (Penninx, 2015).

This process continued during the centre-right government led by Silvio Berlusconi, when PNA was fully institutionalised by Law 189/2002 (called *Legge Bossi-Fini*) and renamed SPRAR (Protection System for Asylum Seekers and Refugees). Despite some recent changes in its name and access criteria (see Sect. 15.5), SPRAR has remained the Italian ordinary reception system until now. Its facilities are set up on a voluntary basis by local authorities that apply in response to the periodic calls for projects issued by the Ministry of the Interior which covers the largest share of the costs. Reception facilities, usually managed by local NGOs that participate in local authorities' public bids, are generally articulated in apartment-based solutions; they include integration measures and are highly regulated and monitored. Despite being largely acknowledged as a good practice, SPRAR has always suffered from the limited number of places available, constantly below the reception demands, and an uneven distribution throughout the country, as a consequence of its bottom-up and locally based approach.

Law 189/2002 also set up the National Fund for Asylum Policies and Services and reformed the asylum procedure, instituting the Territorial Commissions for the Recognition of International Protection operating throughout the country so that interviews could be done even outside of Rome. The Commissions were made up of representatives of local authorities, the police, the Prefectures and UNHCR. UNHCR participated in interviews and decisions, and was called on to play a key role over the following years by providing guidelines and information on the origin countries and courts' judgements, compensating for the other members' lack of specialised knowledge of asylum or human rights matters.

At the same time, Law 189/2002 introduced restrictive provisions similar to those adopted in older asylum countries, including stricter procedures for expulsion and detention of migrants caught without residence permits and a fast-track asylum procedure at the border (Hein, 2010; Marchetti, 2014).

The other major legislative change in the frst decade of the 2000s was the transposition of the EU Directives representing, together with the Dublin and EURODAC Regulations, the core of the Common European Asylum System (CEAS) launched at the European Council of Tampere in 1999. Despite the several national bills presented through the 1990s and the early 2000s, asylum provisions had always been included in migration acts so that the laws transposing the EU Directives have so far been the only acts specifcally addressing asylum, confrming the key role of the EU in shaping the Italian asylum legislation (Finotelli, 2018).

The concern for the implementation of the new provisions was scarce, however, since their adoption was perceived as responding to European rather than national interests. Consequently, at the end of the 2010s Italy was by no means equipped to process and host even a few tens of thousands of refugees, as was soon revealed by subsequent developments.

#### **15.3 Adolescence, When You Protest Against the Adults: The Management of the Arab Spring's Refugees**

Given that the above-mentioned advancement in the national legislation was largely driven by top-down pressures from Europe and was not accompanied by a shift in national policy frames, the old emergency approach soon resurfaced. Yet, the usual instruments and practices went through a series of failures which gradually led to a change in policy goals and frames.5 Indeed, this time Italy had to act in a new context, where infows grew signifcantly and the country was fully integrated into the Schengen Area and the CEAS. Being exposed to a similar problem pressure and regarding Italy as a full partner of the European mobility regime, the other countries became less tolerant towards its old practices. Employing the biological metaphor, we could say that during the "adolescence" stage, the adoption of emergency solutions by Italy appeared to be a sort of revolt against the "adults" of the club it had striven to join.

In 2011, in the face of the increasing infows unleashed by the so-called Arab Spring – with sea arrivals growing from 4400 in 2010 to 62,700 in 2011 (see Fig. 15.1) – the Italian government tried to export the emergency frame on a European level in order to suspend the regular procedures and activate the exceptional instrument of temporary protection set forth under EU Directive 2001/55/ EC. However, both the European Commission and the EU Council rejected the proposal and opted for providing fnancial and technical aid, framing the crisis as an ordinary infux of irregular migrants and calling for stronger border-control measures by the Italian authorities (Campesi, 2011).

As a result, on 12 February 2011 the Italian government adopted a nationalbased ad hoc solution: it declared the state of humanitarian emergency, labelled as the "North Africa Emergency" (ENA, in the Italian acronym).6 The Italian authorities considered Tunisian nationals arriving on the Italian shores neither irregular nor

<sup>5</sup> Indeed, according to Hall (1993), third-order change of policy follows the accumulation of anomalies and a series of policy failures.

<sup>6</sup>Order of the Prime Minister, 12 February 2011, OPCM 3933/2011.

**Fig. 15.1** Asylum applications and sea arrivals in Italy (2009–2020). Source: elaboration by the Author of data from Eurostat (https://ec.europa.eu/eurostat/databrowser/view/MIGR\_ ASYAPPCTZA/default/table?lang=en) and openopolis (https://www.openpolis.it/numeri/gliarrivi-di-migranti-in-italia-dal-1997-al-2020/)

subject to international protection (ENM, 2011), and arbitrarily detained them outside any lawful procedures while working with other Member States and the Tunisian government to negotiate diplomatic solutions. On 5 April the government signed an agreement with Tunisia on police cooperation and readmission of irregular migrants. On that same day, the Italian Prime Minister issued a decree whereby temporary humanitarian protection was issued according to Law 40/1998 for the Tunisian citizens who had landed since the beginning of the year (Campesi, 2011).7

The same decree provided the migrants with a travel document allowing them to freely circulate within the European Union (ENM, 2011; Marchetti, 2014). The explicit aim was to encourage their departure, and thus to obtain the consensus of the Northern League which was participating in the government and was reluctant to allow migrants from Tunisia to stay in the country (Campesi, 2011). In response, the French government contested the legality of the temporary visa, returning its holders to Italy, and reinstated border checks along the frontier due to "public order concerns" (Carrera et al., 2011). At the Justice and Home Affairs meeting on 11 April 2011, Italy was explicitly accused of violating the "Schengen spirit". As underlined by Campesi (2011), this was an occasion to open up the debate on the revision of Schengen governance. Indeed, in 2011 the European Commission proposed a series of amendments to the Schengen Borders Code that came into force in 2013.8 Specifcally, after those changes, Article 26 of the Code provides for

<sup>7</sup>DPCM 5 April 2011. Actually, only 11,800 out of the 23,500 Tunisian migrants who reportedly entered Italy during that period were granted this status (Marchetti, 2012).

<sup>8</sup>Regulation (EU) No 1051/2013 of the European Parliament and of the Council of 22 October 2013 amending Regulation (EC) No 562/2006.

exceptional circumstances allowing for the suspension of Schengen where the overall functioning of the free-movement area is put at risk as a result of persistent serious defciencies relating to external border control.

In the meanwhile, all of the subsequent newcomers who arrived after 5 April 2011, mainly Sub-Saharan Africans residing in Libya when the Gheddaf's regime collapsed, were somehow *a priori* classifed as asylum seekers,9 and automatically channelled into the reception plan set up within ENA. The government established specifc reception facilities outside the ordinary reception system SPRAR and put those facilities under the coordination initially of the Prefectures and then of the Civil Protection Service. Most of them were large-size collective centres concentrated in bigger cities and providing poor or no integration measures. This plan turned out to be a big failure. The majority of people hosted there were still homeless and jobless when, at the end of ENA in March 2013, they were abruptly expelled from those facilities so that some of them illegally occupied empty buildings (Marchetti, 2012; Ponzo et al., 2022; Pogliano & Ponzo, 2019).

It became clear that the rooted policy practices, i.e., fostering secondary movements and adopting reception emergency solutions, did not work so well in the new context, where unplanned fows towards Italy were boosted by the geopolitical instability in the Mediterranean area and the European partners became more demanding.

Against this backdrop, the state-managed Search and Rescue operation Mare Nostrum, started by Enrico Letta, Prime Minister of the then centre-left government, on 18 October 2013, in response to the tragic shipwreck of Lampedusa that had occurred 2 weeks earlier, can be regarded as the starting point of a new course.10 Mare Nostrum, by bringing people to the Italian shores under the coordination of the state, required that they be channelled through more structured paths, overcoming the contingent responses to spontaneous arrivals (interview with the Head of Italian Caritas' Immigration Offce, 11 December 2018). Most important, Mare Nostrum marked a shift in the Italian position towards Europe. It was indeed aimed at putting pressure on the EU within the struggle of Enrico Letta and his successor Matteo Renzi to fx the principle that the infows triggered by the Arab Spring could not be an exclusive concern of Mediterranean Member States (Pastore & Roman, 2014; Panebianco, 2016). Hence, the idea of responsibility-sharing, namely the traditional concern of Northern European countries (Boswell & Geddes, 2011; Bendel & Ripoll Servent, 2018), suddenly became Italy's main claim. The Joint Operation Triton that took over Mare Nostrum in October 2014 was a rather poor result in this respect since, beyond its much more limited assets, area of patrolling and mission, it kept bringing rescued people only to Italian ports (Pastore & Roman, 2014). Nevertheless, the revision of the old policy frames and practices did not stop.

<sup>9</sup> In October 2013 a general provision granted some form of protection to all the migrants assisted by ENA (*Circolare* of the Ministry of the Interior no. 400, 31 October 2012).

<sup>10</sup>The operation cost more than nine million euro per month, mainly covered by the Italian Navy's ordinary budget, and led to the rescue of more than 130,000 migrants (Pastore & Roman, 2014).

A further confrmation of the new emerging course came from the epochal change that occurred in public discourse. Still in the frst decade of the 2000s, the mechanisms for handling unwanted immigration in Italy were embedded in the economic legitimation of immigration (Finotelli & Sciortino, 2009), whereas in 2011–2013 the prevalent emphasis in public discourse shifted towards the humanitarian character of the crisis and Italian commitment (Marchetti, 2014), recalling the logics adopted by older asylum countries such as Germany (see Chap. 7).

Nevertheless, the established policy practices and frames were not dismantled altogether: secondary movements as a legitimate response to increasing infows persisted. A case in point is that of over 80,000 Syrians and Eritreans stopping a few days in Milan to organise their journeys towards other countries. Almost none of them got fngerprinted and since October 2013 the Ministry of the Interior provided economic support to the Municipality of Milan through the local Prefecture in order to establish reception facilities to host transit refugees, while frst-aid and information on journeys were provided at the Central Station (Pogliano & Ponzo, 2019). However, this practice became increasingly hard to implement and stopped in 2016, as we will see in the next section.

#### **15.4 Coming of Age, When You Cannot Step Back Anymore: The European Refugee Crisis**

The new course was reinforced and eventually institutionalised during the so-called European refugee crisis. Despite the skyrocketing arrivals registered between 2014–2016 (see Fig. 15.1), the emergency frame was abandoned in the public discourse and asylum was depicted as a structural phenomenon. Consistent with the strategy inaugurated by Enrico Letta, the emphasis was instead on the lack of support from the rest of Europe (Marchetti, 2014). On 23 June 2015, Prime Minister Matteo Renzi wrote in the national newspaper *La Stampa:* "The demand for peace and food that forces thousands of women and men, sometimes with their children, to risk death to reach Europe did not start today, it will not end tomorrow (…) We are a great country that does not get hysterical because a few thousand more refugees arrive in a year. If forced to go it alone, we will not step back (…) But having a European response would be useful to Europe frst, then to Italy"11 (translation by the Author).

In terms of regulations and practices, the Italian government started to pursue a "normalisation" in the management of asylum (interview with a high offcial in the Department for Civil Liberties and Immigration of the Ministry of the Interior, 12 October 2018). Indeed, after the end of ENA, the Italian government, in collaboration with the National Association of Italian Municipalities, the main national

<sup>11</sup> https://www.lastampa.it/esteri/2015/06/23/news/renzi-sui-migranti-l-italia-non-rinuncera-maia-salvare-vite-umane-1.35255023

NGOs and the UN agencies started a phase of feverish reforms (interview with a high offcial of SPRAR Central Service, 17 July 2018) and avoided the adoption of any further emergency plan, whereas several older asylum countries, from Finland to Germany, implemented such plans in 2015–2016 to accommodate the booming numbers of asylum seekers (Caponio et al., 2019; Ponzo, 2022).

As a frst step, the government mobilised in order to make up for the lack of places in the SPRAR system: in 2014, it adopted a National Plan for the relocation of asylum seekers which entitled the Prefectures to set up governmental reception facilities, called CAS (Extraordinary Reception Centres). Regrettably, the quality of services provided by CAS was extremely heterogeneous: CAS projects went from SPRAR-like solutions organised in apartments and aiming at peoples' empowerment, to large-size isolated centres with no integration services. However, this also happened in old asylum countries because of the high pressure to rapidly multiply the available places (Caponio & Ponzo, 2022).

In order to reduce the gap between CAS and SPRAR, internal administrative regulations (*Circolari*) issued by the Ministry of the Interior in 2014 asked the Prefects to follow SPRAR's guidelines and some of its key principles when setting up CAS centres.12 Nevertheless, implementation remained jeopardised since the Prefectures, enjoying a high level of discretion and being poorly equipped for managing the governmental reception, complied with those directions to a very different extent (Ponzo et al., 2022).

A key step in this process of normalisation was the "Agreement between the Government, Regions and Local Authorities on the Implementation of the National Plan to face the extraordinary infow of third country nationals, adults, families and unaccompanied minors" signed on 10 July 2014. It rationalised reception by articulating it into three categories, i.e., frst aid and identifcation centres at disembarkation points, frst-level reception at governmental facilities and second-level reception. Moreover, the Agreement acknowledged the principle of dispersed accommodation to prevent concentration in big cities as had occurred with ENA, and established an annual National Operational Plan on Asylum aimed at estimating the need for reception places and setting their distribution across the Italian regions.13

Many of those provisions were ratifed by the Legislative Decrees adopted in 2015 and transposed the recast EU Directives on asylum issued in 2013. Hence, those Directives were mainly regarded as a "window of opportunity" to rationalise and gather into single laws the decisions made until then, rather than top-down impositions (Ponzo, 2022).

In the meantime, the pressure around secondary movements coming from the EU and the European partners increased. In September 2014 the Ministry of the Interior,

<sup>12</sup>*Circolare* 104, 8 January 2014; *Circolare* 14,100/27/I, 2 May 2014; *Circolare* 0005484, 27 June 2014; *Circolare* 14,906, 27 December 2014.

<sup>13</sup> It refned the criteria fxed during ENA when, in March 2011, the Ministry of the Interior established a collaboration with the Regions to set regional quotas based on the number of residents (Department of Civil Protection – Presidency of the Council of Ministries, 12 April 2011), although with rather disappointing results.

acknowledging the complaints of other Member States in this regard, issued a *Circolare* that established tighter identifcation procedures14 (Caritas Roma, 2014). But it did not prove to be enough so that Italy's full membership in CEAS and Schengen was called into play. In December 2015 the European Commission took the frst step of an infringement procedure against Italy on the implementation of the Eurodac Regulation which provides for effective fngerprinting of asylum seekers and transmission of data to the Eurodac central system within 72 h.15 Moreover, between September and October 2015, neighbouring countries suspended Schengen and restarted border controls, with Germany and Austria raising the failure of Italy and Greece to apply the external border controls in their notifcations to the European Commission. On 1 December 2015, the European Council published a proposal entitled "Integrity of the Schengen Area" (14,300/15) for the application of Article 26 of the Schengen Borders Code as modifed in 2011 (see Sect. 15.3) and the reintroduction of internal border controls (Guild et al., 2015).

In this context, behaving as a transit country was no longer a viable option. Hence, the Italian government tried to bargain and used the systematic fngerprinting of newcomers as a currency of exchange to obtain more responsibility-sharing. This negotiation was ratifed by the "European Agenda on Migration" launched at the end of 2015 under the EC Presidency of Jean-Claude Juncker. Together with Greece (see Chap. 16), Italy accepted the adoption of the so-called hotspot approach at disembarkation ports which is based on tight cooperation between Italian police forces, EU agencies (Frontex, EASO, EUROPOL) and international organisations (IOM and UNHCR) with the aim of registering and fngerprinting every newcomer who irregularly crosses the sea borders. As a consequence of this decision, the gap between sea arrivals and asylum applications lodged in Italy started to diminish in 2016 and completely closed in 2017, as shown in Fig. 15.1. In exchange, Italy obtained more responsibility-sharing, mainly embodied in the European "emergency relocation mechanism" of third-country nationals in clear need of international protection (i.e., belonging to nationalities with a high recognition rate) from frontline countries to other Member States.

Yet, the unbalanced implementation of the two components of the agreement was regarded as a sort of betrayal by the Italian government: on one hand, the number of asylum applicants fngerprinted and registered in Italy grew signifcantly; on the other hand, the relocation of asylum seekers was poorly implemented by other Member States providing the frontline states with a limited amount of relief. "The only ones that hosted refugees in a serious way with a great sacrifce of the population and elected politicians, as you can see from the last elections, are Italy, Germany, Sweden, even Norway and Switzerland, although the last is not in the EU. The others have refused to host anyone, despite everyone having signed the Juncker Agenda, leaving Greece and Italy completely alone and failing to honour their obligations"

<sup>14</sup>*Circolare* of the Ministry of the Interior No. 28197, 25 September 2014.

<sup>15</sup> https://www.statewatch.org/media/documents/news/2015/dec/eu-com-infringements-10- 12-15.pdf

(interview with a high offcial of the Department for Civil Liberties and Immigration of the Ministry of the Interior, 12 October 2018).

In fact, secondary movements partially persisted but changed in nature: migrants started to apply for asylum in Italy and then moved towards other Member States on an irregular basis to seek jobs, returning to Italy to renew their permits (Hatziprokopiou et al., 2021). Therefore, Italy remains the state responsible for the asylum procedure, and secondary movements appear circular and related to the weak controls of both Italy and the destination countries, mainly France and Germany, rather than the result of a proactive Italian strategy to circumvent the Dublin Regulation, as when the state supported Syrians who were passing through Milan on their way to other European countries.

Meanwhile, the issue of the redistribution of asylum seekers across the country took a step forward. The idea of asylum seekers' mandatory relocation was opposed by local authorities since it would have undermined the bottom-up approach of SPRAR which was rooted in the initiatives set up in the 1990s to accommodate former Yugoslavian refugees. A compromise was found at the end of 2016 and ratifed in the so-called "Bari Agreement" signed by the National Association of Italian Municipalities and the Ministry of the Interior, establishing a specifc quota of refugees per municipality, i.e., a ratio of 2.5 hosted asylum seekers per 1000 residents. This mechanism resembles the redistribution methods adopted in Northern Europe, such as the German distribution key (*Königsteiner Schlüssel*) where the distribution quota is calculated annually according to the tax revenue and the population of single federal states (Glorius, 2022). Moreover, this decision was accompanied by the so-called "safeguard clause" (*clausola di salvaguardia*) 16: the municipalities whose SPRAR reception places met the above ratio would be exempted from the setting up of CAS by the Prefectures.

Thanks to the "safeguard clause" and an increase in the central government's co-funding from 80% to 95%, in 2016 SPRAR places grew by 20% and reached a total of 26,000. However, this growth was still not enough to keep pace with new arrivals which were over 181,000, so CAS continued to cover the largest share of reception (Ponzo et al., 2022). Despite the limits of CAS facilities mentioned above, they have in fact played a key role in closing the gap with older European asylum countries and following their same rationales, i.e., providing accommodation to every newcomer regardless of the size and variation of the infows.

Moreover, different from the limited-in-time solutions adopted in the 1990s and in 2011–2013 during ENA, CAS facilities have become a rather permanent component of the Italian reception system. This allowed the national authorities, generally under the pressure of civil society organisations, to develop strategies aimed at increasing CAS standards. Those improvements started in 2014, as explained above, and accelerated when Marco Minniti became Minister of the Interior in December 2016, under Paolo Gentiloni's centre-left government. Since the end of 2017 the

<sup>16</sup>*Circolare* of the Ministry of the Interior, 1 October 2016; Directive of the Minister of Interior, 11 October 2016.

organisations managing CAS have been requested to submit balance checks together with supporting documents to the Prefectures, whereas before that they used to submit invoices only, with little accountability about how the money was spent.17

In that same year, the Ministry of the Interior adopted a new bid scheme for governmental centres18 according to which CAS should provide reception and integration services similar to those offered in SPRAR facilities, with the exception of legal assistance to prepare for the interview, vocational training and support for job seeking and housing seeking. Although CSOs criticised the lack of these services, the 2017 bid scheme contributed to harmonising service provision in CAS and to reduce the gap with SPRAR facilities.

With regard to the asylum procedures, as early as 2014, Law 146 increased the number of Territorial Commissions for the Recognition of International Protection from 10 to 20, with 30 local sections, and established that their members had to undergo a dedicated training delivered by UNHCR and EASO. The process for increasing the effciency of asylum procedures accelerated with Legislative Decree 220/2017 (called the *Minniti-Orlando Decree*), converted into Law 46/2017, which entered into force on 31 January 2018 and introduced some crucial changes: the recruitment of ministerial offcials with specifc expertise to become permanent members of the Territorial Commissions, overcoming the turnover of different public entities' representatives with little knowledge of asylum matters; the replacement of the UNHCR representatives with experts proposed by the UN agency but employed by the government; the establishment of 26 specialised court sections responsible for examining asylum appeals only, without however providing additional funding for this task; the suppression of the possibility to appeal a negative Civil Court decision before the Court of Appeal so that the applicant can only lodge an appeal before the Court of Cassation (for details see Roman, 2000).

To conclude, we can affrm that the Italian asylum regime has come of age: the country has fnally recognised to be an asylum country and its general goals and rationales became similar to those of traditional European asylum destinations. Elements that were completely absent in the early 2000s are now taken for granted, such as codifed disembarkation procedures, systematic fngerprinting, professional commissions for the recognition of international protection, specialised court sections and the right to reception for all asylum seekers without fnancial means. Moreover, greater responsibility-sharing within Europe has become the cornerstone of every Italian government's migration policies regardless of its political colour.

<sup>17</sup> Inter-Ministerial Decree of the Ministry of the Interior and Ministry of the Economy, 18 October 2017.

<sup>18</sup>Ministerial Decree, 7 March 2017.

#### **15.5 Weak Political-Institutional Capacity: The Italian Asylum Regime's Main Hurdle**

Whereas modifcations in the social phenomenon of migration and the new institutional settings have fostered policy change, the country's weak political-institutional capacity, which results from a mix of the governments' instability and a weak state capacity, has pushed towards the opposite direction. Indeed, it appears to be a substantial obstacle to the consolidation of the new policies, practices and regulations.

Governmental instability and the high politicisation of migration have always been crucial in defning the contradictions of Italian immigration policy (Sciortino, 2009; Finotelli & Echeverria, 2017). This has become particularly evident in the feld of asylum over the last few years. In the second half of the 2010s, the frequent changes in the political colours of the short-lasting governments and their use of migration as an ideological fagship undermined the consolidation of the newly established policy practices. Reforms concerning asylum occurred almost every 2 or 3 years, with each piece of legislation partially dismantling the previous one – i.e. the laws adopted in 2015 to transpose the recast EU Directives, Legislative Decree 113/2018 on Security and Migration adopted by the Ministry of Interior Matteo Salvini, Decree 130/2020 adopted by the Ministry of Interior Luciana Lamorgese (for details see Ponzo et al., 2022). At the same time, those reforms seem to take the idea of Italy as a destination country for granted so that they offered the different governmental coalitions' recipes to better manage the phenomenon and, as in other asylum countries, mainly focused on asylum seekers' rights, i.e., the national forms of protection and eligibility for reception.

At the same time, the implementation of asylum policies has suffered from a weak state capacity which, in her seminal works, Skocpol (1979 and 1985) defnes as whether a state is able to implement offcial goals. Following Geddes (1996), we could say that state capacity can be equated to the implementation power of the state, a task that falls under its bureaucracy. More precisely, Skocpol and Finegold (1982) argue that the success of a certain state programme depends largely on the availability of a professional independent bureaucracy made up of employees with proper training and experience in public policies, endowed with fnancial resources, able to collect data and conduct policy-oriented research and analytical work indispensable for all control programmes without relying on extra governmental experts and organisations, and holding a public-service perspective while maintaining constant relations with the major interest groups.

As for Italy, a comparative overview of public administration characteristics and performance in the EU28 (Thijs et al., 2018) shows how the country scores low on transparency and accountability, professionalism of civil servants, human resources management capacity and performance. In the specifc feld of migration, Finotelli and Echeverria (2017) underline how in Italy the lack of administrative staff, its ineffcient selection procedures and the weak coordination of the bureaucratic apparatus have been major impediments to effective implementation of labour migration policies. Asylum policies seem to suffer from the same weaknesses. Yet, state capacity cannot be regarded as a fxed attribute: it varies over time and across domains (Skocpol & Finegold, 1982; Skocpol, 2008), as developments in the management of arrivals, asylum procedures and reception show.

The management of arrivals at sea has substantially improved since 2013, with fngerprinting becoming systematic and procedures after disembarkation being harmonised and codifed with the contribution of UNHCR, IOM and EU agencies (Giannetto et al., 2019). On the other hand, the state still depends on UNHCR and other NGOs to provide assistance and information to migrants at arrival points, identifying vulnerable people and carrying out referrals.

State capacity has grown signifcantly in asylum procedures too, thanks to the establishment of specialised court sections and the recruitment of ministerial offcials to make up the Territorial Commissions on the Recognition of International Protection in 2017, allowing the phasing out of the UNHCR, as explained in the previous section. At the same time, the backlog is still substantial: in January 2021 there were almost 34,000 pending claims in the Territorial Commissions and in June 2020 there were nearly 95,000 pending appeals in Courts, with a waiting time for the court's decision between 15 months and 3 years.19

Regarding reception, state capacity has greatly expanded so that all asylum seekers without means now have access to some form of accommodation. Yet, governmental facilities are not under the coordination and monitoring of a migration-specifc governmental agency or department, as in most other large EU receiving states, but rather under the responsibility of the Prefectures. Since the latter have not been adequately expanded and trained to properly manage those new tasks (Ponzo et al., 2022), several central governmental decisions over those facilities have remained on paper or have been implemented with wide territorial disparities and low accountability (Openopolis and Action Aid, 2018). Moreover, monitoring has been largely contracted out to UNHCR and other nongovernmental organisations through specifc projects substantially co-funded by the EU (Giannetto et al., 2019).

To sum up, the state's implementation power has substantially improved in various asylum domains but still falls short in some respects while national and international nongovernmental organisations compensate for the state's lack of internal skills in performing specifc tasks such as assistance at disembarkation and monitoring of reception. This limited state capacity, together with the above-mentioned regulation revisions promoted by every new government, undermines the consolidation of the new practices and hampers the ongoing policy change.

<sup>19</sup>Updated statistics on the average duration of the procedure are not available (AIDA, 2020). The data are drawn from an investigation report of the newspaper *Il Sole 24 Ore* (https://www.ilsole24ore.com/art/immigrati-attesa-una-risposta-richieste-140mila-richiedenti-asilo-ADkGZyEB).

#### **15.6 Concluding Remarks**

The chapter challenges persistent stereotypes depicting Italy as a deviant laggard compared to older asylum countries. With this aim, I have analysed the policy change that has occurred over the last three decades by singling out policy frames and practices, showing how the country has passed from what I have called, using a biological metaphor, infancy to adulthood.

Indeed, after a long period of stability, over the last decade policy frames and practices have been evolving in response to the failure of the old approaches.20 The main factors that led to those failures seem to be the modifcations in the social phenomenon of migration (the sharp increase of unplanned infows) and in the institutional settings where negotiations among Member States occur (the full inclusion of Italy into the Schengen Area and the CEAS). The resulting policy change consists of a new policy frame that acknowledges Italy as a destination country for asylum seekers and the implementation of new practices and regulations which overcome ad hoc emergency solutions. Moreover, Italy joined the older European asylum countries in the call for greater responsibility-sharing. Hence, European solutions are now actively pursued instead of being passively received as top-down impositions and possibly circumvented in practice, as happened in the past.

However, the country's weak political-institutional capacity, by undermining the consolidation of the new practices, has hampered this policy change. This weakness has become one of the main explanatory factors of the gaps existing between Italy's and older asylum countries' asylum regimes whereas strategic divergence from EU partners has lost relevance.

In the current context, with the EU struggling to agree upon the management of refugees and increasing geopolitical instability in third countries, including Ukraine, it is diffcult to say how Italy will deal with the dual pressure coming from a conficting Europe and the unpredictable infows, for which further research is needed.

#### **References**


<sup>20</sup> Indeed, according to Hall (1993), third-order change is always preceded by a series of policy failures that progressively undermine the consensus around the established policy paradigm (in this case, Italy as a transit country) and open the way to the adoption of a new one (Italy as a destination country).


Hein, C. (Ed.). (2010). *Rifugiati. Vent'anni di storia del diritto d'asilo in Italia*. Donzelli Editore.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 16 The Greek Asylum Regime: From Latecomer on Reception to Inspirational Model on Asylum Procedures**

**Angeliki Dimitriadi**

#### **16.1 Introduction**

Since the 1990s, the European framework has sought to enhance the mobility of EU citizens while restricting the (irregular) entry of third country nationals (Kostakopoulou, 2000), including asylum-seekers. The "gatekeepers" in this endeavour are frst and foremost the frontline states, specifcally Greece, Italy, and Spain. Due to their geographical location, they receive the majority of irregular arrivals, and they act as buffers to the rest of the EU. Unlike other frontline states, Greece lacked, until 2011, a suffcient institutional framework to process asylum applications and host asylum applicants. The country had been severely criticized in the period 2000–2010, for failing to provide to asylum applicants a functioning asylum process and dignifed conditions of stay (Cabot, 2014). Signifcant changes take place since 2015. Greece functions as the main entryway to the EU, with 850,000 migrants crossing its maritime border, according to UNHCR. The majority do not remain in Greece, opting to transit to other EU Member States (MS). This is evident in the rather limited number of asylum applications; only 13,197 applications were submitted in 2015 to the asylum service. Greece, in turn, was criticized for indirectly assisting transitory movements, by failing to register on EURODAC1 migrants on arrival to the land and/or sea border.

At the start of the "crisis", in May 2015, the European Commission announced the Agenda on Migration and Asylum (henceforth the Agenda). The Agenda

A. Dimitriadi (\*)

College Year in Athens, Athina, Greece e-mail: angeliki.dimitriadi@cyathens.org

© The Author(s) 2023

<sup>1</sup> In 2015 the Commission initiated infringement procedures against Croatia, Greece, and Italy for failing to register migrants in the EURODAC upon arrival to the continent. The Commission noted that in Greece, 492,000 migrants were registered and only 121,000 fngerprinted of the estimated 890,000 that crossed (Zalan, 2015).

C. Finotelli, I. Ponzo (eds.), *Migration Control Logics and Strategies in Europe*, IMISCOE Research Series, https://doi.org/10.1007/978-3-031-26002-5\_16

incorporates the notion of the "hotspot approach". The latter is presented as a migration management tool, to ensure that those who reach the EU external borders are registered before processing their asylum claims. The "hotspots" originally are envisaged as both sites of disproportionate pressure as well as sites where the streamlined operation of screening and registration takes place.2 The hotspot approach renders newcomers' registration in the EURODAC database systematic (Tazzioli & Garelli, 2020). Greece establishes the frst registration facilities (known as hotspots) in June 2015, long before the EU-Turkey Statement of 2016. Yet, bureaucratic delays, and objections from locals on the islands, delay the project. The frst registration facility, Moria in Lesvos, is only introduced in October 2015, with the rest following throughout 2015 and early 2016. The EU-Turkey Statement (henceforth the Statement) of March 2016, *de facto* alters the nature of the hotspots. From temporary screening sites, they transform overnight into spaces of indefnite stay, where newly arrived migrants remain until such time as they receive asylum in Greece or are returned to Turkey as per the Statement. The latter infuences in unexpected ways the asylum processing in the country. Greece is the only Member State in the EU that implements two parallel asylum procedures, i.e., a regular procedure in the mainland and a fast-track border procedure on the islands,3 with signifcant repercussions on asylum applicants. It is the implementation of the Statement that offsets a period of unprecedented change in Greece as regards the asylum regime, from the processing of applications to reception.

To ensure implementation of the Statement, a swift transfer took place of all asylum applicants on the Greek islands that arrive prior to March 18, 2016, so that their applications are processed under the regular procedure (in contrast to those arriving post March 18, 2016). Their transfer to the mainland, in conjunction with the closure of the Western Balkan route, created an emergency, as regards reception, to which international organisations, local and non-state actors responded, resulting in synergies and partnerships. Through formal and informal practices of cooperation, Greece acquires a reception system that in 2021 passes under the full control and administration of the Greek state.

The different dynamics regarding the reception of asylum seekers but also the asylum procedures, constitute part of the broader asylum regime in Greece. The notion of regime derives from the feld of international relations but has been utilised in migration research since the early 1990s. The regime implies the exploration of the governance structures through a set of negotiating practices that can be formal but also informal (Rass & Wolff, 2018). In other words, public and private actors, state, and non-state actors, produce and implement various decisions as regards migration and asylum through a continuous interaction and negotiation (Cvajner et al., 2018, Finotelli and Ponzo, introduction). In the period 2015–2019, the Greek asylum regime is characterised by divergent implementation between the

<sup>2</sup> In the Agenda, the "hotspots" are borders facing disproportionate pressure (initially Greece and Italy) where the approach of screening and registration is implemented in facilities set up to enable the various EU Agencies (Frontex, Europol, EASO) to deploy and perform their tasks.

<sup>3</sup>Since 2020, the fast-track border procedure has been extended also on the land border in Evros.

border areas and the mainland. Different asylum processing and different reception conditions and accommodation options exist between the border areas. This divergence is reinforced by actors involved; non-state actors are actively involved in reception, and state and EU actors in the asylum procedures. Eventually by late 2020, all aspects of the asylum regime are administered by the state and a logic of deterrence is incorporated. Aspects of said logic are visible in the legislative reforms undertaken in the context of the International Protection Act (IPA) and subsequent amendments affecting both asylum processing and reception. The aim is to make Greece a less-attractive destination for asylum seekers. In turn, elements already applied in Greece have now found their way to the EU New Pact on Migration and Asylum of 2020, indicating parts of Greece's asylum regime are in line with the EU's overall asylum policy aims and objectives.

The chapter discusses the development of the asylum regime with a particular focus on the reception system in the mainland and the processing of asylum claims at the maritime border. Section 16.2 offers a brief discussion of the Europeanisation of Greece's migration governance prior to 2015, identifying the main approach of Greek migration and asylum policy until the refugee crisis. Section 16.3 focuses on the reception system that emerges from the implementation of the EU- Turkey Statement of March 2016; the core components and the synergies developed. Both sections are based on empirical research carried out in 2018 in the framework of the CEASEVAL project.4 Data collection identifed the legal changes in the asylum system, complemented by ffteen (15) semi structured interviews with key stakeholders at national and local level as well as non-state actors (e.g., international organisations, NGOs, researchers). Here, Greece maintains the role of the "student", developing reception services already in place by its European partners. Section four shifts the focus to asylum procedures, the legislative changes implemented both for the application of the Statement in 2016 and particularly in 2019–2020 with the hopes of re-invigorating core aspects of the deal, namely returns. The chapter concludes with a look at the elements of the New Pact that appear "modelled" on the experience of Greece, suggesting that the country functions both as a "student" when it comes to reception but also as an inspirational model for the EU as regards asylum procedures.

#### **16.2 Greece's Asylum Prior to 2015**

In the 1990s Greece lacked a functioning asylum system. Greece endorses and adopts the status of a "transit country" (Papadopoulou-Kourkoula, 2008; Dimitriadi, 2018), thereby assuming that migrants would seek to settle in other EU Member States. The Greek state, based on the logic of ethnic homogeneity and the desire to

<sup>4</sup>H2020 Project "CEASEVAL" (Evaluation of the Common European Asylum System) Grant Agreement No 770037.

maintain it, does "not trust or encourage the establishment of non-nationals in the country" (Papageorgiou, 2013, p.76). This "mistrust" according to Papageorgiou (2013), is incorporated in the governance of migration and asylum, resulting in the absence of a functioning asylum system until 2011. The Hellenic Police was responsible for asylum processing and routinely rejected asylum requests on submission (Cabot, 2014). No service provision was included, no fnancial assistance to asylum applicants nor accommodation options with one exception: a formal camp in the outskirts of Lavrio (port) in Attika. By reducing access and incentives, asylum applicants are "encouraged" to transit rather than settle in the country. This system is partially disrupted in 2010 as Greece undergoes a process of renewed Europeanisation5 of its asylum and reception system. Europeanisation is more than the adoption of institutional structures, or the transposition of European legislation (Sotiropoulos, 2007) but also the normative (a human-rights based asylum system) and policy processes, in this case refugee protection (Lavenex, 2001).

In Greece, the Europeanisation period of 2010–2014 is brought about by the decision of the European Court of Human Rights in *M.S.S. v Belgium and Greece*. The Court condemned Greece for failure to uphold Article 3 of the European Convention on Human Rights (ECHR) as regards the degrading detention and living conditions of the applicant (M.S.S). Belgium was also condemned for enforcing a Dublin return, despite the numerous and public information of Greece's inability to uphold the standards of the Convention (Moreno-Lax, 2011). The Court's decision resulted in suspension of Dublin returns for Greece (AIDA, 2014), extended until mid-2019. It also kicked off the Europeanisation process of the Greek asylum system, starting with the National Action Plan on asylum of 2010 and culminating in the adoption of Law 3907/2011 creating an independent Asylum Service and a First Reception Service.

Of importance is Law No. 3907 of 2011 "on the establishment of an Asylum Service and a First Reception Service", transposition into Greek legislation of Directive 2008/115/EC "on common standards and procedures in Member States for returning illegally staying third country nationals" and other provisions (see also AIDA report 2013–2020 for all the legislative changes). L. 3907 serves as an example of Europeanisation, as the governance structure of asylum in Greece aligns with EU practices. L. 3907 ends arbitrary detention in scattered locations that function also as *de facto* reception sites. It establishes a reception service responsible for new structures (frst reception centres), new detention facilities and a streamlined process from arrival to asylum. Two stages of reception emerge, frst and secondary, with L. 3907 explicitly referring to "frst reception". The latter indicates the shortterm supply of basic material conditions and services offered to asylum seekers on arrival, especially those deemed most vulnerable (elderly, women with children, unaccompanied children). In principle, those applying for asylum enter a "secondary reception" framework where material conditions are offered for the duration of their asylum process, while those who opt out and/or are rejected, remain outside

<sup>5</sup>The frst having taken place on accession to the EU.

the scope of any assistance (and often detained). However, it is important to note that these clear distinctions are not always implemented in practice, as asylum seekers do not always access frst and/or secondary reception, despite their eligibility.

The institutional and legislative changes coincide with the worst period of the Greek fnancial crisis (2010–2015). The fscal limitations imposed by Greece's creditors, impact reforms but most importantly, result in a gap between the design of the asylum procedures and reception and their implementation. The various agreements between Greece and its creditors prohibit additional hires in the public sector, limiting the staffng ability of the Asylum Service and the First Reception Service (European Commission, 2014). With staffng shortages and fnancial limitations, the Asylum Service becomes operational only in 2013, and the Regional Asylum Offces (RAO) open gradually over the next few years. The frst RAO outside of Attika, opens its doors in July 2015 in Thessaloniki, with offces in Thrace, Epirus, Thessaly, Western Greece, Crete, Lesvos, Chios, Samos, Leros and Rhodes opening from 2015 through 2019. The First Reception Service opens the First Reception Center (FRC) in the land border of Evros (Fylakio center) in 2013. Fylakio is a screening and registration center, where asylum seekers can remain in dignifed conditions up to 25 days. No FRC exists on the islands in 2015 when the refugee "crisis" unfolds. Throughout the mainland, detention, and pre-removal centers (PROKEKA) are also gradually set up. Asylum seekers whose applications are pending, rejected asylum applicants or migrants that have opted out of the asylum process are often detained for an indefnite period (AIDA, 2014, p.61), evidence of different implementation of the law (Rozakou, 2017). Despite attempts to improve the legislative and institutional framework asylum applicants encounter divergent practices across different entry points on asylum and detention but also in the mainland regarding their reception.

Reception is part of asylum management. It is applicable to asylum seekers only (hence the importance of registration) and thus, a way of distinguishing between the refugee and "unwanted" irregular migrants, with the former worthy of assistance and the latter undeserving of protection (Mantanika & Arapoglou, 2021; Holmes & Castaneda, 2016, p. 17). As prescribed in the Common European Asylum System, reception entails service provisions, and begins with identifcation, screening and submission of an asylum application rendering the individual under the care of the host State. It includes access to legal assistance, counselling, healthcare, shelter, food, education particularly for minors, special accommodation provisions for vulnerable groups, food, hygiene, interpretation. Despite the codifcation in L. 3907 of the material provisions for asylum applicants, most asylum applicants are left unassisted. In September 2014, a total of 1160 reception places are available) operated by 16 NGOs and funded through the European Social Fund (ESF), although the National Centre for Social Solidarity (NCSS) responsible for processing requests for accommodation of asylum seekers registers 4269 requests in that same year. Overall, Greece, at the start of the refugee "crisis" in 2015, lacked a reception system capable of addressing the needs of asylum applicants (accommodation and material provisions) and as a result the emphasis was placed on improving reception conditions and particularly accommodation. As the numbers of those stranded in Greece increased post March 2016, a humanitarian crisis unfolded necessitating a reception system put in place.

#### **16.3 The EU-Turkey Statement 2016 and Its Impact on the Greek Reception System**

Until March 2016, hundreds arrived from the islands to the Greek mainland on a weekly basis, joining thousands already waiting to register for asylum, or trying to organize the next leg of their journey. The Syriza government commits, in late 2015, to 30,000 reception places in collective accommodation schemes (i.e., camps), mostly in the mainland, to respond to the urgent need created by the refugee "crisis" as well as the transfer of the population of the hotspots before the implementation of the EU-Turkey Statement. Greece has a total reception capacity of 36,910 as of March 20, 2016. Around 10,000 are transferred to Northern Greece in newly set up camps in the mainland and/or near Athens (Human Rights Watch, 2016). In total, an estimated 50,000 (UNHCR, 2016) require accommodation and services provision.

Various problems are reported, particularly as regards accommodation provided by the Greek State (in camps set up by the military and in the hotspots), such as accommodation in inappropriate structures (e.g., tents or containers) for a prolonged period; the use of former factories with hazardous industrial residues; inadequate sanitary conditions and the lack of infrastructure, such as hot water and heating during harsh weather conditions (Greek Ombudsman, 2017, pp. 41–45). The Greek state is unable to ensure the provision of decent living conditions for all asylum seekers (*ibidem*), failing to respond to its obligations deriving from European and national law. This left signifcant room for international organisations, NGOs, and cities to step up and function as implementing partners, seeking to offer a holistic reception system that includes accommodation, social and health services provision, and assist, where possible, with integration.

#### *16.3.1 Non-state Actors in Reception*

International organisations do not immediately expand their presence in Greece. UNHCR and IOM's growth and expansion takes place towards early 2016, in parallel with the closure of the Western Balkan route (Dimitriadi & Sarantaki, 2019) and the implementation of the EU relocation scheme. The restrictions imposed on border crossings from end of 2015 through March 2016, combined with the transfer of thousands from the islands to the mainland, create an unprecedented demand for but also a large-scale provision of housing and services. UNHCR and IOM become implementing partners to the EU Relocation programme, with UNHCR responsible for housing those eligible and IOM for their transfer from the islands to the

mainland and/or their relocation fights. The accommodation scheme is known as ESTIA6 (I & II). ESTIA initially provides housing and cash assistance to the benefciaries of the relocation programme. By late 2016 the benefciaries expand to include Dublin family reunifcation candidates and vulnerable applicants. It is the frst attempt to offer holistic reception services to benefciaries. To implement the ESTIA programme, UNHCR cooperates with NGOs and local authorities.7 By the end of June 2020, UNHCR had established 25,792 accommodation places in 14 cities and 7 islands across Greece. ESTIA provides rented housing, in parallel with access to social workers, interpreters, assistance with accessing medical services, employment, language courses and recreational activities but most importantly cash-assistance. The latter reached until 2021, 70,000 asylum seekers and refugees. In December 2020 ESTIA transitioned to the Greek State and is now part of the national reception system (UNHCR, 2020). Similarly, in 2021, cash assistance transitioned under national administration. What began as a distinct activity by international organisations, currently forms almost 50 percent of the Greek state's capacity on reception.

In 2015, responding to the "emergency", NGOs (national and international) redirected their services to assist refugees in Greece (Skleparis, 2015). Many NGOs function as implementing partners to international organisations, in offering various services related to reception. Alongside the formal NGOs, informal/grass-root initiatives emerge, supporting asylum seekers. Self-organised movements of radical Left and anarchist spaces already active in supporting migrants' rights (Papataxiarchis, 2016) step in and occupy abandoned spaces in urban centres, particularly Athens, to address accommodation shortages.

The level of involvement of non-state actors in Greece is unprecedented and requires a continuous negotiation between the Greek state, the international organisations and non-state actors as well as the local level that will eventually also step in and assist with the buildup of reception.

#### *16.3.2 The Local Actors*

Though municipalities and regions have no legal authority over matters of migration including reception, they are an important actor that can either facilitate or hinder the setup of reception services in an area by objecting or supporting the transfer of refugees in their municipal limits. There is no redistribution mechanism of asylum applicants in Greece (Dimitriadi & Sarantaki, 2019). This has not prevented municipalities from participating (Sabchev, 2021a) but it has resulted in a patchwork of responses across the mainland with some cities taking a very active role and others

<sup>6</sup>ESTIA stands for Emergency Support for Integration and Accommodation. It includes urban accommodation and cash assistance. See http://estia.unhcr.gr/en/home/

<sup>7</sup>For example, in 2018 10 NGOs and 9 local authorities partnered with the programme. The number of local authorities has increased since.

actively opposing the presence of asylum seekers (e.g., protests, blocking hotels that would function as accommodation sites). For some cities, like Athens and Thessaloniki, reception is critical as in 2015, they are on the receiving end of signifcant numbers of asylum seekers arriving from the islands. The two cities opt to partner with NGOs and international organisations, drawing from best practices acquired through city networks (e.g., EUROCITIES) (Dimitriadi & Sarantaki, 2019).

The City of Thessaloniki forms the Refugee Assistance Collaboration Thessaloniki (REACT) programme (Sabchev, 2021b), for the reception of asylum seekers and refugees in private apartments in partnership with UNHCR (funded by the EU Civil Protection and Humanitarian Aid). It is, thus, part of the ESTIA programme. The municipality is responsible for organising accommodation places in private apartments and covering the utility costs with different NGOs assisting. Along the lines of ESTIA, it offers a holistic approach to reception.

For the City of Athens, 2015 marks a turning point in how migration is integrated in the administration of the City (OECD, 2018). The position of Vice Mayor for Migrants, Refugees and Municipal Decentralisation is created, indicating that migration is fast becoming a priority despite the absence of a legal framework. There are no accommodation places available in Athens, in 2015. Migrants transferring from the islands end up sleeping rough in public parks and squares (Squire et al., 2017), necessitating the City's engagement with reception. The City of Athens, in consultation with the Ministry for Migration, offers space for the setup of the frst reception center for asylum seekers (Elaionas), established under the auspices of the Ministry of Migration Policy. The lease with the Ministry is renewed annually per agreement by the municipal council, until 2021.8 The city also joins the ESTIA programme, cooperating with UNHCR for the distribution of accommodation spaces in different areas of Athens.

All the aforementioned examples focus on the mainland. However, both NGOs and International Organisations are also active on the islands, and in some cases (e.g., Lesvos) municipalities seek early on to cooperate with non-state actors in establishing reception processes, standards, and services. Kara Tepe on the island of Lesvos is one such example. The camp is created and managed by the Municipality of Mytilene and UNHCR in collaboration with the NGOs that offer for services provision. The camp offers temporary housing for asylum seekers waiting their registration process. Overall, there are fewer examples of partnerships between local and non-state actors on the islands that are unwilling to support further hosting of refugees. This is partly due to the presence of the hotspots but also the EU-Turkey Statement that stranded a large population of asylum seekers and refugees on the islands, thereby transforming them into a buffer zone (Tazzioli, 2018). This resulted in a divergence between mainland and the islands, where different reception practices are applied, largely due to different levels of mobilisation but mainly due to the different legal framework on the islands.

<sup>8</sup>Elaionas is scheduled to close down by early 2023 with the City Council of Athens having voted not to renew the license.

#### **16.4 Beyond Reception: The Greek Asylum System After the Statement**

Asylum drastically changes following the agreement on the EU-Turkey deal of 2016. In the words of a senior manager in the Asylum Service, '*Let's be clear, [Greece] right now is not a best practice, it is a unique practice*" (interview with Asylum Service staff, June 2018 Athens, in Dimitriadi & Sarantaki, 2019), referring to the application of border procedure on the islands and regular procedure in the mainland. The following section tries to identify the major changes that have taken place in Greek legislation and practice. It is important to note that this is not a detailed and exhaustive account of the changes and their impact, but rather a zooming in on specifc elements that the recent proposal on the New Migration and Asylum Pact appears to replicate.

#### *16.4.1 Reforming Asylum to Fast-Track Returns*

A key component of the Statement is that all new irregular migrants crossing to the Greek islands as of 20 March 2016 can be returned to Turkey, following an individual examination of their asylum application on the basis of (in)admissibility.9 The aim is to speed-up returns of Syrians. To implement the Statement, Greece reforms in 2016 its asylum and border procedures through Law 4375/2016, (adopted in April 2016). L.4375 provides for shorter deadlines and fewer safeguards to the asylum procedure and simultaneously establishes complex processing for different types of applicants. Fast-track border procedure initially applies only to the Syrian nationals. Gradually the procedure is extended to nationalities with a rate of international protection under 25 percent (AIDA, 2017). In contrast to the Syrians, the asylum applications of the latter are examined on merit. All other asylum claims are examined on the basis of merit through the regular procedure.

The law also regulates the organisation and operation of the Asylum service and Appeals Authority. The latter consistently rejected inadmissibility decisions, deeming Turkey not a "safe third country". The reform alters the composition of the Appeal Committees, now made up of judges (previously lawyers and trained staff), in an effort to enforce the provisions of the Statement. L. 4375 further undertakes administrative changes, renaming the First Reception Service into Reception and

<sup>9</sup>Admissibility is the procedure thanks to which the Member State that receives an application decides whether it is responsible to examine it or not. Per Statement, Greece can deem inadmissible- i.e. not examined on merit, the applications of Syrians since they are eligible for protection in the country they arrived from, in this case Turkey. Inadmissibility can also be applied to individuals that have remained in Turkey without facing persecution or risk of harm prior to their arrival in Greece. Merit refers to the actual examination of the person's claim to check if they fulfll the criteria to be recognized as a refugee.

Identifcation Service (RIS), responsible for overseeing the RICs (hotspots) and establishes the legal framework for the operation of the RICs.

The law transposes the recast Asylum Procedures Directive, establishes the Reception and Identifcation Centers (RICs) of the fve islands of Eastern Aegean (known as the hotspots) and clarifes the application of a fast-track border procedure on the maritime border (Petracou et al., 2018). Fast-track procedure is based on the examination of the admissibility of the asylum claim, rather than the merit. In the L. 4375 it is an exceptional procedure, applicable only to the fve islands of Samos, Kos, Leros, Chios, Lesvos for those arriving after 20 March 2016 and subject to the EU-Turkey Statement, and in particular *vis-à-vis* Syrians. Since April 2016, asylum seekers on the islands are subject to a "restriction of freedom of movement" decision issued by the Head of the RIC (FRA update 2019) which is revoked once registration completes. A deportation order is then issued, which is also suspended but a geographical restriction is automatically imposed on the island and/or in the RIC until the completion of the asylum (Greek Council for Refugees, 2018) or the return procedure.

On the islands, the screening takes place in conditions of detention and with limited legal assistance provided; interviews, particularly of Syrians under the accelerated border procedure, often take place at the moment of disembarkation, with interpreters frequently speaking different dialects (Dimitriadi & Sarantaki, 2019), and with little information offered to the applicants. The asylum interview is conducted by either Asylum Service or EASO staff. The role of EU agencies is not clarifed in the Greek legislative framework (AIDA, 2017) and remains murky to this day. Particularly EASO has been criticized for the quality of its recommendations, with the EU Ombudsman highlighting that "[…] when conducting admissibility interviews EASO fails to comply with the provisions on "the right to be heard" in the Charter of Fundamental Rights (Article 41), as well as EASO's own guidelines" (European Ombudsman, 2017). The fast-track asylum procedures in the period 2016–2019, have a time limit of 2 weeks (AIDA, 2017) that is impossible to meet. In fact, the entire asylum process lasts for years in most cases, with the exception of the Syrians whose claims are routinely prioritized, in an effort to implement the Statement.

The design of the Statement envisages swift processing and swift returns that will function both as a deterrent in reducing the number of irregular migrants but also ensure that the hotspots have space for new migrants. In practice, this does not happen. By the end of January 2020, only about 2000 people are returned to Turkey in line with the Statement. Pakistanis make up the largest group of returnees, many volunteering for return. Absence of returns impacts reception. While the capacity of the hotspots remains the same, the population increases and there are no transfers to the mainland (until 2019). Conditions deteriorate rapidly, and migration is a key campaign issue in the national elections of 2019 that results in the victory of the center-right party of New Democracy. Priority is given to legislative changes to support further the implementation of the Statement and reduce not only the number of arriving migrants but also the overall number of asylum seekers in Greece.

#### *16.4.2 The International Protection Act*

On 1 January 2020, Greece's International Protection Act (IPA) enters into force (voted in 2019, Law 4636/2019). It is Greece's ffth legislative asylum reform since the implementation of the Statement (GCR & Oxfam, 2020). The IPA seeks to regularise aspects of the Statement deemed exceptional measures in the previous legislation, and to accelerate the implementation of the Statement and develop further a restrictive asylum system.

Fast-track border procedure is not new (see above discussion on the L 4375/2016) but it is established initially as an emergency process that the IPA now normalizes. The law foresees that the fast-track border procedure "can be applied for as long as third country nationals who have applied for international protection at the border or at airport/port transit zones or while remaining in Reception and Identifcation Centres, are regularly accommodated in a spot close to the borders or transit zones" (Article 90 (3) IPA; AIDA, 2021). (In)admissibility is determined based on Turkey being a "safe third country"10 or a "frst country of asylum" for the applicant. Those inadmissible are also returnable. The fast-track border procedure is now disentangled from the Statement and can be applied to any location meeting the above requirements (for criticism of the IPA see ECRE, 2019; UNHCR, 2019; Greek Ombudsman, 2019). The new law, among others, severely restricts asylum safeguards, increases detention, limits the right to family unity and marks a return to the period of 2012–2014, when Greece prioritised deterrence and detention of migrants including asylum seekers.

Further amendments were added in May 2020. The registration and examination of all new asylum applicants in 2020 becomes a priority. As a result, those who arrive prior to 2020 and whose interviews are scheduled for 2019 and 2020, have their appointments postponed, in some cases for 2022 and 2023. The Covid-19 pandemic further exacerbates the situation, with some applicants receiving dates as late as 2025. Until such time, they will remain in detention in one of the new RIC and PROKEKA11 centres being built on the islands and land border. Nationality differentiation appears to be the norm. In some cases, AIDA reports (2020) that asylum seekers with specifc profles (for example from Eritrea, or vulnerable groups from Afghanistan) are granted refugee status without an asylum interview. Others are rejected following a quick and brief interview on disembarkation. These diverse

<sup>10</sup>Greece did not have a list of safe third countries before June 2021. Instead, the asylum service applied the criteria on an individualised basis of the safe third country rule that already existed in Greek legislation per transposition of the recast Asylum Procedures Directive and only for Syrians that in principle could access protection in Turkey under the Temporary Protection regime. Turkey was only formally declared a safe third country on June 7, 2021, for asylum-seekers originating from Afghanistan, Bangladesh, Pakistan and Somalia, in addition to Syria.

<sup>11</sup> In 2022 the procedure requires that migrants that disembark on the islands or land border go through the Reception and Identifcation Center and those whose application is rejected, will be transferred to a PROKEPA (predeparture center) situated within the broader space of the RIC.

practices, often random and dependent on which regional offce processes the asylum application, produce an unprecedented level of diversity in decisions.

The numerous changes in the administrative aspects of asylum processing make navigating the asylum system impossible without legal assistance. The law allows for the rejection of an application without interview, if there is no available interpreter in the language spoken by the applicant. It "foresees an extended list of cases in which an application for international protection can be rejected as 'manifestly unfounded' without any in-merits examination and without assessing the risk of *refoulement*" (AIDA 2020). In addition, the IPA introduces the notion of "fctitious service" (*πλασματική επίδοση*, AIDA 2020) of frst instance decisions. An appeal against a frst instance decision should be submitted in a written form (in Greek) and it should mention the "specifc grounds" of the appeal. Otherwise, the appeal is rejected as inadmissible without further examination. In practice, only those with access to lawyers can launch an appeal. Though legal assistance is free, it is extremely rare that applicants have access to it.

Throughout the appeal, applicants remain in administrative detention12 until their asylum is fully processed. A link is now established between rejection and return. If the appeal is rejected, the Independent Appeals Board issues a return decision at the same time as the rejection; thus, administratively the two are now processed by the same authority. In principle, this facilitates the process of returns. If rejected on appeal, the individual has the right to request cancellation before a court, but the request does not have a suspensive effect, i.e., it is possible to deport the individual while the examination is pending.

Overall, the impact of the Statement on the asylum procedure should not be underestimated. The Statement began a process whereby asylum increasingly changed, differentiated based onentry points and nationality. It is a process that has slowly come to be normalized and, despite the severe impact the Greek model has had on asylum seekers rights, it is partially adopted in the New Pact on Asylum and Migration.

#### **16.5 The New Pact on Migration and Asylum: Greece an Inspirational Model for Europe?**

A few days after the catastrophic fre in Moria in September 2020, the New Pact on Migration and Asylum13 was announced. The Vice President of the Commission, Margaritis Schinas, described the Pact as a house with three foors. The frst foor is the external dimension; the second foor are the front-line states that apply border

<sup>12</sup>Though the law transposes the minimum requirements of EU law, detention became the de facto policy for all new arrivals (Papastergiou, 2021).

<sup>13</sup>Five legislative proposals make up the New Pact on Migration and Asylum: the Asylum and Migration Management Regulation; the Screening Regulation; the amended Asylum Procedures Regulation; the amended Eurodac; the Regulation addressing situations of crisis and *force majeure*.

procedures seeking to identify those who can be returned to the frst foor. And the third foor is the internal dimension, of solidarity between Member States. The Pact establishes a pre-entry phase, where several elements from the Greek practice since 2015 are incorporated.

#### *16.5.1 Screening*

The pre-entry phase includes a screening procedure, upon which the remaining process hinges on. Screening is proposed for all third-country nationals who reach the external borders and/or after disembarkation or Search and Rescue. Here the screening adopts the hotspot approach. It includes a health and vulnerability screening, identity check, registration, security check.14 EU agencies, from Frontex, Europol and EASO alongside national authorities in the hotspots undertake the screening and registration of the individual that has the right to apply for asylum but only after the process concludes. Screening should be completed within 5 days. Though the proposal incorporates as standard practice the hotspot approach, the experience of Greece suggests this is unattainable, particularly in times of large-scale arrivals where the screening sometimes takes weeks if not months. On the other hand, systematic registration in the EURODAC is achieved through the hotspots, and this is part of the aim of the pre-entry phase proposed in the New Pact.

#### *16.5.2 Border Procedure*

Once screening is completed, an accelerated border procedure is triggered and made mandatory for most cases. This is already the current practice in Greece since the IPA came into force. The accelerated border procedure becomes obligatory in the Pact for all applicants arriving irregularly through the external border and/or in a Search and Rescue. Thus, it is normalised, instead of an exceptional measure. According to the EU Commission, the aim of the screening, is to "accelerate the process of determining the status of a person and what type of procedure should apply". It will also render diffcult to abscond after entrance in the territory; this is also the justifcation utilized by Greece for the restriction of movement imposed on individuals. Yet, the border procedure has already been found controversial; as noted by the EU Fundamental Rights Agency (FRA 2019) "almost three years of experience [of processing asylum claims in facilities at borders] in Greece shows, [that] this approach creates fundamental rights challenges that appear almost insurmountable".

<sup>14</sup>The security check takes place through the Schengen Information System to ensure the person does not constitute a threat to internal security.

The border proposal in the New Pact incorporates the options for the refusal of entry (Art. 14(1) screening proposal) as well as expulsion. In principle a bias is established from the moment screening takes place, based on one's nationality. Individuals arriving from a country with recognition rates below 20 percent, as well as those coming from a "safe country of origin" or "safe third country" (Cassarino & Marin, 2020) will be referred to the accelerated or/and border procedure, instead of the regular procedure. The aforementioned proposal is not unlike what takes place (or is designed to take place) in the Greek borders (in Greece currently the threshold is 25 percent, see discussion above on the IPA).

In the Pact, during the screening procedure and accelerated border procedure, asylum applicants are considered as not having entered the country (Article 4, draft Screening Regulation, Article 40, new draft Asylum Procedure Regulation). A version of this already applies in Greece, through the Statement. The geographical restriction of movement on the Greek islands that prevents asylum applicants from reaching the mainland, limit as much as possible entry to the country, with the islands functioning as buffer zones. Countries that designate transit zones for prescreening, will likely do so before border areas or at the actual border, which raises the potential for many "Moria" style camps to emerge where asylum applicants wait for their pre-screening to complete and/or for their deportation to take place.

Finally, the Asylum Procedure Regulation proposes that a rejected application should be accompanied with a return decision. It is an administrative model already applicable in Greece15 since the IPA. As in Greece, in the original proposal of the New Pact, there is no suspensive effect of appeal established. Instead, the individual's appeal can be processed following her/his return.

#### *16.5.3 Force Majeure*

One of the least discussed elements of the proposal of the New Pact is the introduction of the right to derogate from the rules due to *force majeure*. It was proposed by Greece, in a non-paper circulated during the negotiations (ΑΜΝΑ, 2020).16 In February 2020, thousands of migrants moved towards the Greek-Turkish land border of Evros. In response, Greece closed its border, and suspended access to asylum in its territory, in violation of international law. Greece announced it was invoking article 78 (3) to justify the suspension of asylum, even though this would require a proposal to be submitted by the EU Commission and agreed on by the EU Council. The Greek proposal to the New Pact was to establish a mechanism that allows for derogation from the rules in case of crisis.

<sup>15</sup>The other countries applying similar practices are Finland, France, Lithuania, Luxembourg, the Netherlands, and Slovenia.

<sup>16</sup> https://www.amna.gr/en/article/463581/Koumoutsakos-proposes-emergency-fexibility-clausesfor-European-Migration-and-Asylum-Agreement

In the New Pact *force majeure* triggers several derogations from the Asylum Procedures Regulations but also as regards registration of applications for international protection. The problem with the *force majeure* is that it allows for derogation from the rules in an emergency that is left undefned. If triggered at times of an increase in asylum seeking fows, it will likely impact the asylum applicants that will remain stranded in multiple "Moria-type" centers along the external borders. Yet elements of the *force majeure* have already been tested in Greece following the events in Evros in March 2020.

The aforementioned elements suggest that, to an extent, Greece serves as an inspirational model, particularly as regards the strengthening of the links between asylum and return.

#### **16.6 Conclusion**

The impact of the EU-Turkey Statement on asylum seekers has been well documented, yet the impact on the Greek asylum regime has been less explored. The chapter has highlighted the attempted responses to unfolding emergencies, contributing to the argument raised by Finotelli and Ponzo that a rather than Europeanisation a "hybridisation of strategies, logics of action, and practices" (see Chap. 1 in the volume) is witnessed since 2015. In Greece this kicks off as a response to the divergence between mainland and border areas in asylum processing; a product of the EU-Turkey Statement.

The legislation aims to facilitate the implementation of the Statement on the maritime border. In the period 2015–2019 this produces delays and inconsistencies particularly between border areas and mainland (exempt from the Statement). The limited capacity of the Asylum Service (and Appeals) poses an additional challenge, since it is unable to meet the increasingly high level of applications launched throughout the period in question. Overall, the asylum legislation post- 2016 seeks to implement a more restrictive and deterrence-based approach. This logic of deterrence, initially a response to the refugee "crisis", is now formalised and normalised in the IPA and its amendments.

Although reception emerges in the same context as asylum processing, the former is driven by a logic of humanitarian emergency (Dimitriadi & Malamidis, 2020). It is oriented towards services-provision and accommodation, in an effort to address the needs created from the swift transfers of all the population in March 2016 from the islands to the mainland, the relocation mechanism and the border closures. Unlike asylum, the state and its institutions have little role in the development of the reception system that develops mainly by non-state actors, including cities, international organisations and NGOs. In 2021 the reception system becomes fully part of the national reception system. As numbers of arrival are at an all-time low, little progress has been made to move away from the "crisis" model into a sustainable secondary reception system that results in integration. Overall, despite the positive steps undertaken between 2015–2019, reception today falls short of what is

on offer in other EU Member States and is gradually taking backward steps. This is in line with the overall logic of deterrence, increasingly applicable to all elements of the asylum regime since 2020. The aim is to render Greece an undesirable destination for asylum applicants, by reducing available spaces of accommodation, closing many of the camps that offer shelter to applicants, and limiting who is entitled to cash assistance (only those residing in the few remaining state-run accommodation sites).

Despite the shortcomings of Greece on reception, there are aspects of the asylum regime that render Greece a "pioneer" and inspirational model for the EU, as evident in the initial proposal of the New Pact. Accelerated border procedures, prescreening, and derogations from the rules in case of "emergencies", to name a few, suggest that the deterrence logic applied in the country is very much in line with the EU and its Member States, and a potential-albeit problematic- model for the way forward.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 17 Concluding Remarks: Towards a New Conceptualisation of Similarities and Differences in European Migration Controls**

**Claudia Finotelli and Irene Ponzo**

#### **17.1 From the North-South Divide to Blurring Boundaries**

This volume challenges the idea that the North-South divide is the deepest cleavage in the way European countries manage migration fows, and, with that, contests the dichotomous thinking that pervades the literature on migration control. By analysing cases from both geographical areas – particularly Germany and the Netherlands, on one hand, and Italy and Spain, on the other – each section has revealed the *shared* logics of action, strategies and practices on the ground pointing to unexpected similarities and blurred boundaries among European migration regimes.

Relevant similarities also emerge in areas where we would expect the difference between the two blocs to be larger in light of structural constraints and the low level of Europeanisation. Labour-market policies and internal controls are the primary examples of this. Jan Schneider and Holger Kolb, for instance, showed how Germany's labour policies have come back to target the origin of immigrants as a selection criterion after 20 years, during which time they had been driven by the guiding principle of individual qualifcations. As the authors noted, distinctive traits of Italian and Spanish labour migration policies such as "source-country particularism" and "preferential bilateralism" as well as the coexistence of universalistic (i.e. country-blind) and particularistic principles, have taken on increasing importance in Germany. At the same time, Camilla Devitt demonstrated how Italian labour migration policy has shown growing similarities with Northern European states: the country is increasingly reliant on the free movement of workers from Romania and non-EU family migrants for low-medium skilled labour needs, while adopting a

C. Finotelli (\*)

Complutense University, Madrid, Spain e-mail: cfnotel@cps.ucm.es

I. Ponzo

FIERI (International and European Forum of Research on Immigration), Torino, Italy e-mail: ponzo@feri.it

more restrictive approach to infows of non-seasonal workers from outside the European Union. Similarly, Jeroen Doomernik, Blanca Garcés-Mascareñas and Berta Güell showed that both in the Netherlands and in Spain, the seasonal labour demand in agriculture has increasingly been met by migrants already living in the respective country and circulating Eastern European workers, against a background of increasing deregulation of the sector. Overall, the boundaries between labour migration policy of Northern and Southern European countries appear more blurred than expected.

Other chapters highlight similarities regarding the management of irregular residence. Gabriel Echeverría, while acknowledging the substantial differences between Spain and the Netherlands due to deep-rooted political, cultural and historical dynamics as well as different underlying policy objectives, showed how policies towards irregular migrants in both countries have been driven by societal demands. Indeed, the shift in the Netherlands from a more relaxed to a stricter enforcement of internal migration controls during the late 1990s should not be attributed to a greater generic propensity to stick to the rules, but rather to a long-term process of deteriorating social and political acceptance of migrants as well as a slow transformation towards a more skilled and more formal labour market. Similarly, Jorge Malheiros and João Peixoto argued that in Portugal, the changing approaches and declining consensus regarding regularisations have been driven by changing societal demands. Moreover, they demonstrate how the so-called Southern European migration regime is both less exceptional and less homogeneous than it is generally presented. First, Malheiros and Peixoto pointed out that the principle of general mass regularisations was also adopted in the 1990s and 2000s by Northern European countries such as France, Austria, Luxembourg and Netherlands, albeit in a more limited way. Second, the system in Portugal and Spain – as well as in France and Switzerland – evolved towards a case-by-case approach to regularisation during the 2000s, while this did not happen in Italy and Greece undermining the idea of a common Southern European model.

Other similarities emerge in the welfare-migration nexus, driven by social expectations regarding equal access to welfare. Welfare scroungers coming from other EU countries are perceived as a problem in both Northern and Southern countries, though this problem takes diverse forms at different latitudes. South-North migration, especially of young people, has resumed as a consequence of the Great Recession and has contributed to the overall growth of intra-EU migration statistics. This has helped to sustain anti-EU concerns, as witnessed in the UK (see Alessio D'Angelo). On the other hand, the healthcare costs generated by Northern European retirees moving to Southern Europe has contributed to a redefning of healthcare access among non-labour-motivated intra-EU migrants, as the case of EU retirees in Spain shows (see Claudia Finotelli). In both cases, the hostility towards intra-European migration was not so much framed in terms of generic xenophobia, but rather as welfare chauvinism; this has led to efforts to redefne the migration-welfare nexus in exclusionary terms. Overall, the stratifcation of welfare rights has materialised as a mechanism to regulate the entry and settlement of EU migrants in an increasing number of national migration regimes. However, the actual effects were

far more extreme in the United Kingdom, where the stratifcation of welfare rights became a deeply rooted political project that ultimately led to the country's exit from the EU.

Alongside accusations of being the origin of welfare scroungers, Southern countries have periodically been blamed for being Europe's "soft underbelly" for immigrants. This is hardly true nowadays. Irene Landini and Giuseppe Sciortino's chapter highlighted how Southern European countries have been progressively incorporated into control mechanisms shaped by the control objectives of Northern European countries, which are mainly aimed at preventing the transit of unwanted infows of asylum seekers from Southern European borders to Northern European countries where they could apply for asylum. Against this backdrop, Southern European countries ended up imposing visa obligations to former colonies even (see Lorenzo Gabrielli) and countries with privileged tourism and foreign policy relations (see Federica Infantino) Infantino also noted that Italian consulate offcials have increasingly understood migration control as a key mission, and have adopted restrictive practices that frame migration as a "risk", just like their Northern European peers. Interestingly, as Landini and Sciortino highlighted, convergence dynamics can be observed not only in the case of a relatively successful harmonisation area (such as visa policy), but also in areas with very low harmonisation (such as the enforcement of returns). In such cases, however, convergence does not indicate increasingly similar policies, but similar policy outcomes. The diffcult implementation of readmission agreements, the stagnating cooperation with countries of origin and transit, administrative diffculties and liberal constraints keep return rates low in all European countries.

Moreover, some Southern countries have gone beyond the mere convergence towards the policy objectives and practices of the Northern ones: they have developed some of the strictest asylum and border control management procedures in Europe. This is the case in Greece, where newcomers arriving since 2016 have been kept in conditions similar to detention and cannot leave northern Aegean islands until their asylum application has been processed (see Angeliki Dimitriadi), not to mention violence, including rape, taking place in the overcrowded reception camps. Spain, with its wall at the border of Ceuta and Melilla as well as the externalisation of border controls (often entailing violent control practices) to Morocco, acts as an equally draconian guard of European borders (see Gabrielli).

The so-called European refugee crisis revealed further unexpected dynamics. Since then, Italy has changed how it frames migration and has started to perceive itself as a main destination for refugees. As a result, the country abandoned its traditional approach based on *ad hoc* emergency solutions, instead developing structural measures so that nowadays all people claiming protection go through the ordinary asylum procedures and reception in public facilities, irregular migrants and asylum seekers are systematically fngerprinted, and the country has fully joined Northern European countries' call for greater responsibility-sharing (see Irene Ponzo). At the same time, the German asylum system managed infows far less effciently than expected in 2015–2016. Newcomers were housed in tents, and the Federal Offce for Migration and Refugees (BAMF) was exposed to harsh criticism

for long delays in applicants' registration and the poor management of asylum decisions that were largely reversed by courts (see Dietrich Thränhardt).

Next, since the establishment of the EU Trust Fund for Africa in 2015 Valletta Summit, Northern European countries, which had traditionally been marginal actors in border externalisation projects leaving the stage to Southern European states, have become more active in the African context (see Lorena Gazzotti, Mercedes Jiménez Álvarez & Keina Espiñeira). Moreover, the principles guiding relations with third countries in the feld of migration are becoming increasingly homogeneous across Europe (see Gazzotti, Jiménez Álvarez & Espiñeira; Gabrielli). These principles include effective control over the arrival of irregular migrants through cooperation with origin and transit countries; economic development in origin and transit regions to deter (irregular) emigration; and development cooperation and trade preferences serving as the main levers to secure third-country commitments in the feld of return and readmission.

In light of such complexity, chapters in this book point to the emergence of a surprising trend, with some migration control approaches developed by Southern European countries actually inspiring the rest of the EU. One case in point is Spain's approach to the external dimension of migration policies. As Gabrielli explained in Chap. 5, during the 1980s and 1990s, the Iberian country adopted its frst immigration law, introducing visa requirements for the countries of Maghreb and starting fortifying Ceuta e Melilla in order to meet EU requests and join the Schengen Area. Nowadays, Spain has instead become a model, inspiring the EU in various ways, such as by introducing "migration conditionality" in its relations with third countries, linking development aid, foreign direct investment and trade, etc. with cooperation in migration issues.

Similarly, Greece, the once worst example of asylum practices in Europe, clearly served as a model for the New Pact on Migration and Asylum presented by the European Commission in 2020, as Dimitriadi illustrated in Chap. 16. As a matter of fact, several elements from the Greek practice have been incorporated into the Pact, such as the hotspot approach to carrying out pre-entry screening, accelerated border procedures as the standard procedure rather than an exceptional one, and the simultaneous issuing of application rejection and return decisions with no suspended effect of appeal. More generally, the Pact considers asylum applicants during the screening and accelerated border procedure as if they had not entered the territory, similarly to what happened in Greece after the signing of the EU-Turkey Statement in 2016 that blocked thousands of people on the islands, preventing them from "entering" the mainland. Moreover, the Pact allows for a derogation from the ordinary rules due to *force majeure*; Greece did precisely this in February 2020, when thousands of migrants moved towards the Greek-Turkish border of Evros; the derogation was then proposed by the Hellenic country in a non-paper circulated during the Pact's negotiations.

With less success, Italy has been trying to shape EU migration policies as well, especially with regard to joint management of mixed infows. As Ponzo explained, the *Mare Nostrum* Search and Rescue operation started by Italy in 2013 was the forerunner of the subsequent EU's *Triton*, *Themis* and *Sophia* operations.

Overall, Southern European countries have changed their policy frames and accepted their geopolitical fate as the guards of Europe's external borders. Obviously, this shift has deeply affected their strategies, policies and practices, and turned them from recalcitrant recipients of EU norms to active shapers of the EU migration regime, suggesting that a "Southernisation" of models of action may be at work in certain felds.

To conclude, the empirical evidence provided in this book challenges the classic North-South divide, and shows how European boundaries are blurrier than expected. This is the outcome of two combined dynamics: the converging composition of migration infows and Europeanisation, as Sects. 17.2 and 17.3 illustrate below.

#### **17.2 The Converging Composition of Migration Infows**

With globalisation, the number of countries of origin of non-European fows has been increasing while migrants have started to concentrate in a reduced number of destination hubs (Czaika & de Haas, 2014). In such a context, Europe has turned into an immigration magnet where, despite the diversity of individual EU Member States, migration fows have been increasingly converging in terms of the composition and types. This is largely due to exogenous factors that do not concern migration regimes specifcally, but have changed the composition of the migrant population over time, with a signifcant impact on migration policies.

First, the immigrant population in both Northern and Southern Europe has become more and more European in the last 30 years. In fact, Europeans accounted for more than half of the immigrant population in Europe in 2017 (UN, 2017). Eastern European Union enlargement and the remarkably Eurocentric orientation of the European Neighbourhood Policy, with its visa facilitation for non-EU Eastern European countries, certainly had a major impact on migration fows and their composition, turning Eastern Europe into the main source of foreign labour for almost all Southern as well as Northern EU countries.

This has affected both labour migration policies and the management of irregular migration. Specifcally, those processes have led migrants to exploit EU internal mobility and easier circulation paths for Eastern non-EU countries; as a result, the need for non-European immigrants declined and recruitment rules for them became stricter. Those trends ended up producing a sort of "ethnic turn" in labour migration policies in both Southern and Northern Europe that clearly looked East (see Devitt; Schneider and Kolb; Doomernik, Garcés-Mascareñas & Güell). At the same time, the relaxation of visa rules for Eastern Europeans promoted by the German Federal Foreign Offce at the beginning of the new century as part of the European Neighbourhood Policy facilitated irregular circular migration systems from Eastern European countries, especially Ukraine and Moldova (see Malheiros & Peixoto). Although this system is different from that of Southern Europe where Eastern Europeans usually overstay their visa and then wait for the next regularisation (see Finotelli; Malheiros and Peixoto), the logics according to which Northern and Southern Europe have exploited irregular movement from Eastern Europe are similar.

The economic crisis that erupted in 2008 was another turning point in the shifting composition of migration infows, as it curbed the huge appetite for newcomers in Southern European countries. In the 2010s, Italy and Spain (together with New Labour's UK) had been the top importers of immigrant labour in Europe. With the Great Recession and public debt crises, the GDP of Spain, Italy, Portugal and Greece fell; labour demand followed, leading to a drastic growth of migrant unemployment in all Southern European countries (Ponzo et al., 2015; Finotelli & Ponzo, 2018). As Devitt's analysis of the Italian case study showed, the economic crisis allowed Italy to fulfl the bulk of its reduced foreign labour needs through a set of functional equivalents, including immigrants already living in the country, intra-EU mobility, family and forcibly displaced migrants. Hence, the decrease in the demand for foreign workers allowed for restrictions in labour recruitment policies towards third-country nationals (TCNs).

The same happened for agricultural seasonal workers in Spain, as the chapter by Doomernik, Garcés-Mascareñas and Güell makes clear. With unemployment rising quickly, the Spanish government decided to freeze circular migration programmes such as the GECCO in order to promote employment of those workers already in the country. In truth, arrivals from third countries did not stop completely, but indeed decreased, partially compensated by migrant workers who were already resident in Spain but had been made unemployed by the economic crisis, together with EU migrants coming from the new Member States such as Romania and Bulgaria.

The Great Recession might even have contributed to closing the gap between Northern and Southern European countries in terms of irregular migration. As Echeverría showed for the case of Spain, employment opportunities for irregular migrants became very limited during the economic crisis; this caused many of them to leave the country and thus reduced the labour supply, so that irregular migration fnally appeared to be contained. This change also opened up opportunities for tougher policies against irregular migrants: the rightist government that came into power concomitantly with the abrupt collapse of the economy exploited rhetoric against irregular migration and started a sensationalist yet ineffectual policy of identity controls, selling these as a migration control success when, in fact, it was the result of self-regulating socioeconomic dynamics. Portugal followed a similar path, as highlighted by Malheiros and Peixoto: the shrinking of the Portuguese economy, especially between 2011 and 2014, led to a substantial reduction in the supply and arrival of irregular immigrants. The political dynamics of the Lusophone country followed those of Spain: once the socioeconomic decline became severe, the prevailing dominant logic of consensus around migration and regularisations gave way to more Left-Right polarisation in policy discourse and in Parliamentary voting. This confrms the hypothesis formulated by Malheiros and Peixoto that irregular migration levels depend on economic cycles and the type of economic demand, rather than on the implementation of policy mechanisms facilitating regularisation.

Finally, the Arab Spring and the Syrian civil war triggered other remarkable changes. The increase of asylum seekers in Southern Europe due to the geopolitical instability in Mediterranean third countries since 2011 and the halt to secondary movements as a result of the Schengen's suspension in several Member States at the end of 2015, has made the composition of infows more similar to that of Northern countries. This in turn has affected how Southern European countries perceive and frame themselves, i.e., from transit contexts to refugees' fnal destinations (see Ponzo). At the same time, turning a blind eye to asylum seekers is not always a viable option: their management requires greater state intervention (from initial screenings to the processing of claims, to reception) than that needed for labour and family migration. Because of this, a *laissez-faire* approach becomes almost unfeasible, especially when arrivals suddenly spike, as happened in 2014, and the Common European Asylum System (CEAS) came into force. Mediterranean countries ended up adopting some of the solutions already implemented in older asylum countries, while developing fresh strategies to deal with the challenge posed by an increasingly complex external blue border (see Ponzo; Dimitriadi).

In sum, factors exogenous to migration policies – such as the Eastern Enlargement, the European Neighbourhood Policy, the Great Recession and the Arab Spring – changed the composition of migration fows, and consequently contributed to further blurring the boundaries between Northern and Southern European migration regimes. The liberalisation or facilitation of circulation for Eastern Europeans (EU and non-EU alike) made them the main source of foreign labour, with corresponding restrictions for less-needed non-European TCNs in both Northern and Southern European countries. Then, the 2008 economic crisis substantially reduced the demand for foreign workers, allowing for further curbs in TCN recruitment. More recently, the growth of asylum fows towards Southern Europe has inevitably triggered new ways by state and non-state actors to observe and act upon migration fows. Finally, the war between Russia and Ukraine could presumably become another external factor affecting infows and fostering convergence of immigration policies.

#### **17.3 Europeanisation as Hybridisation**

Another explanation of the blurring of boundaries between North and South has to do with Europeanisation. If we conceive of the European migration regime as a space of negotiation where several actors as well as practices – formal and informal – are involved (Klepp, 2010), we can bring together elements that are generally analysed separately: on one hand, the formal and informal negotiation processes between Member States; on the other hand, a sort of "Europeanisation by practice" or "ways of doing things" (Radaelli, 2004; Schmidt & Radaelli, 2004; Musliu, 2021) taking place at the implementation stage.

As for the frst type of dynamics, the most well-known process is that by which Northern countries have pushed their practices upon Southern countries through political pressure and by "uploading" their policy preferences at the EU level (Boswell & Geddes, 2011). Against this backdrop, the incentive to join the Schengen Area has apparently played a greater role than legal harmonisation *per se*. Especially in the 1990s, Southern European countries, as the new "border guards", were well aware of the European need to balance free internal circulation through stricter external controls (see Gabrielli; Ponzo). Concerns related to the enforcement of the Schengen Agreement led Northern European countries to impose their preferences on their Southern European counterparts through the formulation of common lists of countries whose nationals required a visa and the approval of a common visa code that defned procedures and conditions for issuing short-term visas.

Surprisingly, Schengen membership has remained a key priority even today. In 2011, immediately after the increase in the secondary movements that followed the Arab Spring, the EU DG for Justice and Home Affairs explicitly accused Italy of violating the "Schengen spirit" and opened up a debate on the revision of Schengen governance as well as at the end of 2015 several European countries has restored border controls (see Ponzo).

However, Southern European countries should not be viewed as passive recipients of pressure from the North and the EU. For instance, despite the pressure from Northern European countries, Italy left substantial components of the Common European Asylum System, such as the Dublin and EURODAC Regulations, on paper until their implementation became a currency of exchange to obtain asylum seekers' relocation in 2015 (see Ponzo). Similarly, Spain has engaged Europeanisation in a way consistent with and functional for its own interests. In the 1990s, it took symbolic measures to meet the requests of European partners by starting, with signifcant EU funding, the erection of border fences around Ceuta and Melilla, while in more recent years it has exploited the EU's growing emphasis on externalisation, leveraging its established experience in this feld (see Gabrielli).

Alongside EU directives and regulations, Europeanisation also develops via "soft" tools, e.g. non-legally-binding guidelines, such as handbooks and instructions for processing visa applications at consulates which enter into the realm of national sovereignty (see Infantino), as well as the 2015 European Agenda on Migration that introduced the "hotspot" approach in Italy and Greece (see Dimitriadi; Ponzo). Similarly, the Global Approach to Migration (GAM) adopted in 2005, renamed the Global Approach to Migration and Mobility (GAMM) in 2011, as well as the 2015 EU Trust Fund for Africa (EUTF) established at the Valletta Summit, have been key tools of the EU migration containment strategy (see Gazzotti, Jiménez Álvarez & Espiñeira; Gabrielli). Those soft instruments, often regarded as secondorder tools of Europeanisation, might matter more than legally binding EU directives, and allow the Union to introduce regulations where it would be otherwise impossible, expanding Europeanisation beyond the scope laid down in the Treaties (Cini, 2001; Ponzo, 2022; Roman, 2020).

That said, Europeanisation happens to a large extent at the implementation stage, where logics of action and practices are constantly hybridised. Here, the role of street-level bureaucrats and NGOs in the development of the European migration regime is signifcant, and triggers dynamics that go beyond Member States' centralised strategies and control. For instance, growing similarities in implementing visa policies have derived from recurring formal and informal exchanges between streetlevel bureaucracies of Northern and Southern European countries' consulates. These have led to the sharing of narratives, frames, meanings and practical knowledge, and have triggered policy change from below (see Infantino).

Another transformative dynamic originates from the role of international organisations that have compensated for the weak state capacity in Italy and Greece at the onset of the European refugee crisis. In Greece, UNHCR and IOM became implementing partners of the EU relocation programme in 2016 without the direct involvement of the national government, with UNHCR setting up the ESTIA programme – initially designed to provide housing and cash assistance to benefciaries of the relocation programme and then expanded to include Dublin familyreunifcation candidates and vulnerable applicants (see Dimitriadi). Similarly, in Italy, UNHCR has played a signifcant, though declining role in raising standards and flling the gap with the rest of Europe by participating in almost all asylum phases, from disembarkation to the examination of asylum applications, from refugee integration to assistance in the drafting of codes, handbooks and national plans (see Ponzo).

In this "Europeanisation by practice", while international organisations act somewhat like state entities, some state entities increasingly resemble nongovernmental organisations in their logics of action. With regard to the external dimension of migration policies, we have observed a sort of NGO-isation: the tendency of Member States' agencies to act like NGOs insofar as they become the implementing actors of EU development funding (see Gazzotti, Jiménez Alvaréz & Espiñeira). For instance, EUTF-funded projects are implemented through a "delegated cooperation" whereby public development cooperation agencies belonging to individual Member States bid on and act as contractors for large-scale EU-funded projects, just as NGOs generally do, following EU's rationales and guidelines instead of those of Member States. Hence, this kind of Europeanisation has emerged as an unintended consequence of the EU principle of transparency and fair competition rather than a planned outcome of overarching strategies concerning migration controls. Overall, practices on the ground do not appear to simply be the outputs of clear-cut EU or national strategies: instead, practices and strategies interact in complex cycles of recursive interactions where established routines, a variety of rules, and the actors' diverse capacity may undermine or reinforce any intended Europeanisation.

All in all, Europeanisation appears to be more similar to a hybridisation of strategies, logics of action and practices, rather than to a top-down adoption of common regulations issued by EU-level entities or to horizontal convergence among clearcut national approaches. The Europeanisation of migration control analysed from a regime-analytical perspective appears more complex than the rational efforts of Member States to upload and download their policy preferences to and from EU policies, and more multifaceted than a mere convergence between different national models driven by the necessity to answer similar challenges or to implement common EU rules; it cannot be reduced to the response to deterministic impulses. Rather, the adoption of rules from above goes hand in hand with the institutionalisation of practices from below. Actually, Europeanisation appears rather fuzzy and partially incoherent, a multiactor, multilayered process where states are not commanders in chief, a sidewinding process which blurs the boundaries that are supposed to divide migration regimes in Europe.

#### **17.4 Framing Difference: Competing Interests and Internal Constraints**

The European migration system is clearly a complex, ambiguous reality, where convergence dynamics must come to terms with persisting variance. The way nationstates deal with migration is highly dependent on the different geopolitical, economic and institutional constellations they are embedded in and the structural limitations they entail. Such embeddedness is key to understanding the functioning of migration control and to framing divergence beyond the traditional opposition of "weak" versus "strong" control capacity. On the one hand, nation-states are not isolated units, but interconnected political organisations that often need to negotiate their migration control goals with a variety of interests and social actors, depending on the migration systems in which they are involved. On the other hand, nationstates observe and act upon international migration differently, depending on "the design of these states, their way of reproducing sovereignty, and the related conceptualisation of loyalty and power relations" (Bommes & Thränhardt, 2010: 210). Hence, nation-states' migration-control goals are shaped by different nationbuilding histories, interests and structural designs (e.g. the dimension of the informal economy or different types of institutional traditions) that might at least partially offset the power of convergence processes based on changes of infows and Europeanisation. Against this backdrop, another contribution of this book is to show that the mechanisms of migration control can be only understood by taking into account the connections of nation-states with other types of systems (e.g., international relations or the economic system) and the power of internal constraints.

#### *17.4.1 The Role of Competing State Interests*

Nation-states as crucial engines of complex migration regimes are not impermeable units, but political organisations with different types of (often contradictory) interests that must deal with different social organisations. Such interests vary not only from state to state, but also within individual states and are key when shaping the strategies and logics of various migration controls and therefore contributing to the confguration and functioning of migration regimes.

As various chapters of this book have shown, migration control measures have been clearly affected by geopolitical interests in several ways. Infantino's chapter pointed out that countries such as France were initially reluctant to introduce restrictive visa regulations fearing a negative impact on their bilateral relations. In fact, it was not irregular migration but a series of terrorist attacks carried out in the 1980s

that provided a "policy window" to introduce visa restrictions. Nonetheless, current trends show that geopolitical interests still play a very important role in visa-policy implementation; pre-existing bilateral relationships clearly had an infuence on Schengen visa application trends. As the data presented in the chapter by Landini and Sciortino showed, differences in visa rejection rates do not seem to refect different attitudes to migration control, but rather different types of geopolitical and historical considerations. According to the authors, this turns visa policy into a particularly suitable example to show that states are not independent but interdependent units whose control policies do not depend on more or less effective implementation, but rather on the role played by each nation-state in the control system in which it is involved. This variety of migration systems also has implications for the implementation of enforced returns. In fact, as Landini and Sciortino argued, differences in return rates have more to do with the composition of the undocumented immigrant population in a given country than with the institutional capacity to return migrants – nationals from some countries are easier to return than others, depending on the existence or not of bilateral agreements between immigrants' countries of destination and their countries of origin.

The close connection between the struggle against irregular migration and other (mainly geopolitical) interests is particularly evident in the case of the complex negotiations on migration cooperation among the EU, Spain and Morocco (Carrera et al., 2016). Successive Spanish governments have had to combine their negotiations on border controls at the EU level with issues such as trade agreements or questions regarding Western Sahara, which are both very important for Morocco's geopolitical agenda in Northern Africa, as the text by Gabrielli noted. And as Gazzotti, Jiménez Álvarez and Espiñeira showed, European efforts to protect its external borders have thus become deeply intertwined with Morocco's attempts to increase its geopolitical infuence through cooperation with the EU on migration issues; this has turned Morocco's "migration diplomacy" into a crucial element in the construction of the European border regime.

Of course, geopolitical interests did not only infuence the negotiation of migration issues with third countries but also shaped intra-European migration policies as well. As D'Angelo pointedly argued in his chapter, welfare chauvinism directed against European citizens and the related restrictions and oversight of intra-EU migrants became useful instruments for the UK to "repudiate" its role as a major node of the EU regional system. Geopolitical interests, such as the need to compensate certain countries for asylum restrictions or to foster cooperation on migration issues with new transit countries, as it occurred for the Balkans or Georgia in the case of Germany, have also affected the implementation of labour migration recruitment schemes by opening the way to "ethnic preference" at the expense of personal qualifcations (see Kolb & Schneider).

Finally, it should be noted that geopolitical interests have been frequently complemented by economic interests in the design and implementation of migration control. As Malheiros and Peixoto emphasised in their chapter, regularisations probably represent the most evident example of how state control imperatives confict with pressure from employers to open (even though *a posteriori*) admission channels to employ migrant workers, thus guaranteeing the functioning of labourintensive economic sectors. Infantino's chapter furthermore reminds us that Italy's pre-Schengen visa policy was shaped not only by foreign policy interests, but also tourism. Indeed, the relevance of tourism and economic interests for visa policy can still be observed in the designation of visa-exempt countries and visa rejection rates, which are lower for countries that generate large fows of tourism and business, such as China and Russia before the pandemic and Ukraine-war (Finotelli & Sciortino, 2013). Likewise, tourism and other economic interests probably had a role in limiting control ambitions on intra-European fows. For instance, the relevance of tourism for the Spanish economy helps to explain the government's inertia in discussing the compensation system for healthcare services provided to Northern EU and UK citizens on the Spanish territory. As Finotelli suggested, nobody seems to have a real interest in provoking a debate that could negatively affect Spain's attractiveness to tourists and "lifestyle migrants" from other EU countries. In this case, accumulation logics embedded in the economic and tourism sector seem to be more relevant than the "protection" of the public healthcare sector within the EU citizenship regime.

#### *17.4.2 Different Types of Internal Constraints*

The previous section shows that nation-states are interdependent units where migration control policies are deeply interconnected with other types of state interests and policy felds. Yet migration control logics and practices are not only shaped by different types of state interests, but also by different types of state designs. In fact, the functioning of migration control cannot be adequately understood without taking into account the fact that nation-states are political organisations with different characteristics that often act as structural constraints on similar imperatives. It is against this backdrop that the chapters presented in this book provide some insights on how national constraints such as nation-states' geographical position, the political-institutional capacity and their labour-market structures can shape different migration control strategies.

Geographical position and the evolution of migration routes were obviously key factors behind Italy's and Greece's exposure to refugee fows during the migration crisis in 2015 (see Dimitriadi; Ponzo). On the other side of the Mediterranean, the Spanish border regime has obviously been infuenced by its land and maritime border with Morocco, while geography has kept Portugal fairly isolated from all major European migration crises (see Gabrielli; Gazzotti, Jiménez Álvarez & Espiñeira).

Besides geographical position, other types of structural determinants inevitably have an infuence on the regulation of wanted and unwanted migration fows. For instance, the design and implementation of labour migration policies clearly depend on the type of labour demand. Different types of labour demand inevitably determine different types of labour recruitment. This is, for instance, the case of Southern European countries such as Italy, where recruitment schemes for highly skilled

foreign workers have been rather unsuccessful due to the reduced demand for highskilled labour in the country (see Devitt; Doomernik, Garcés-Mascareñas & Güell).

Labour recruitment policies in Southern Europe have also been affected by a weak political-institutional capacity that has prevented the effcient match between labour demand and supply, as the case of Italy demonstrates (see Devitt); this was not the case in Spain, due to its more coordinated bureaucracy and its lower level of political instability (Finotelli & Echeverría, 2017). The same seems to apply to the feld of asylum, where states' political-institutional capacity has clearly infuenced the way in which they have dealt with unwanted migrants such as refugees and other irregular migrants. A weak political-institutional capacity has limited improvements to the Italian and Greek asylum procedures (see Ponzo; Dimitriadi). As already noted, Germany experienced cumbersome asylum procedures and diffculties in managing asylum fows, similar to its Southern counterparts. However, Germany's troubles seemed to be more related, at least in later phases of the asylum crisis, to a political desire to reduce Germany's attractiveness to new asylum seekers than with weak institutional capacity (see Thränhardt). In light of this, the new government coalition led by Olaf Scholz has called for "a paradigm shift". The approved coalition agreement states that one of the goals of the German government in the feld of migration and asylum will be to turn over a new leaf with "fair, fast and legally secure" asylum procedures (Coalition Agreement 2021, 139–140.); this should take place by taking pressure off the Federal Offce for Migration and Refugees (BAMF) and introducing a new countrywide, independent asylum counselling system (*ibid*.).

The interrelation between institutional capacity and social expectations regarding the state's control capacity or the legitimacy of policy options such as amnesties represents a further important aspect to understand the way regimes observe and act upon irregular migrants. German governments have been traditionally reluctant to immigrant regularisation, since this would undermine the traditionally high expectations from society about the state's control capacity. That is why *ex-post* regularisations in Germany are primarily embedded in the functioning of the asylum system and do not have an economic but rather a "humanitarian" background in the form of "exceptional leave to remain (*Duldungen*)" and "old-case regulations" (see Finotelli). In contrast, regularisations are less controversial measures in Italy and other Southern European countries since their public acceptance is deeply embedded in the traditionally modest expectations concerning the state's institutional capacity (see Finotelli; Malheiros and Peixoto). Against this backdrop, regularisations and amnesties have been carried out not only to reduce the number of irregular migrants, but also to legalise formerly illegal buildings, rectify tax evasion or even address the illegal possession of archaeological artifacts and exotic animals. Similarly, Echeverría showed that the different approaches to internal control in Spain and the Netherlands also depended on historically different modes of state penetration in society. In the Netherlands, this has resulted in systematically higher levels of ambition concerning the state capacity to control irregular migration than in Spain. At the same time, he showed that political resistance to internal identity checks, higher in the Netherlands and Germany than in Spain, may be related to different perceptions of institutional legitimisation when interfering with personal freedoms.

Nevertheless, the fact that Southern and Northern European countries deal with undocumented migration differently does not mean that they actually achieve different results. Policies designed to produce and manage knowledge on irregular migration are examples of this. For instance, regularisations of labour migrants in Italy and *ex post* regulations of rejected asylum seekers in Germany have allowed these states to recover control of irregular migration, showing that different ways of handling state ignorance of unauthorised residence follow similar regime logics (see Finotelli). The comparison of the German and Italian cases suggests that functional equivalence rather than divergence may represent a more useful analytical tool for understanding the way nation-states with different histories and internal designs handle the presence of undocumented migrants.

#### **17.5 Conclusion and Outlook**

The analyses of migration control logics and strategies in this book make it possible to explain migration control outcomes not as the pure, direct products of state control policies, but rather as the intended and unintended consequence of strategies pursued by a wide variety of public and private actors operating on various geographical levels. To this end, the concept of "migration regimes" has proved to be a useful heuristic tool for researching immigration detached from normative assumptions and for understanding how immigration controls actually work. By using the concept of migration regimes, the chapters unravel policy practices and organisational strategies of nation-states, demonstrating that migration controls are not simply a reaction to the social phenomenon of immigration, but also the outcome of a negotiation process where the governments of individual states are only one of the many possible actors involved, ranging from street-level bureaucrats and local governments to Europe-wide policymakers, and from local NGOs and private actors to major international organisations. Against such a backdrop, nation-states do not appear as independent actors, but rather as highly interconnected political organisations, deeply embedded in a dynamic interplay of changing external contexts, different geopolitical and economic interests, and internal constraints.

The use of the regime concept in this book not only empirically challenges the idea of clear-cut national models defned by more or less policy effcacy. It also contributes to questioning the North-South divide as an analytical lens, and helps pave the way for a new conceptualisation of similarities and differences in European migration controls. In fact, the chapters in this book bring to light a multifaceted reality, where convergence dynamics overlap with persisting differences. If, on the one hand, they shed light on the blurring of boundaries between national migration regimes as the result of Europeanisation processes and external dynamics that have made infows more homogenous, on the other they show that migration controls and their outcomes remain dependent on internal structural designs and the role played by nation-states in the geopolitical or economic systems in which they are involved. This allows scholars to identify the many patterns of convergence and variance in the European control systems, despite common control imperatives.

Overall, the analyses presented in this volume contribute to overcoming clear-cut and often Manichean representations of the functioning of migration control in Europe. Yet there are still open questions with regard to the dynamics of convergence and divergence observed in the previous pages. One of these points is surely the impact of visa policy on other policy felds. As previously mentioned, the new Visa Code creates a new connection between visa policy and the enforcement of returns, whereby third countries that do not collaborate in readmitting their own citizens may become the object of visa restrictions. It is certainly premature to assess the effectiveness of this "conditionality clause" with regard to readmissions. Nevertheless, linking visa policy (with its high degree of harmonisation) to collaboration on enforced returns (with a very low degree of harmonisation) may represent an important step towards policy convergence.

Another question concerns the "Southernisation" trend observed in some policy felds. *Refoulements* in the East and Central Mediterranean, together with the limitations imposed on NGO interventions, have become quite a frequent (and tacitly accepted) practice – which has been adopted by Eastern countries, e.g., in the *refoulement* of refugees coming from Belarus in 2021 and the establishment of a no-go area for NGOs along the Polish border. Moreover, the European Court of Human Rights in Strasbourg found in *N.D. and N.T. v. Spain* (13, February 2020) that "hot returns" at the Spanish-Moroccan are not equivalent to a violation of collective expulsions under Protocol 4 Article 4 of the European Convention of Human Rights (Carrera, 2020). It therefore remains to be seen how certain types of "dirty" border practices implemented in Southern Europe will become mainstream in the European Union.

Third, it goes without saying that a proper understanding of the functioning of the European migration control system and the related convergence and divergence dynamics will inevitably require looking at Eastern and South-eastern countries, especially taking into account the consequences of the recent war between Russia and Ukraine. This has already started to impact the European asylum regime with the activation, for the frst time ever, of the Temporary Protection Directive for Ukraine refugees. The convergence triggered by this momentous decision has been reinforced by what we called "Europeanisation by practice": Member States have been facilitating Ukrainians' mobility and settlement by adopting rather swift procedures, while almost everywhere NGOs and civil society have massively complemented the public intervention. The way of handling the Ukrainian crisis was initially regarded as a sort of turning point for the European asylum regime since the rules and practices set up to deal with Ukraine refugees had not been applied to the people crossing the East and Central Mediterranean and the Belarus border.1 Yet, the smooth reception of Ukrainian refugee fows might also represent a further example

<sup>1</sup>On the question of the Ukrainian war as a turning point in the European asylum system see also the contributions by Thränhardt (2022); Thym (2022); Benton and Selee (2022).

of "source-country particularism" following a trend that has already been highlighted in this volume with regard to labour migration. This development emphasised how control policies are affected by intervening factors such as policy frames and geopolitical interests. Moreover, in a period in which the European Union reaches one of the lowest unemployment rates ever and the labour demand is skyrocketing, the skilled and socially accepted labour force arriving from the Eastern border is more than welcomed by the economic actors. Indeed, the Ukraine crisis further confrms the heuristic usefulness of the concept of "migration regimes" to read the complex and dynamic scenario of migration controls.

How thoroughly is the EU's migration regime going to be shaped by Southern, Eastern and South-eastern European priorities, is a fascinating question that will deserve a great deal of research attention in the future.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.