The Tools of External Migration Policy in the EU Member States: The Case of Italy

This article focuses on the external dimension of migration policy (EXMIPO) in the EU member states (MSs), through a policy tool approach. It offers an analytical framework for conceptualizing MS EXMIPO, by (1) unpacking its tools and (2) understanding their relationship with EU-level migration governance, as based on competition, convergence and complementarity. Empirically, it examines the case of Italy, building on an original dataset spanning over 30 years and 125 instruments. Through the Italian case, this article demonstrates that MSs have developed an intense external action at the bilateral level in the ﬁ eld of migration, which has expanded quantitatively and qualitatively, alongside (and despite) EU initiatives.


Introduction
In recent decades, the instruments of migration policies have garnered significant attention in policy and academic debates.Striving to tackle migration effectively, the European Union (EU) and its member states (MSs) have diversified their migration policy tools, both within their borders and beyond.As migration has risen to the top of both domestic agendas and foreign policy priorities, the 'external dimension' of migration policy in Europe has gained salience (Geddes et al., 2020;Niemann and Zaun, 2023).
The 'external dimension of migration policies' (EXMIPO) involves the engagement of countries of origin and transit in managing migration.Extensive literature has explored the many instruments (such as dialogues, arrangements, statements and deals) that constitute these co-operative efforts, their interplay and their evolution over time (Carrera, 2019;Carrera et al., 2015;Czaika et al., 2023;Longo and Fontana, 2022;Reslow, 2017Reslow, , 2019;;Trauner and Wolff, 2014;Zardo, 2022).
However, research on this aspect within the European context has predominantly centred on the EU level, often neglecting analyses of individual MSs.The emphasis on the supranational level can be partly attributed to the initial expectation that the EU would lead in shaping external migration policies, serving as a mediator for agreements on behalf of all MSs and wielding greater negotiating leverage with third countries compared to individual MSs (Cassarino et al., 2023, p. 48;Weinar, 2011).Whilst scholars have observed that these initial expectations were not met and that MSs have developed a 'resilient bilateralism' (Cassarino, 2011) alongside EU initiatives, research at the MS level has mainly focused on return and readmission agreements.The external dimension of MS migration policies, including the diversity of tools used in bilateral relations, their configuration and the way they affect the EU external dimension, has received limited attention.
In this article, we contend that MS EXMIPO needs to be unpacked further.Not only is EU foreign policy still largely in the hands of the MSs, but migration is also a matter of shared competence between the Union and the MSs.Our research is thus driven by the following questions: how can MS EXMIPO be conceptualized?Specifically, what tools characterize the external dimension of MS migration policies, and how have they evolved?How does MS EXMIPO relate to EU initiatives?
In examining the above, we propose a new analytical framework for unpacking the tools of the EXMIPO at the MS level.Departing from the literature on EU 'externalization instruments' (Longo and Fontana, 2022), we reorient the focus from the EU level to the MSs and back.We analyse MS EXMIPO in terms of the tools used, temporal evolution and relations with the EU, conceptualizing MS EXMIPO as being related to EU EXMIPO through competition, convergence and complementarity.We contend that the examination of MS external migration policies is enriched by a deeper understanding of the empirical referents of the policy itselfnamely, the tools that MSs adopt to manage migration in co-operation with third countries.A policy tool approach to the study of MS EXMIPO not only reveals how national policy-makers translate intentions into practices but also allows exploration of the spatial and temporal evolution of external migration policies and how these tools interact with EU instruments.
We ground the analysis in original empirical material, focusing on the case of Italy as a key instance of an MS that represents a crucial hub of migratory flows to the EU.We rely on an original dataset 1 that is the first comprehensive effort to systematically identify and classify the instruments of Italy's external migration policy.The dataset contains Italy's agreements with 17 countries in the broader Mediterranean, between 1990 and 2022, for a total of 125 entries. 2  Through the case of Italy, this article demonstrates that MSs have developed a rather intense set of external activities at the bilateral level, which has expanded quantitatively and qualitatively over time and space, alongside (and despite) EU initiatives.It shows that MS EXMIPO is not limited to bilateral readmission agreements (RAs).Instead, it is far more sophisticated, encompassing a wide range of policy tools such as quota schemes, labour agreements, military missions and political dialogue, which simultaneously complement and hamper EU action.
In this article, we explore the theoretical concept of 'external dimension' of migration policies, going beyond the EU level and reorienting the analysis to the MS.We then outline our analytical framework and typology of MS EXMIPO instruments and how these relate to EU initiatives.Next, we illustrate the methodology.Finally, we apply our typology to the Italian case and discuss the empirical results, examining the evolution of Italy's EXMIPO and its interaction with the EU's external migration policy. 1 The dataset was developed in the context of the Project DEPMI -Dimensione Esterna Politica di Migrazione Italiana, conducted by The Siracusa International Institute for Criminal Justice and Human Rights and implemented with the support of the Italian Ministry of Foreign Affairs.The project takes inspiration from, and expands, previous research on the mapping of the external dimension of migration policies in the case of the EU, in the framework of the H2020 project 'PROTECT -The Right to International Protection'. 2 The dataset is available in the form of an interactive map at https://bit.ly/3vsO0iR.

I. The 'External Dimension' of Migration Policies: Reorienting the Analysis From the EU to the MS (and Back)
The concept of the 'external dimension' of migration policies originally developed in EU studies and refers to the incorporation of migration in the scope of the Union's external relations and to the engagement of countries of origin and transit in managing migration to the EU, through various tools and instruments (Boswell, 2003;Carrera et al., 2015).In this sense, the external dimension emerges when policies traditionally belonging to the domain of internal affairs, such as migration and asylum, go 'abroad' and get an external resonance (Carrera et al., 2015;Geddes, 2009).
In the extensive literature on this topic, particularly at the EU level, four main analytical bodies can be identified.The first explains EU externalization practices as the 'natural continuation' (Boswell, 2003) and the 'external face' (Lavenex, 2007) of the process of Europeanization, in terms of the transfer of traditional migration policies from the EU to other settings outside Europe and their impact on actors other than the MSs (Lavenex and Uçarer, 2004).Just as Europeanization describes the processes of change inside the EU, including the reconfiguration of territorial boundaries in the process of European integration (Lavenex, 2007, p. 246), externalization is framed as the 'extraterritorialization of territorial boundaries' (Rijpma and Cremona, 2007) or 'deterritorialization of migration control' (Van Munster and Sterkx, 2006).
The second body of literature focuses on the nexus between EU foreign affairs and migration management (Carrera, 2019;Eisele, 2016;Martenczuk, 2014) and how issues traditionally belonging to the domestic agenda are translated into foreign policy initiatives (Dimitriadi, 2016).Externalization practices are explained as shifting 'outwards' towards foreign policy co-operation (Lavenex, 2006), in the attempt to secure third countries' co-operation and make migration policies more effective, whilst balancing the lack of EU internal frontiers with greater action at the external border (Boswell, 2003;Cardwell, 2013).
The third explores externalization as a case of 'securitization' of migration in terms of both discourses and practices (Huysmans, 2000;Léonard and Kaunert, 2022;Longo and Fontana, 2022).As migration is increasingly framed as a security issue in EU agendas, co-operation with third countries emerges as an extraordinary means to manage the complex entangling of external and internal aspects of security (Bigo, 2001;Boswell, 2003), through forms of 'remote control' (FitzGerald, 2020;Panebianco, 2022).
Finally, the fourth body focuses on the diverse policy tools used to incorporate surrounding states and regions into the objectives and instruments of EU policy (Carrera et al., 2015;Geddes, 2005;Reslow, 2017).Migration dialogues, visa facilitation, mobility partnerships, RAs, common agendas, EASO and Frontex arrangements are just some of the many instruments that constitute the EU's 'toolbox of externalization' (Longo and Fontana, 2022).These tools can be legal (e.g., formal agreements), political (e.g., declarations and memoranda) and operational (e.g., technical co-operation agreements) (Longo and Fontana, 2022).In terms of content, the literature identifies three main clusters of tools, aimed at (1) preventing and managing irregular migration, (2) addressing the root causes of migration and (3) co-operating on legal pathways (Boswell, 2003;Carrera et al., 2018;Eisele, 2016;Chou, 2009).
The tools of external migration policy in the EU member states: The case of Italy Most of these tools are considered external as they are decided and implemented in co-operation with partner countries (Carrera et al., 2015).However, there are also other instruments that, whilst being 'internal' because they are autonomously decided by the EU without the involvement of third countries, still have an 'external' impact (Carrera et al., 2015), with implications in terms of mobility, asylum or negotiation of RAs.This is exemplified by concepts like 'safe country of origin' or the inclusion of a country in the Schengen visa exemption lists.
Whereas the concept of external dimension of migration policies is extensively explored in the case of the EU, less attention was paid to the analysis of the same concept in case of MS migration policies.This is surprising for three reasons.First, bilateral co-operation between the MSs and third countries of origin and transit was already well consolidated before the external dimension of migration policies was introduced by the European Council of Tampere in 1999, and a rapid proliferation of bilateral agreements concluded outside the scope of the EU had been occurring since the early 1990s3 (Cassarino, 2010).Although most of these agreements exclusively focused on return and readmission, they represented the initial attempts to externalize migration policies, linking migration and foreign policy goals to compensate for the failures of unilateral state action (Castles, 2004;Teitelbaum, 1984).Given the shortcomings of internal migration policies (Rosina, 2022), MSs began exploring alternatives, by shifting migration policy-making to the foreign policy arena and addressing migration control dilemmas through co-operation with third countries (Boswell, 2003;Niemann and Zaun, 2023;Weinar, 2011).
Second, as migration remains a shared competence, MSs play a crucial role in shaping EU externalization strategies (Enriquez et al., 2018) through variable geometries of bilateral agreements.Notable examples include the Italy-Libya Deal of 2017 and the Spain-Morocco agreement of 2012.Whilst these agreements are bilateral tools concluded outside the EU framework, they have come to represent a much-debated cornerstone of EU external migration policies in the Central and Western Mediterranean.Similarly, as mobility continues to be an underdeveloped dimension in EU external migration policies, MS bilateral labour agreements and quotas take on greater significance.Moreover, whilst EU RAs with certain countries hold legal precedence over RAs concluded by individual MSs with the same countries, numerically, EU agreements are still fewer than those concluded by MSs, due to third countries' preference for bilateral co-operation outside the EU's purview (European Court of Auditors, [ECA], 2021).
Third, due to the intertwining of policies, actors and levels of action within EU EXMIPO (Garcia Andrade, 2019), analyses of which specific tools are used by MSs, and whether and how they are linked to the EU's initiatives, provide crucial insights into overall EU migration governance.Whilst a focus on the tools of the MS inherently prioritizes the state level, potentially underemphasizing the roles played by non-state or international actors, it allows going beyond political rhetoric and engaging with actual policy instruments.This can shed light on a multitude of aspects including diverging or converging policy choices, cases of policy learning and emulation and underlying power dynamics.
Despite the relevance of expanding the investigation of the tools of the external dimension of migration policies to the MS, curiously, only a few studies explore such aspects.
Of the few existing works, Panizzon (2012) delves into the division of power between the EU and the MSs in the case of RAs.Cassarino (2010) relatedly explores the determinants of bilateral readmission co-operation in terms of geographical proximity, migration salience and incentives, arguing that neighbouring countries may have a higher propensity to co-operate on readmission, especially when migration is not a politicized issue in the sending country and the destination state is providing adequate incentives.However, these studies are focused exclusively on RAs.The 'policy toolbox dimension' (Longo and Fontana, 2022) in terms of the variety of tools being adopted at the bilateral level and how these are arrayed in a mix remains unexplored.
This article aims to address this gap by redirecting the analysis of the external dimension of migration policies from the EU to the MS and back.It adopts a policy tool approach to investigate the various instruments of the MSs' externalization toolbox and how they fit into the EU's external dimension.The following section offers a novel analytical framework and a typology, representing the first comprehensive effort to unpack and take stock of the external migration policies of the MS.

The Policy Tools of MS EXMIPO: In Search of a Typology and Conceptual Framework
In public policy studies, policy tools are the key means through which governments intervene to achieve policy goals and implement policy options (Capano and Howlett, 2020).The analysis of policy 'toolboxes' has received increasing analytical and empirical attention due to its focus on the role of policy instruments and how they relate to each other when combined, and an instrument-based approach is now recognized as a perspective able to enlighten the nature of policy dynamics (Capano and Howlett, 2020;Lascoumes and Le Gales, 2007).
This work starts from the assumption that the analysis of policy tools as the empirical referents of external migration policies can enhance our understanding of the latter.Specifically, 'deconstruction through instruments' (Lascoumes and Le Gales, 2007, p. 4) enables the identification of the tools that underpin MS EXMIPO and facilitates assessments of its transformations, in both scope and scale.Additionally, employing an instrument-based approach 'allows us to address dimensions of public policy that would otherwise not be very visible' (Lascoumes and Le Gales, 2007, p. 4), contributing to a broader comprehension of a country's overall migration diplomacy (Adamson and Tsourapas, 2018), shedding light on the (often-concealed) means by which such diplomacy is conducted.
Therefore, readapting the fourth strand of the literature on the external dimension of EU migration policies to the study of MSs, we aim to understand the various instruments that shape MSs' co-operation with third states in managing migratory flows.
Regarding the EU, scholars have provided insightful categorizations of the diverse tools that constitute EU EXMIPO (Table 1).Trauner and Wolff (2014) propose a typology distinguishing nine instruments according to their naturewhether incentive-based, operational or rooted in international law.Carrera et al. (2015) identify legal instruments (e. g., RAs), soft policy tools (e.g., mobility partnerships), political dialogue on migration, multilateral policy processes (e.g., Budapest process), accompanying policy tools (e.g., Frontex working arrangements) and specific programmes.Czaika et al. (2023) investigate EU EXMIPO by considering the instruments' geographic orientationwhether they The tools of external migration policy in the EU member states: The case of Italy pertain to internal or external affairsand their implicit or explicit targeting of migration issues.Lastly, Longo and Fontana (2022) propose a 'toolbox' with 13 specific elements, spanning from binding and formal agreements to softer or even informal tools.
We argue that the notion of external dimension is not exclusive to the EU but does exist at the national level too.In particular, we understand the external dimension of migration policies in the case of the MSs as including a rich toolbox of instruments employed to co-operate with third countries for the attainment of internal policy goals.
To map and classify the instruments of MS EXMIPO, we consider instruments and tools as synonyms for the various means that facilitate the attainment of the objectives of MS external migration policies.We adhere to Lascoumes and Le Gales' (2007, p. 4) understanding of policy instruments as a device that is both technical and social, that organizes specific social relations between the state and those it is addressed to.
Expanding on Longo and Fontana (2022), we identify nine elements that constitute the 'policy toolbox' of MS external migration policies (Table 2). 4 Whilst this list may not be 4 Starting with the categories of tools identified in the EU literature, we (a) selected the categories that have comparable tools or counterparts at the bilateral level (e.g., EU RAs → MS RAs); (b) adjusted categories that, whilst being characteristic of the EU, could point to similar tools at the MS level (e.g., EU visa facilitation agreements suggesting exploration of bilateral tools promoting legal mobility, like quotas and labour agreements; or EU protection programmes suggesting exploration of bilateral tools encouraging legal pathways of protection, such as humanitarian corridors); (c) excluded tools exclusive to the European context, such as Frontex/EASO working arrangements and mobility partnerships.Lastly, we have intentionally omitted the discussion of aids and development programmes.This deliberate exclusion is based on the belief that these initiatives constitute a highly specialized dimension that warrants a distinct and comprehensive analysis, complete with its own dedicated toolbox.exhaustive, it provides a comprehensive effort to conceptualize EXMIPO tools, for an improved empirical understanding of MS external actions concerning migration.
Quotas (2) and labour migration agreements (3) concern regular mobility.Quotas are quantitative restrictions set by MSs on the number of migrants admitted each year.These measures are primarily aimed at managing labour migration from third countries, with quota limits being determined on the needs of the domestic labour market (e.g., in Italy, Portugal and Slovenia) or other criteria (e.g., in Austria or Estonia) (European Migration Network [EMN], 2014).Quotas can be included in bilateral agreements or implemented unilaterally by the MSs.Even when set unilaterally, they still play a role in MSs' externalization toolbox, potentially acting as an incentive for other agreements.Labour migration agreements establish co-operation on labour market demands and access, seasonal or circular mobility and partnerships for graduates and young professionals (EMN, 2021).We incorporate here visa agreements too.Examples include the France-Mauritius Agreement on the Stay and Circular Migration of Professionals (2008) or Italy-Egypt Agreement on Bilateral Migration Flows for Employment Reasons (2005).
Resettlement schemes and private sponsorship partnerships (such as humanitarian corridors) (4) facilitate the safe and legal transfer of vulnerable refugees to Europe (e.g., Irish-Afghan admission programmes or Swedish resettlement programmes).Whilst they may not entail bilateral agreements and instead result from multilevel co-operation amongst national and international, institutional and non-institutional actors, these programmes remain key instruments of co-operation with countries of origin and transit (European Union Agency for Asylum, [EUAA], 2023).
Technical and operational agreements (TOAs) (5), migration-specific agreements (6) and migration clauses in broader bilateral agreements (7) are cross-cutting tools.TOAs encompass protocols, memoranda or exchange of notes, focusing on operational and technical co-operation such as capacity building, training and provision of equipment, in areas like border management or police co-operation.The Spanish-Mauritania Memorándum de Cooperación Policial (2009) or the Italy-Libya Technical Protocol of Cooperation ( 2007) are a case in point.Migration-specific agreements (6) refer to agreements The tools of external migration policy in the EU member states: The case of Italy that are specifically dedicated to migration matters but that are broader than technical-operational, readmission and labour mobility agreements and that integrate several aspects.Examples include the Spain-Mali Framework of Cooperation on Migration Issue (2008) or France-Burkina Faso Agreement on Concerted Management of Migratory Flows and Joint-Development (2009).Lastly, under 'migration clauses in broader bilateral agreements' (7), we include bilateral agreements that, whilst addressing various objectives and issues, also incorporate migration co-operation.Examples include treaties of friendship and good neighbourhood, strategic partnerships and agreements of cultural and scientific co-operation.
Political dialogue (8) involves diplomatic exchanges and high-level meetings that include migratory issues in their agenda.Lastly, military missions (9) refer to MS' missions integrating migration components into their mandates and reflecting the increasing deployment of the military to address migration challenges (Biscop and Rehrl, 2016) (e. g., Italy's military mission in Niger).
Whilst different, these instruments share potential similarities in their content and nature.Content refers to the specific areas the instrument is focused on.These can be irregular migration (border control, fight against human smuggling and readmission and return) or regular migration (mobility and legal pathways).In terms of nature, they can be formal or informal, depending on factors such as outputs (e.g., a declaration/memorandum, rather than an international agreement), processes (e.g., political dialogue that bypasses parliamentary scrutiny or treaty-making norms), actors involved (e.g., agencies rather than diplomats) or non-binding forms (see Cardwell and Dickson, 2023;Ott, 2020;Pauwelyn, 2012).
The identification of these nine categories of instruments, differentiated by their content and nature (Table 3), offers valuable insights to take stock of MS EXMIPO across various dimensions, including time, geography and policy priorities.
Moreover, in dialogue with the literature that characterizes EU migration governance as a multi-layered system (Cardwell, 2011;Kunz et al., 2011;Scholten, 2013;Scholten and Penninx, 2016;Scholten and van Ostaijen, 2018;Spencer, 2018), the identification of these nine instruments enables an understanding of the interplay between MS and EU external migration policies.It allows an examination of how tools interact, conflict or (dis)engage, leading to either co-ordination or fragmentation (Fakhoury, 2019;Panizzon and van Riemsdijk, 2019).MS EXMIPO does not occur in isolation, and MSs' tools have varying impacts on the overarching EU external migration policy (summarized in Table 4).They Iole Fontana and Matilde Rosina can either complement or hinder EU EXMIPO, whilst maximizing their national autonomy through 'the use' or 'non-use' of Europe (Slominski and Trauner, 2018).In this regard, our theoretical expectation is that MS EXMIPO instruments interact with EU-level tools in three primary ways: competition, convergence and complementarity.
Competition emerges when antagonistic dynamics arise, with MS withdrawing from the supranational level to safeguard their national prerogatives (Hanke et al., 2019).This leads to dysfunctional interactions (Niemann and Speyer, 2017) and inefficient overlaps between EU and MS tools, impeding the development of EU EXMIPO.Competition in external policy-making indicates either forms of disjointed governance, where policy instruments operate in a contradictory manner (Curry, 2018), or de-coupled governance (Scholten, 2013;Scholten and Penninx, 2016), characterized by instruments functioning in a separate and non-reinforcing way.In both scenarios, they transmit mixed signals to policy targets, such as transit or origin countries, fostering venue shopping and cherry-picking by third countries (Hanke et al., 2019).
We expect competition as disjointed governance to be particularly evident in the realm of RAs (1), where bilateral agreements between MSs and third countries might hinder EU negotiations, diminishing the latter's bargaining power.MSs retain the right to negotiate an RA until the EU decides to do so.Indeed, EU's RAs (EURAs) take precedence over any previous bilateral RA made by the MSs: when the European Commission (EC) receives a negotiating mandate for EURAs, MS are expected to discontinue any ongoing or future negotiations (ECA, 2021).However, EU history is replete with cases of bilateral RAs being pursued despite the EU being already in the negotiation process (Carrera, 2019;Garcia Andrade, 2019).This practice can hinder EU negotiation efforts, potentially diminishing the EU's leverage, as third countries may not see a clear advantage in The tools of external migration policy in the EU member states: The case of Italy pursuing a EURA over co-operating with individual MSs.Moreover, when the EU enters into informal return arrangements, their non-binding nature does not hinder MS-level initiatives, granting MSs enough flexibility that could potentially sideline EU efforts (Slominski and Trauner, 2018).
On the flipside, we anticipate competition as de-coupled governance to be found in quotas (2) and labour agreements (3).As MSs retain the power to determine the number of labour migrants, the EU's capacity to offer legal migration opportunities largely depends on MS actions.Quotas and labour migration agreements can act as conditional incentives to enhance not only bilateral co-operation but also EU collaboration with third countries.However, limited incentives for legal migration opportunities at the national level, or their use as rewards to advance individual MS' interests and exclusively promote bilateral co-operation, often undermine the effectiveness of EU EXMIPO and its ability to leverage these tools (European Parliament, 2015), leading to forms of venue shopping by third countries.
Convergence occurs when MSs and the EU share common goals and priorities, and their respective tools are mutually reinforcing.This reflects dynamics of multilevel governance with tools being synchronized without a clear hierarchy (Scholten, 2013).In these cases, national and EU tools align, resulting in a co-operative approach to addressing migration challenges.We expect this to be the case for resettlement programmes (4), which complement the EU EXMIPO's approach of establishing legal pathways through humanitarian admission (EC, 2017), or for military missions (9) that contribute to EU migration management objectives through activities like border security, capacity building and anti-smuggling efforts in third countries (EC, 2015).Finally, the enhancement of bilateral political dialogue (8) could also synergize with the objectives of the EU EXMIPO.
Lastly, complementarity refers to instances where MSs, either explicitly or implicitly, act on behalf of the EU, opening new avenues and facilitating the development of EU EXMIPO.This is exemplified by situations in which MSs with well-established bilateral relations could act as facilitators or intermediaries in expanding dialogue with certain countries at the EU level.Complementarity can also arise when MSs compensate for the lack of EU-specific tools within a certain country or in a certain area.These are cases where national quotas or opportunities of legal mobility are established to facilitate the conclusion of EU agreements with third countries, showing complementarity of EU and MS EXMIPO efforts.
Because of their multifaceted nature, we anticipate that tools (5), ( 6) and ( 7) can impact the EU EXMIPO in different ways.They may complement it, for instance, through TOAs established with countries where the EU is unable to get an agreement.They may converge with it, when reinforcing other EU tools.Or they might pose a challenge, by integrating return co-operation within broader strategic frameworks or other arrangements (e.g., police co-operation agreements and memoranda of understanding), even when the EC has a negotiating mandate (Cassarino, 2010).
Building on this typology, in the remainder of the article, we analyse Italian EXMIPO as a case study of MS external migration policies.

II. Methodology
The article centres on Italy to investigate MS EXMIPO, employing a case study approach.Although a case study analysis is narrower in scope, when compared to a large-n study, it offers a comprehensive understanding of a process and is a vital instrument for theory building and advancement (George and Bennett, 2005;Kohlbacher, 2006).
Specifically, Italy was selected as a crucial case of an MS that (a) is a vital hub of migratory flows to the EU and (b) prioritizes its EXMIPO.Because of its strategic location at the centre of the Mediterranean and role as a transit state for many of the migrants travelling further north,5 Italy's response to migration has crucial relevance for the whole EU.Moreover, although Italy is often considered a recent country of immigration, existing evidence suggests that its EXMIPO plays an important role, especially in the realm of returns, as Italy is the second MS by number of bilateral RAs with third countries (after Francesee Cassarino, n.d.).Questions therefore emerge on the extent to which Italy's EXMIPO extend beyond returns, what are its specific tools and its interaction with EU EXMIPO.It is a crucial case for testing our theoretical framework, providing rich evidence to explore the nuances of the external dimension of migration policies at the MS level, and whose actions hold significant implications for other MSs and the EU at large.
Having selected the Italian case, the empirical research involved a substantial effort to identify and classify all the instruments of the country's EXMIPO.Due to limited pre-existing studies, our first step involved retracing its development, through a qualitative analysis of primary and secondary sources.To that end, we reviewed all Italian immigration laws6 and relevant secondary literature, to identify key trends and dynamics.Then, employing our analytical toolbox, the team7 mapped the actual instruments of Italy's EXMIPO, identifying and cataloguing the many memoranda, treaties, agreements, protocols and so forth that make up Italy's EXMIPO. 8Building on such data, we compiled a dataset of Italy's external migration tools.We analysed the text of all the agreements and categorized each entry by year, country, region, instrument type, main objective and focus area(s).We also transformed the dataset into a map, for easy consultation and visual communication.9Finally, we conducted descriptive statistical analyses to identify temporal and geographical trends and compared our data with a similar dataset on EU externalization tools,10 to explore the relationship between the Italian and the EU EXMIPOs.
Overall, we analysed Italy's EXMIPO between 1990 and 2022, with 17 partner countries (Figure 1).We selected 1990 as the starting point for the analysis, as that is when Italy's first migration law was adopted.Country selection was based on an effort to cover a variety of subregions within what is often referred to as the 'broader Mediterranean' macro-region 11 : North Africa (Algeria, Egypt, Libya, Morocco and Tunisia), sub-Saharan Africa (Ivory Coast, Niger and Nigeria), Eastern Europe (Albania and Moldova), Middle East (Iran, Iraq, Lebanon and Turkey) and South Asia (Afghanistan, Bangladesh and Pakistan).Although the data collection covered 17 countries, for Afghanistan, no tools were found. 12

III. The Development of Italy's EXMIPO: Priorities and EU Context
Italy's commitment to enhancing collaboration on migration with third countries began in the late 1990s (Zotti and Fassi, 2020), at a time when the EU's external migration policy had not yet fully developed.Whilst EU EXMIPO was formally adopted at the 1999 Tampere European Council, where EU leaders urged the integration of migration and asylum into EU external policies, Italy's Turco-Napolitano Law (1998) had already established co-operation with third countries as a central strategy for managing both regular and irregular migration.This law created a comprehensive toolbox for migration management encouraging the interior and foreign affairs ministries to establish agreements with countries of origin to expedite the return of irregular migrants.Importantly, it also introduced annual 'entry quotas' for seasonal or employed work, prioritizing countries with 'agreements regulating entry flows and readmission'.In doing so, the law marked one of the earliest national efforts to systematize and institutionalize collaboration with third countries as a core approach to managing migration flows.
During the 2000s, concerns about employing the right incentives to secure migration co-operation became increasingly prominent in both MS and EU debates.In 2002, the Bossi-Fini Law reinforced the conditional link between preventing irregular migration and providing legal mobility channels, by establishing numerical restrictions on workers 12 Country selection was informed by the following considerations.For North Africa, all countries were selected due to their relevance for migratory flows to the EU.For sub-Saharan Africa, the Middle East and South Asia, all the studied countries were identified as key partner states, or mentioned for special co-operation, in the 2016 EU MPF.For Eastern Europe, we selected two countries with a relevant history of migration to the EU: a candidate country (Albania) and one that was not so at the time of country selection (Moldova).from states deemed un-co-operative 'in combatting irregular migration or in readmitting their own citizens'.Thus, conditionality was established as a fundamental aspect of Italy's EXMIPO early on.At the EU level, discussions about leveraging development aid for returns were also initiated that same year at the European Council in Seville, although formal adoption would occur only 13 years later, in response to the 2015 asylum crisis (Carrera et al., 2015).Furthermore, whilst the use of quotas and temporary work permits to support EU efforts in securing RAs had been discussed since the Franco-German proposal for a New European Policy in 2006 (Carrera et al., 2015), the credibility of these regular channels remained compromised due to the limited legal migration opportunities offered by the MSs (European Parliament, 2015).
Overall, during the late 1990s and early 2000s, whilst the EU EXMIPO was still taking shape, Italy engaged in intense migration diplomacy, beginning with Albania and later extending to North Africa.Between 1998 and 2010, Italy signed RAs with eight countries, namely, Albania, Algeria, Egypt, Moldova, Morocco, Nigeria, Pakistan and Tunisia, and established a dedicated 'Returns Fund' (2009 Security Package: art.1).Simultaneously, the 'Flows Decree' (Decreto Flussi) offered preferential entry quotas to citizens of countries 'that had signed migration cooperation agreements' (e.g., Decreto Flussi, 6 June 2003).
After the Arab Spring events in 2011, migration assumed even greater importance as a foreign policy issue for both the EU and the MS.In 2011, the General Approach to Migration and Mobility (GAMM) recognized the necessity of providing incentives to third countries and aimed to align the EU's and MSs' visa policies with broader migration considerations (EC, 2011, p. 3).In line with the EU's heightened focus on external migration policies and dialogue with North African nations, Italy relaunched dialogue with Tunisia, concluding a new memorandum with the transitional government in April 2011.Simultaneously, the 2012 Tripoli Declaration laid the foundation for political dialogue with the Libyan Interior Ministry to ensure the implementation of a previous 2008 agreement.In the same year, a technical agreement on search and rescue (SAR) operations was reached with Algeria.
The 2015 'migration crisis' brought the external dimension of migration policies into even sharper focus in the EU and Italy and across the country's entire political spectrum.In 2016, the Migration Compact proposal,13 presented to the EU by Matteo Renzi's centre-left coalition government, heralded migration co-operation with third countries as a pivotal strategy encompassing border protection, security, legal migration opportunities and resettlement.This reaffirmed the Italian government's commitment to addressing migration as both a domestic and foreign policy matter.Furthermore, the proposal aligned with two pivotal EU documents that became the cornerstone of EU EXMIPO amid the migration crisis: the EU Agenda on Migration (2015) and the Migration Partnership Framework (MPF) (2016).The former stressed the vital role of collaborating with third countries and emphasized the EU's readiness to employ 'all leverage and incentives' to secure and enforce RAs (EC, 2015, p. 10).The latter placed additional emphasis on using a combination of positive and negative incentives and urged MS to complement EU incentives with their own to strengthen the overall strategy (EC, 2016).
The subsequent centre-left cabinet, led by Paolo Gentiloni, intensified this approach to 'foreign immigration policy' (Zotti and Fassi, 2020).Building on years of bilateral collaboration, Gentiloni and Interior Minister Minniti led negotiations with the Libyan Government of National Accord.These negotiations resulted in the controversial 2017 Italy-Libya Memorandum of Understanding (MoU), which was also sponsored by the EU.The agreement followed the EU MPF's call for swift and practical returns, 'not necessarily [through] formal readmission agreements', but leveraging the 'special relationships' that MSs might have with third countries, to ensure that those with well-established bilateral ties would assume a central role in the EU's discussions with specific countries (EC, 2016).
Under the Renzi and Gentiloni cabinets, Italy's EXMIPO expanded towards sub-Saharan Africa.The former facilitated a police agreement between Italy and Nigeria, on trafficking in human beings and returns.The latter engaged in discussions with the Interior Ministers of Niger and Chad and promoted a '3 Ps' approach to migration management, focusing on 'partnerships' with countries of origin and transit, 'protection' of vulnerable refugees and migrants and 'prosperity' through investments in Africa. 14n 2017, the Gentiloni cabinet also established the €226 million 'Africa Fund' (Fondo Africa) to enhance co-operation with African states playing a pivotal role in migratory routes (Law 232/2016: art.1, c.621).The fund aimed to support development projects, facilitate voluntary returns, conduct information campaigns and prevent irregular migration (Decreto 'Fondo per l'Africa' 200/2017).Part of this fund was allocated to contribute to the EU Trust Fund for Africa (EUTFA) (see ActionAid, 2017).
To a significant extent, the intensification of Italian co-operation with North Africa, and its broadening to sub-Saharan countries, was driven by growing concerns of the domestic public, as 66% of Italians in 2017 prioritized border security and migration control in Italian foreign policy (Di Filippo and Palm, 2018, p. 67).However, it also aligned with evolving EU priorities: after the Arab Spring, EU documents such as the GAMM focused on political dialogue with North African partners, and since 2016, the MPF has promoted collaboration with countries beyond the immediate neighbourhood, especially in sub-Saharan Africa.
Subsequent populist governments continued to place a strong emphasis on Italy's EXMIPO.In 2019, Interior Minister Salvini introduced two security decrees, which, amongst other things, created a new fund to reward countries collaborating effectively on returns (Decreto-Legge 53/2019: art.12).In 2020, the Africa Fund was expanded to include countries beyond Africa and was renamed the 'Migration Fund' (Law 160/ 2019: art.1, c.878).More recently, the technical government under Prime Minister Draghi intensified co-operation with Libya and Niger in 2021, whilst establishing humanitarian corridors through new protocols with Iran, Libya and Pakistan.
The external dimension of migration policies played a significant role in Giorgia Meloni's 2022 electoral campaign too.Co-operating with North Africa, combating irregular migration and creating hotspots abroad were all essential goals of her electoral manifesto.The manifesto also advocated linking EU development funds to co-operation on returns and utilizing quotas as conditional incentives. 15Reflecting the above, Meloni's 2023 immigration decree introduced new quota schemes reserved for nationals of states engaging in information campaigns on the risks of irregular migration (Decreto-Legge 20/2023: art.1).

IV. The Toolbox of IT-EXMIPO: Evolution, Characteristics and Interaction With the EU Level
Having discussed the development of Italy's EXMIPO, we now explore its instruments.The analysis of the dataset reveals that Italy's EXMIPO has evolved quantitatively and qualitatively, alongside and despite EU initiatives.Specifically, the country's EXMIPO toolbox includes all nine elements from our analytical framework.
Between 1990 and 2022, Italy developed 125 instruments with the 17 countries we analysed.The data demonstrate that Italy's EXMIPO began in the late 1990s, with 12 instruments adopted in 1997-1999.The strategy was then further developed in the 2000s, when 51 EXMIPO tools were launched (including 12 in 2000 alone).
In contrast, the EU EXMIPO (towards the 17 partner countries studied here) was still in its infant stages in the 1990s, with only one instrument adopted (Figure 2).It then gradually evolved in the early 2000s and experienced rapid acceleration during the mid-2010s, especially after the Arab Spring and the 'migration crisis'.Notably, between 2015 and 2017, both Italy and the EU saw a significant surge in EXMIPO, with 17 and 16 new instruments adopted, respectively.Overall, between 1990 and 2022, Italy developed over twice as many EXMIPO tools as the EU (125 compared to 61), demonstrating a substantially broader quantitative reach in MS EXMIPO efforts.

Late 1990s and Early 2000s: Emerging EXMIPO, Instrument Diversification and Conditionalities
Examining the temporal evolution of Italy's EXMIPO in more depth reveals two distinct periods, with the early 2010s serving as a pivotal moment.The first period encompasses the late 1990s and the early 2000s and is characterized by three key features.
The first is the intensive development of Italy's external migration policy.This period marked the initiation of Italy's EXMIPO, starting with co-operation with Albania and later extending to North Africa.Due to increasing migration from Albania to Italy, between 1997 and 2002, the two countries negotiated eight migration-related agreements, covering seasonal work, migration control, police co-operation, SAR operations and readmission.Italy also strengthened its migration dialogue with North African countries during this time, engaging in consultations with the Moroccan and Algerian foreign ministries, signing agreements with Egypt and Tunisia on border management and seasonal work and establishing Treaties of 'Friendship and Cooperation' with Tunisia and Libya to enhance collaboration on irregular migration.
The second feature is the diversification of policy instruments: TOAs, RAs, political dialogue, quotas and labour migration deals were all introduced during this initial phase.As a recent country of immigration, Italy still needed to establish the external dimension of its migration policies, which led to the expansion and diversification of tools.
Consequently, the country's EXMIPO expanded significantly, as it explored various strategies and instruments.
The third characteristic is the use of quotas as incentives.During this period, a strong connection developed between return agreements and quotas: the first series of RAs were established between 1998 and 2002, and they often went hand in hand with the adoption of country-specific quotas to encourage co-operation on returns, as set out by the Turco-Napolitano and Bossi-Fini laws.Indeed, for countries like Albania, Morocco, Tunisia and Moldova, RAs and annual entry quotas were introduced in the very same year (1998 for the first three countries and 2002 for Moldova).The Decreto Flussi allocated up to 1500 quotas for Moroccan and Tunisian citizens, 'considering the previously established bilateral agreements' signed that year with these two countries, and in 2002, it introduced 500 entry quotas for Moldova for the first time, following the signing of an RA.Quotas were used as both incentives and rewards: in the case of Egypt (and, later, Ivory Coast), quotas were granted first, seemingly to encourage migration dialogue and co-operation.In the case of Nigeria and Pakistan, reserved quotas were granted only a few years after signing an RA.In the case of Algeria, entry quotas were provided 7 years after the signature of the RA, when the agreement finally became effective.

Post-2010s: Increasing Informalization and Irregular Migration Concerns
The 2010s marked Italy's EXMIPO second phase, characterized by a growing trend towards informalization and an increasing focus on irregular migration.
In this second phase, the informalization of migration policies became increasingly prominent, as evidenced by the decline in the establishment of formal instruments such RAs, quotas and labour agreements (Figure 3).Political dialogue has become the most frequently used tool since 2010, with notable peaks in 2012 and 2019.Instruments have also become less diverse, with approximately 66% of the tools adopted between 2011 and 2022 falling into the categories of TOAs or political dialogue.
The slowdown in formal agreements during the early 2010s might appear surprising, given the escalating migratory flows to Italy and Europe during this period.However, despite the urgency to collaborate on migration issues in the aftermath of the Arab Spring,  This trend to informalization at the MS level, as exemplified by the Italian case, reveals that increasing informality is not unique to the EU EXMIPO (Slagter, 2019).Given the heightened politicization of migration since 2015 and the difficulties in establishing formal co-operation agreements, informality enabled MSs to adapt to evolving situations, circumvent legal and institutional constraints, bypass common EU policies and gain more flexibility (Cardwell and Dickson, 2023, p. 3124).
The shift that began in the early 2010s is evident not only in the types of instruments used but also in their content.In the late 1990s and early 2000s, Italy's EXMIPO tools addressed regular and irregular migration concerns equally.However, they then evolved differently, with regular migration instruments decreasing and irregular migration tools continuing to increase (Figure 4).The parallel trend that characterized the first period, particularly 1996-2004, reflects the use of quotas as leverage to secure return agreements.The bifurcation of instruments addressing regular and irregular migration post-2005 mirrors the increasing emphasis placed on combating irregular migration at both the EU and national levels.This divergence becomes particularly evident in 2017, when eight tools were introduced to address irregular migration, whilst none were focused on regular migration.Therefore, with the shift towards informality, there was a significant reduction in labour migration agreements and quota schemes after 2010.
The only tool in contrast to the general move towards informality and irregular migration is represented by humanitarian corridors.First introduced in 2015, they provided for safe and legal mobility of people in need of international protection, from a country of first asylum to Italy (Caritas, 2019).

Iole Fontana and Matilde Rosina
In summary, Italy's EXMIPO can be divided into two phases.The late 1990s and early 2000s witnessed significant development and diversification of policy instruments and the regular use of quotas as incentives or rewards for return agreements.Conversely, the 2010s saw a shift towards informality and a heightened emphasis on irregular migration.Ironically, as Italy's focus on irregular migration intensified, its ability to establish formal return agreements declined, with only two such tools being established since 2010.

Competition, Convergence or Complementarity? Tools' Characteristics and Interaction with the EU EXMIPO
Having explored the evolution of Italy's EXMIPO, we now consider the specific instruments used and how these interacted with the EU EXMIPO.
Introduced in 1998 by the Turco-Napolitano Law, return and readmission agreements were an early component of Italy's EXMIPO.As discussed in the section on The Policy Tools of MS EXMIPO: In Search of a Typology and Conceptual Framework, the literature often highlights the competitive dynamics between the MS and the EU regarding the signing of RAs, as bilateral return agreements can pose a challenge to negotiating a EURA, diminishing the EU's bargaining power.This was evident in several cases.For instance, in the Moroccan case, despite the Commission's mandate to negotiate readmission and visa facilitation agreements with the country in 2000, only one round of negotiations took place, with relations mostly remaining at the bilateral level.Italy's migration relationship with Morocco involved a range of instruments, including an RA in 1998 and quotas and a labour migration agreement in 2005.Whilst the bilateral RA predates the Commission's mandate and does not violate the principle of loyal co-operation, it aligns with other bilateral agreements limiting the Commission's negotiating authority.A similar situation is observed with Nigeria, which signed RAs with Italy and other MSs and prefers bilateral co-operation (Zanker et al., 2019).
In some cases, disjointed governance was even more visible, with Italy continuing to pursue RAs even after the Commission received a negotiating mandate.One such example is Algeria: bilateral co-operation began in the late 1990s, resulting in an RA signed in 2000.Although the Commission received a negotiating mandate for Algeria in 2002, negotiations never started.Meanwhile, Italy's agreement with Algeria came into effect in 2008.As mentioned above, according to EU law, once the Commission receives a negotiating mandate for an RA with a third country, MSs are expected to halt their independent negotiations with that country (ECA, 2021).However, despite the Commission's mandate in 2002, Italy's bilateral agreement became effective shortly thereafter.Similarly, in the case of Tunisia, whilst the EU obtained a negotiating mandate for an RA in 2014, a new bilateral (though informal) agreement between Italy and Tunisia was signed in 2020.
The above findings confirm the theoretical expectation that, through competition, bilateral RAs can hinder the development of the EU EXMIPO.Partner countries like Morocco, Tunisia, Algeria and Nigeria have demonstrated a preference for bilateral negotiations because they can exert more bargaining power when dealing with individual MSs rather than the EU as a whole (Abderrahim, 2019;Kaiser, 2019).This preference is also due to individual MSs being able to offer immediate incentives, such as Italy promptly providing quotas to Algeria, alongside the RAs' implementation with clear patterns of The tools of external migration policy in the EU member states: The case of Italy de-coupled governance.Since the EU, as a collective entity, cannot provide such immediate incentives, the Commission's capacity to offer enticing rewards is limited.
These limitations, coupled with the increasing politicization of migration following 2015, compelled the EU to acknowledge, if not openly endorse, MS bilateralism and informality.In the 2016 MPF, the Commission explicitly condoned both approaches, emphasizing that the goal is not necessarily to pursue formal RAs but to achieve quick returns and encouraging the MSs to explore bilateral opportunities (European Commission, 2016, pp. 7, 8).Consequently, whilst MSs' reliance on bilateral RAs may impede EU co-operation efforts, it has come to be accepted as a 'necessary evil' to pursue the broader objective of enforcing returns.
Turning to other EXMIPO tools, as previously discussed, RAs were closely tied to the use of quotas, especially during the initial phase of Italy's EXMIPO.Quotas served as positive incentives for bilateral return agreements.Interestingly, however, in some cases, quota schemes served as national incentives for the conclusion of EU agreements, with bilateral tools complementing the EU EXMIPO's multilevel governance.Bangladesh's case is illustrative.Italian-Bangladeshi talks began in 1994 (Cassarino, 2005, p. 17), but no agreement was finalized.Consular co-operation to ease return procedures outside a specific agreement was initiated in the early 2000s, 16 and from 2003 to 2012, Bangladesh was formally reserved entry quotas for seasonal and employed work.However, in 2013, the country was removed from preferential lists, possibly due to the ongoing failure to reach an RA and establish 'a program of returns with the country'. 17Bangladesh was eventually reinstated on preferential lists in 2020, 3 years after informal standard operating procedures for return with the EU were signed.The case of Bangladesh supports our theoretical expectations that national quota schemes can complement the development of the EU EXMIPO, acting as national incentives or rewards, for EU-wide agreements.
In terms of resettlement, Italy operates mainly through humanitarian corridors, which are currently active for refugees in Lebanon, Libya, Niger, Iran, Pakistan, Ethiopia and Jordan.Humanitarian corridors reinforce EU resettlement endeavours, as explicitly recognized in the 2021 MoU for Afghan refugees in Pakistan and Iran, 18 which aligns Italy's strategy with the EU's recommendation to resettle 50,000 refugees and create private sponsorship programmes.Italy has been a pioneer in implementing humanitarian corridors, and its community sponsorship programmes were highlighted as 'good practice' models by the EC (2020).Not only has this new model been replicated in other MSs, but it has also directly contributed to the overarching EU strategy for the development of legal pathways, confirming the convergence of national and EU resettlement programmes.
In terms of political dialogue, a key instrument for enhancing bilateral relations leading to the conclusion of return or other agreements, Italy has often played a pivotal role as a facilitator for the EU.The EU-Tunisia MoU of July 2023 is a prime example of Italy's mediation efforts.Actively engaging in migration talks involving Tunisia and the EU, Italian ministers chaired EU meetings on migrant smuggling in 2020 (attended by EU states and Tunisia, Algeria and Morocco) and accompanied Home Affairs Commissioner Johansson on missions to Tunisia in 2020 and 2021 (EU Council, 2022).More recently, Italian Prime Minister Meloni emphasized her role as an EU spokesperson during her visits to Tunisia, which Johansson depicted as 'crucial'.In the Commissioner's words, 'Italy plays a constructive role in our relations with Tunis.On this, the Commission and Rome are allied in increasing cooperation with this country' (Pascale, 2023).Italy has also facilitated political dialogue for other countries, as exemplified by Italian Foreign Minister Di Maio accompanying Commissioner for Neighbourhood and Enlargement Varhely to Tripoli in May 2021 (EU Council, 2022).These considerations align with the expectation that MSs may complement EU action, serving as facilitators of political dialogue and correspond to the MPF's invitation for MSs with well-established relations to participate in EU negotiations with third countries.
MS tools can converge with, and complement, EU instruments through other migration agreements and TOAs too.The 2017 Italy-Libya MoU best exemplifies this, showcasing the close alignment between EU and Italian actions in developing the agreement.First, the EU promptly endorsed the MoU, with the European Council declaring its full support to the implementation of the agreement, the day after its signature (European Council, 2017).Second, the MoU's objectives closely matched the EU's own goal of reinforcing Libya's borders and coastguards (European Parliament, 2020, p. 44).Third, the memorandum specified that expenses would be covered by Italian and EU funds, emphasizing the synergy between national and EU EXMIPO actions.Indeed, the EUTFA allocated over €90 million to the Italian Coastguard in 2017-2018 for enhancing the Libyan coastguards' capacities (European Parliament, 2020, p. 45).In summary, whilst the MoU was signed by Italy, it has become a cornerstone of the EU EXMIPO, showcasing the overlap between national and EU EXMIPO efforts.
Finally, the parallelism between EU and Italian actions can be seen in the realm of military missions.Although only two Italian military missions with a migration component were identified, their alignment with EU actions is noteworthy.In Niger particularly, both EUCAP Sahel and the Italian MISIN missions focused on migration-related objectives: the former aimed to improve border management against traffickers and smugglers 19 ; the latter was meant to 'counter illegal traffics and security threats', including irregular migration, 20 showcasing the complementarity of national and EU EXMIPO instruments.
Overall, the analysis reveals that the external dimension of Italy's migration policies is strictly intertwined with that of the EU, through a complex and dynamic network of relations between national and supranational instruments, encompassing convergence, competition and complementarity.

V. Conclusion: The 3Cs of MS EXMIPO
Whilst the EU and its MSs share a strong interest in effective migration management, our understanding of the external dimension of MS migration policies remains surprisingly limited.The concept of 'external dimension' has been explored so far mainly at the EU 19 See https://www.eucap-sahel.eu/gestion-des-frontieres-et-migration-irreguliere/(last accessed 6 September 2023). 20See https://www.difesa.it/OperazioniMilitari/op_intern_corso/Niger_missione_bilaterale_supporto/Pagine/default.aspx(last accessed 6 September 2023).
The tools of external migration policy in the EU member states: The case of Italy level, but it was never fully developed in the case of the MSs, beyond the issue of return and readmission agreements.In this article, we have argued that the notion of external dimension is not only limited to the EU but also applies to the national level.Specifically, we interpreted the external dimension of MS migration policies as encompassing a diverse set of tools to collaborate with third countries, in pursuit of domestic policy objectives.
This article has thus sought to offer an analytical framework to conceptualize MS EXMIPO, by (1) unpacking its tools and (2) shedding light on their interaction with EU-level instruments.By doing so, it has aimed to enhance our understanding of MSs' role in EU's external migration governance.
Through the analysis of the Italian case, this article has demonstrated that MSs have engaged in intense external action at the bilateral level in the field of migration, which has expanded quantitatively and qualitatively, alongside (and despite) EU action.Whilst Italy is likely to be amongst the MS with the most extensive EXMIPO (as exemplified by its high number of RAs), we would expect France, Germany and Spain to have developed an equally dense network of parallel bilateral agreements, thanks to their role as key receiving or transit countries, 21 22 as well as their extensive bilateral networks, even beyond migration. 23On the flipside, the argument that MSs have developed an intense external action might be less applicable to states that are more peripheral to the geography of migratory flows to the EU and that have traditionally received more limited numbers of asylum applications (such as Scandinavian and Eastern European countries). 24Yet, even for those MSs, co-operating with third countries is likely to play an increasingly crucial role, driven by the failure of internal migration policies, the salience of migration for domestic politics and the socialization of MSs' interests through EU institutions.
Building on the analysis of MS EXMIPO tools, we have argued that these interact with EU-level instruments in three primary ways: competition, convergence and complementarity.Competition was evident in matters related to returns, with Italy's RAs with Tunisia and Algeria being a case in point.This evidences the MSs' reluctance to relinquish national instruments and the accompanying power and underscores the intricate power dynamics that underlie the tools of external migration policy.On the contrary, Italy's humanitarian corridors were a prime example of convergence, in which national and EU tools aligned, resulting in a co-operative approach to addressing migration challenges.Finally, complementarity was exemplified by situations in which Italy served as a facilitator for EU-level political dialogue with third countries (as it did in the lead-up to the 2023 EU-Tunisia MoU) and when national and EU tools were employed to compensate for the other's shortcomings (such as the difficulty for the EU to set up legal migration pathways).
Our theoretical framework can shed crucial light onto other MSs' EXMIPO.As the communitarization of EXMIPO gradually advances (as proposed by the 2020 Pact on Migration and Asylum), we would expect an increasing convergence in MS priorities and tools.Returns may continue being an exception and an example of competition, as exemplified by Germany's recent efforts at securing bilateral agreements with Morocco and Nigeria, 25 although the Commission's endorsement of informality and bilateralism appears to be shifting them towards a case of complementarity.Finally, given the salience of migration across the Union, MSs other than Italy are likely to be acting as pivotal intermediaries, or even on behalf of the EU, to secure agreements with countries where they have well-established relationships.France's co-chairing of migration meetings together with Senegal and Ivory Coast and Spain's central role in the EU's Migration Dialogue with Latin America (EU Council, 2022) are pertinent examples.
In conclusion, the external dimension of MSs' migration policies is intricately linked with the EU's, forming a complex web of interactions between national and supranational tools.A policy tool approach paves the way for new avenues of research, enabling deeper understanding of diverging or converging interests, of instances of policy learning and emulation and of the fundamental power dynamics at play, promising to be a valuable instrument in unravelling the complexities of migration governance.

Figure 2 :
Figure 2: Comparison of EU and Italy's EXMIPO Tools With 17 Selected Partner Countries, 1990-2002.[Colour figure can be viewed at wileyonlinelibrary.com]

Figure 4 :
Figure 4: From Quotas to Migration Prevention: IT-EXMIPO Instruments by Focus Areas.[Colour figure can be viewed at wileyonlinelibrary.com]

Table 1 :
EU External Migration Policies: Categorization of Instruments.
Source: Authors' elaboration.Iole Fontana and Matilde Rosina 6 © 2024 The Authors.JCMS: Journal of Common Market Studies published by University Association for Contemporary European Studies and John Wiley & Sons Ltd.

Table 3 :
Example of the Application of MS EXMIPO Typology.