Keywords

1 Introduction

In the last two decades, South America created a regional regime for human mobility that is regarded as the most developed one after the EU (Acosta 2018; Geddes et al. 2019; Lavenex 2019). This regime is characterized by equal social and economic rights and working conditions, non-criminalization of migration, the enunciation of the ‘right to migrate’, and the right to reside in other Member States. In the last two decades, South America has also become an important regional actor in migration-related international forums, as it sustains and promotes very unique positions. During the negotiations for the Global Compact for Migration (GCM), for example, it called for the global recognition of the ‘right to migrate’, for the universality of migrants’ rights and, instead of expulsions, for migrant regularization as a solution to irregularity. These positions are in sharp contrast with the ones sustained in Europe and the US and challenge the assumption that there is a trade-off between openness to migration and rights (Ruhs 2013) and the ‘need to fight against irregular migration’ that prevails in Europe and in the US (Acosta and Geddes 2014). It is also the only region whose Regional Consultative Process (RCP) do not focus on security issues (Lavenex 2019).

Despite these important advances and the ground-breaking positions that this region sustains at the global level, South America remains understudied in governance, Regional Integration (RI) and migration studies, particularly in the English-language literature. Moreover, most of the regional achievements are largely unnoticed by the literature on the Free Movement of Persons (FM), which is mainly centered on the case of the EU.

The two main RI processes in South America are the Andean CommunityFootnote 1 (CAN, for its initials in Spanish) and the South American Common MarketFootnote 2 (Mercosur, for its initials in Spanish). Despite the varied institutional configuration of these two Regional Organizations (ROs), I will argue that the human mobility agenda in ROs tends to follow ‘generations’ of policies.

My argument is that the ‘first generation’ of policies for the mobility of persons started with the creation of the CAN. It was centered on labor mobility and it was selective, as it facilitated the movement of the most qualified workers. The ‘second generation’ (1991–2001) was highly technical and viewed intra-regional migration only from an economic perspective. During the ‘third generation’ of policies (2002–16), South America developed a regional mobility regime that has a strong discursive component linked to the promotion of a regional identity or ‘regionality’ (Van Langenhove 2011). The regime has a dual character: on one hand it expands and promotes migrants’ rights while creating a right to residence and on the other hand, it improves border controls. Mercosur turned out to be the leader in proposing regional migration policies for the whole of South America by multilateralizing its Residence Agreement in the early 2010s, and by developing more far-reaching measures than the CAN. This can be explained by three factors: 1- Argentina’s ‘thematic’ leadership. Argentina was recognized as a legitimate leader on migration issues by her regional partners, who accepted her proposals, mostly aimed at regularizing her irregular immigrant population; 2- foreign policy strategies of the leading countries (Argentina, Brazil and Ecuador) that prioritized relations with the region; 3- the ideologies of the governments in power in that moment, that had a strong focus on human rights and opposed the US and the EU’s ‘restrictive’ policies (Brumat and Acosta 2019). Mercosur was the leading RO because Argentina, the country with most ‘thematic’ or ‘positional’ power and Brazil, the country with most aggregate power (Zartman and Rubin 2000), are part of it, and not of the CAN. I finally identify the possible emergence of a fourth generation (2015–present), where many South American governments experienced a ‘conservative turn’, evident in more restrictive domestic measures, views and ideas around migration that tend to praise policies of countries in the Global North. I also propose four possible scenarios for the medium and short term.

This chapter will make a contribution to the studies of regional migration governance in the Global South by explaining the development of this regional regime throughout time. I adopt a governance approach to better understand policymaking processes and policy change. I understand governance as a structure and as a process. As a structure, it implies “the architecture of formal and informal institutions” that regulate human mobility in a region and as a process it signifies ‘the dynamics and steering functions involved in never-ending processes of policy-making’, that include a wide range of actors seeking to govern and influence ‘the design of institutions and mechanisms in order to shape choices and preferences’ (Levi-Faur 2012, p. 7).

This chapter draws information from more than a hundred in-depth interviews with key government actors, as well as representatives of the business sectors and unions in Mercosur and CAN that took part in the elaboration of these norms, conducted between 2012 and 2018 within and outside the framework of the MIGPROSP project. It is also based on an analysis of the regional legislation on migration as well as policy documents.

The chapter is divided into six sections. In the first section, I define regional policies for the movement of persons, the concept of generation, and processes of convergence. In the second section, I briefly describe the main characteristics and evolution of migration flows and stocks in South America. In the following three sections, I explain the main characteristics of each generation of regional migration policies and analyze the development of a regional regime for the movement of persons and regional policy convergence. I finally conclude and discuss the possible emergence of a ‘fourth generation’ of policies and propose three possible scenarios for this policy agenda for the short and medium term.

2 Regional Policies for the (Free) Movement of Persons

I consider that regional policies for the mobility of persons are comprised of international (bilateral and multilateral) legislation that regulate migration flows. This legislation can be either adopted by the agreement between two or more states in a specific region, or by the institutions of a specific RO.Footnote 3 The instruments and operative actions taken by the institutions of a RO aimed at executing what was established in the international regulations are also part of regional policies. The regional dimension of regional migration policies is not given by the nationality of the persons whose mobility is regulated, but by the fact that the states of the same region or a RO adopt and develop a common policy.

The instruments and operative actions include three subareas: (1) measures for the (cross-border) entry and exit of persons; (2) measures that regulate the residence of the persons in the territory; and (3) measures that affect the execution of labor activities within a specific region, for the nationals or non-nationals of the member states (Brumat 2016; Brumat and Acosta 2019). Thus, my analysis will only look at the regulations that deal with these dimensions.

The human mobility agenda in ROs tends to follow ‘generations’ of policies. The concept of ‘generation’ is useful to compare regional thematic agendas, as it can better grasp their complexity and understand the ‘coexistence of several kinds of regional arrangements different in quality/content, while meanwhile also acknowledging that some forms of regionalism build upon previous ones’, thus avoiding the ‘strict separation’ of RI processes into ‘chronological clusters’ (Langenhove and Costea 2005, p. 2). These generations of policies, particularly in Latin America, are strongly influenced by the ideological and political context (see Dabène 2012). Generational change is determined by an alteration in 1- one or more of the three dimensions of regional policies for human mobility and 2- in the actors that take part in the negotiation, formulation, adoption and development of the policies.

The processes leading to the adoption of policies for the intra-regional movement of persons involve discursive acts and interactions in which ideas are exchanged. I consider that ideas matter, particularly in South America where ‘rhetorical regionalism’ (see Jenne et al. 2017) characterizes RI processes. These ideas lead to labels and categorizations that result in outputs of regional governance (Geddes and Vera Espinoza 2018). The sustained interactions and exchanges of ideas can lead to policy convergence. Convergence is ‘the tendency of policies to grow more alike, in the form of increasing similarity in structures, processes, and performances’ (Kerr 1983 in Drezner 2001, p. 3). It can be either spontaneous, when a country unilaterally ‘adapts itself’ to other countries’ policies, or it can be ‘intentional’, when countries voluntarily harmonize their legislation (Malamud 2011, p. 225). Migration policy convergence (‘formal convergence’) refers to the ‘increasing commonality of the methods used to control population movements’ (Pellerin 1999a, p. 996). The existence of a ‘regional paymaster’ that is willing to ‘offer’ regional integration and pay a higher cost than its partners for the adoption of regional policies (Mattli 1999) collaborates with policy convergence.

3 Migration Flows in South America

Latin America currently hosts 10 million immigrants in its territory and, at the same time, 38 million Latin Americans are emigrants (within or outside the region), which makes it the third sending region in the world, after Asia and Europe (United Nations 2017). But this has not always been the case: between the late 1800s and until the 1950s, some South American countries such as Brazil, Argentina and Uruguay were net receiving countries of mainly European immigrants.Footnote 4 Extra-regional emigration in Latin America is a relatively recent phenomenon, but intra-regional migration is a ‘classic’ pattern in the region, due to its frequency and historical roots (Martínez Pizarro and Villa 2005; Durand 2009). In South America in particular, mobility in border areas is one of the prevailing patterns, which is facilitated by geographical and cultural proximity and commercial exchanges (Martínez and Vono 2005; Massey et al. 2008). The prevailing migration corridors are from Bolivia, Chile, Paraguay and Uruguay to Argentina and from Colombia and Ecuador to Venezuela.Footnote 5

Migration flows in the region are highly dependent on the economic and political situation in the sending and receiving countries and on migrant networks (Martínez Pizarro and Villa 2005; Durand 2009). Intra-regional flows in Latin America started increasing in the 1970s: stocks went from 1,200,000 in 1970 to 2,000,000 in 1980 (Martínez Pizarro and Villa 2005).Footnote 6 In 1990, 2,200,000 intraregional migrants were registered. This small increase was due to the political instability in the 1980s. After the return to democracy, intra-regional stocks increased again, reaching 2.900.000 in the year 2000 (Martínez and Vono 2005; Martínez Pizarro and Villa 2005). The improvement of the economic situation in some countries such as Argentina, combined with a decrease in emigration to extra-regional destinations such as Spain and the USA, collaborated with the sharp increase in the intra-regional migration stocks in this decade: in 2013, 5,400,000 intra-regional migrants were registered (IOM 2015). The Southern Cone (Argentina, Brazil, Chile, Uruguay and Paraguay) is the subregion where immigration increased the most and where most intra-regional mobility happens (OAS and OECD 2017).

The two main receiving countries in Latin America have traditionally been Argentina and Venezuela. In 2000, these two countries alone hosted almost two-third of intra-regional migrants (Martínez Pizarro and Villa 2005, p. 6) and in 2010, they hosted 46% (IOM 2015). This situation changed dramatically since 2015, when Venezuelans started to emigrate in very large numbers due to the deep economic, social and political crisis in their country. By the end of 2018, Venezuelan emigration had reached a stock of 3.4 million, 95% of which reside in other South American countries (UNHCR 2019).Footnote 7

4 The First Generation: The Andean Pact and the “Unidimensional Migrant”

The Andean Community was part of the second wave of Latin American regionalism. The Economic Commission for Latin America and the Caribbean (ECLAC) promoted the industrialization of the Latin American economies and boosting intra-regional trade as a way of reversing underdevelopment. RI processes were seen as a way of optimizing the costs of industrialization (Malamud 2010; see Granato and Perrotta 2015).

These ideas drove the making of the initial regional policies for human mobility. The first generation of regional migration policies involved only the states (mainly the Ministries of Labor and Interior) and the RO as the main policymaking actors. This generation started in 1977 with the Andean Instrument of Labor Migration (Decision 116), reformulated in 2004 with the Decision 545 (see infra.). It was a foundational document, as it was the first regional agreement that typified migrant categories and migration procedures. It led to an increase in the interaction between member states that resulted in the regularization of undocumented migrants (Mármora 2004). It also stated the preference for nationals of member states (Torales et al. 2003) (what is now known as ‘nationality criteria’). Furthermore, it multilateralized basic rights such as non-discrimination and national treatment (art. 12), and it also guaranteed equal rights for education, housing, health and social security (art. 13).

The Instrument standardized the categories of regional workers that could enter the territory of each member state. However, the legislation prioritized qualified workers, because of the order in which those (four) categories were enumerated, as well as the restricted definition of “worker” which was limited to wage-earning jobs. It also created “Labor Migration Offices”, run domestically by each ministry of labor, which were in charge of selecting migrant workers (art. 7.c). This is one of the main characteristics of this policy generation, which was modified in the following ones: the selectivity of migrant workers and the preference for workers in employments of higher qualifications.

The second piece of legislation was the Andean Instrument of Social Security (Decision 113), reformulated in 2004 with the Decision 583. Its first version regionalized certain rights such as equal treatment (art. 4). But as these instruments do not acknowledge the social, political and cultural dimensions of migration, or even migrants themselves, and were only focused on the economic aspects of migration, they were described as ‘unidimensional’ (Stang 2009).

Both instruments stopped being enforced in the mid-eighties (Martínez Pizarro and Stang 2006), during a period of crisis in the RI process due to the economic and political crisis created by the debt crisis. But due to their continuous reference to national legislation, they never turned into a ‘real’ regional policy (Torales et al. 2003, p. 90).

5 The Second Generation: Open, Uniaxial Regionalism and the Absence of the Social Dimension

The ‘third wave’ of Latin American regionalism was known for ‘open’ regionalism models that combined preferential regional commercial agreements between countries with significant asymmetries, and extra-regional openness. Open regionalism was closely linked to the predominant neoliberal ideology, centered in commercial liberalization as the main and only end of RI, and a means for the countries to take part in a globalized economy. For this reason, they were labelled as ‘uniaxial’ regionalisms (Comini and Tussie 2016). They tried to attract FDI while at the same time they developed policies that were ‘technical’ in appearance and consequently, ‘non-political’. As Perrotta (2013, translation is mine) notes, there is an ‘apparent contradiction between the hiding of the political dimension and the fact that those are experiences driven by governments (they constitute real public policies)’. During the nineties, Argentina, the main receiving country and the one that made most of the proposals for regional policies for the movement of persons, prioritized relations first with the US, and then with Latin America, Europe and Japan (see Colacrai 2004; Tokatlian and Merke 2014).

5.1 The Andean Community During the Nineties/Early 2000s

After the Andean Pact had a structural crisis during the eighties due to the debt crisis, its institutional configuration was reformed and made more ‘flexible’.Footnote 8 Its name was changed to Andean Community (CAN). CAN’s name shows a fundamental difference with Mercosur’s: it aims to be a ‘community’ and not just a common market.

In the CAN, this second generation of regional policies responded to the neoliberal ideology that prevailed in the region and thus, was focused on ‘technical’ aspects of human mobility which involved, mainly, the optimization of border crossings.

The regional legislation that was adopted during this stage was the Andean Migration Card (TAM, for its initials in Spanish) in 1996 (CAN 1996, Decision 397), which seeks to improve the exchange of information about the persons that enter or exit the states’ territories. In 2001 four policies were adopted: the ‘Border Integration Zones’ (ZIF, for its initials in Spanish) (CAN 2001a, Decision 501) and the ‘Binational Centers of Attention at the Border’ (CEBAF, for its initials in Spanish) (CAN 2001b, Decision 502), —both of which are policies seeking to improve the management of the movement of services, goods and persons in border areas- Decision 503 for the recognition of the national documents (CAN 2001c), and the ‘Andean Passport’ (CAN 2001d, Decision 504) which regulated the documentation to enter and exit the states’ territories.

Decisions 503 and 504 introduced some elements that had not been present in regional legislation until that moment: Decision 503 defines the free movement of persons as a ‘right’ and both decisions call for the consolidation of the ‘Andean identity’. The call for an ‘Andean identity’ responds to pre-existing ideas that started to gain more importance in the early 2000s. The pre-existing idea is the one of creating, as mentioned, a ‘community’ and achieving Andean ‘integration’, something that had been present in the subregion historically and that was consolidated in the Cartagena treaty. In the early 2000s, the idea of a regional citizenship and of promoting a sense of belonging, that had already been debated in the regional institutions during the nineties, started gaining ground and was included in regional legislation (official of the government of Ecuador, interview, April fourth 2019).

5.2 Mercosur During the Nineties

In the first 4 years of its existence, and until the adoption of the Ouro Preto Protocol (OPP) in 1994, Mercosur’s objective was to complete a Common Market (CM) and thus, to achieve the free movement of persons. For this reason, the first policies that were adopted during this time tried to facilitate the movement of persons as a way of increasing intra-regional economic exchanges. So, Mercosur’s first measures were the implementation of preferential channels in airports for Mercosur citizens (CMC 1991), integrated border controls (with the so-called ‘Recife Agreement’) (CMC 1993a, b), and the regulation of national documents that were valid for travelling inside the area (apart from passports) (GMC 1994a, b).Footnote 9

The OPP redefined the RI process as an imperfect customs union (CU), downgrading the end goal of the integration from CM to CU, and eliminating the free movement of persons as an objective. Consequently, migration issues were labelled as ‘labor migration’, which reduced the scope of regional policies in this area (Pérez Vichich 2007). The agenda was fragmented in different policy areas, and migration issues started to be treated in diverse regional institutions. The governments’ priority was to ‘adapt’ regional migration policies to the level of the economic integration, an idea that was promoted and sustained by the Argentine government in most negotiations on social and labor issues (see Almeida Freitas 2009, p. 281). As stated by a Brazilian official: “[when] we talked about the movement of persons, there was a discourse that, well, the movement of persons is linked to the movement of the factors of production” (official of the government of Brazil, interview, 13th November 2015).

In 1996, at the same time that the CAN adopted the TAM, Mercosur created its equivalent, the Entry and Exit Card (TES, for its initials in Spanish) (GMC 1996b, c). In 1997 and 1998, two important policies that regionalized workers’ rights were adopted: the Multilateral Social Security Agreement (MSSA) (CMC 1997), which created a transference mechanism for pensions and the Socio Labor DeclarationFootnote 10 (DSL, for its initials in Spanish) (Mercosur 1998). These two agreements resulted from the pressure of a new regional actor: regionally-organized worker unionsFootnote 11 within regional institutions.Footnote 12 The unions’ strategy was to influence decision-making in Mercosur as a way of ‘securing’ workers’ rights at the regional level, as they had lost their influence on national governments in a neoliberal context. A high-level representative of the CCSCS explains that the Mercosur was “a platform where [we] could say what we could not say at the national level” (CCSCS high-level representative, interview, 9 September 2015).

Some of the provisions included in these agreements are related to equal rights for workers’ families, non-discrimination, and national treatment. The DSL promotes the future achievement of the free movement of workers. The mechanisms established by the MSSA, as was pointed by some government officials that were interviewed, helped to reduce the bureaucratic procedures and increase dialogue between national bureaucracies of the member states, as they are basically ‘obliged’ to be in contact almost daily to coordinate actions (official of the government of Brazil, interview, 11th August 2015). Increased interaction led to the development of a sentiment of regional identity (Gómez-Mera 2005) and was a factor that later helped with policy convergence.

As part of the ‘technical’ focus of neoliberalism, Mercosur also adopted measures for rapid border-crossing procedures for residents in border areas (the Neighboring Transit Credential or TVF, for its initials in Spanish) (CMC 2000a), for the exemption of translation for documents for migration purposes (CMC 2000b), and an Agreement for the Exemption of Visas between Mercosur Member States (CMC 2000c). This last agreement was limited to temporary, high- qualified workers in the area of services and was never adopted by all the members, so it was only implemented bilaterally by the ones that had approved it.Footnote 13

The end of this generation of policies was brought by a deep economic-political crisis that almost led to the disappearance of the Mercosur (see Gómez-Mera 2005). It signaled the failure of the neoliberal model and opened a new phase in South American regional integration.

6 The Third Generation of Regional Policies for the Movement of Persons: The Social Turn

This stage began with the end of the 2001–2002 crisis and the change of the governments in the region. The new administrations have received diverse denominations: populist (Freindenberg 2007; Gratius 2007), neopopulists (Walker 2008), progressives and/or redistributionists (Perrotta 2013), leftist (Castañeda 2006; Gratius 2007; Riggirozzi 2012), social democrats (Walker 2008), developmentalist (Alves Texeira and Desiderá Neto 2012). Despite the diversity of denominations, they share certain characteristics, namely: the rejection of the pro-market policies of the previous decade and the greater importance that is given to social issues.

As a result of the traumas created by the 2001/2002 crisis, South American regionalism was reformulated. A more ‘cautious’ attitude towards unilateral trade liberalization was adopted (Bouzas et al. 2007, p. 17), together with policies that were oriented towards state intervention in the economy.

These changes led to the adoption of new concepts in the specialized literature: ‘postneoliberal regionalism’ (da Motta Veiga and Rios 2007; Sanahuja 2012) and ‘posthegemonic regionalism’ (Riggirozzi 2012).Footnote 14 Briefly, these concepts try to explain the main features of this stage: the regional agenda was widened, including issues other than economic, for which it was labelled as ‘multiaxial’ (Comini and Tussie 2016). The new regional priorities were the political agenda, social questions and development. The predominant political discourses rejected open, ‘uniaxial’ regionalism and American hegemony and influence in South America’s affairs. This reflected the foreign policy orientation of the leading countries (Argentina and Brazil), which shared some commonalities: they prioritized relations with the neighbors, they sought to increase South America’s profile in the international scene, they prioritized autonomy and had a strong anti-imperialist and Latin Americanist rhetoric that promoted a regional identity (Simonoff 2009; Levitsky and Roberts 2011; Gomes Saraiva 2012; Majdalani 2013; Tokatlian and Merke 2014; Merke and Reynoso 2016). In the CAN, Ecuador was the thematic leader, and it followed similar foreign policy principles (see Zepeda 2011).

This political ideology deeply influenced a re-orientation in the regional policies for the movement of persons in South America. This stage is inaugurated with the milestone in regional migration policies: the Residence Agreement.

6.1 Towards the Convergence of Regional Policies for the Movement of Persons: The Residence Agreement and the South American Conference for Migration

Towards the end of 2002 there was a significant change in regional policies. Mercosur approved a series of agreements that expanded notably the rights of migrants and liberalized their movement within the region. These agreements were regionalized to the rest of South America.Footnote 15 The most significant and well-known one is the Residence Agreement for Nationals of the Member States of Mercosur and Associated StatesFootnote 16 (2002), an international treaty aimed at facilitating the procedure for obtaining a legal residence to all member state nationals. Its main provision is the creation of a simplified process by which regional migrants get a 2-year temporary residence that allows them to get permanent residence after these 2 years.

The Agreement changed the regional migratory agenda by reintroducing the free movement of persons as an ‘essential’ objective of the RI process and by extending its scope to all the persons (not just workers).

The importance of the Agreement relies on several dimensions. First, not only did it change regional policies, but also domestic ones, as it eliminated the ‘categories’ of immigrants defined by national laws (which were generally linked to the persons’ economic activity), for South American nationals. It thus established the ‘nationality criteria’ as the main requirement for obtaining a legal residence and it eliminated the requirements linked to the socio-labor condition of the migrant. Second, the Agreement aims to prevent South Americans from having an irregular migratory status in the region. It does so by eliminating sanctions and penalty fees that migrants might have to pay in case they have an irregular status and want to change it (art. 3). The documents that are required for the process are simplified and harmonized. It also established meaningful rights such as equal treatment, family reunion and the right to transfer remittances (see arts. 8 and 9).

This milestone policy was formulated in the Office of International Relations of the National Direction of Migration (NDM) of Argentina, the main receiving country in the region. Argentine officials based their proposal on the country’s knowledge and historic experience in migration management. The underlying ideas were, first, that the reason for the large number of irregular immigrants residing in the region was national legislation that was ‘detached from reality’. Also, from this perspective, migration flows would keep entering the country (official of the government of Argentina, interview, 9 October 2014). The approval, regionalization and institutionalization of this regime was due to two main factors. First, Brazil, the biggest country in the region and one of the major economies in the world, supported and promoted the approval of the agreement (official of the government of Argentina, interview, 9 October 2014; Acts 1/02 and 2/02 in Mercosur 2015, pp. 22, 28). Second, both Brazilian and Argentine officials had a friendly relationship and shared ideas and experience in migration and human rights issues, so they worked jointly to convince the rest of their partners to take part in this regime (official of the government of Argentina, interview, 9 October 2014; Acts 1/06, 4/06 and 6/06 in Mercosur 2015). The rest of the regional partners accepted the Argentine proposal because they considered that it was ‘superior’ (Alfonso 2012, p. 51) to previous regional legislation.

At the same time that the Residence Agreement was adopted, the South American Conference for Migration (SACM) was created. The SACM is one of the fourteen consultative regional processes that exist worldwide. It holds annual meetings since its creation in the year 2000.Footnote 17 It currently has twelve members.Footnote 18 Decisions are taken by consensus and its declarations are non-binding. In spite of this, it influences regional policymaking. It also collaborates with the Mercosur and the CAN and fosters convergence between these two via the exchange of experiences and ideas.

The SACM and the Residence Agreement were the two main factors that started a new stage in the making of regional policies for the movement of persons, as they involved most of South America, and incorporated new principles and ideas, which were promoted when the ideologies, political orientation and RI model changed.

In 2004, the Declaration of Santiago about Migration Principles was signed by Mercosur member states, plus Bolivia, Chile and Peru. It stressed the importance of multilateralism for the management of migration, human rights, equal treatment, family reunion and the social and economic contributions of migrants to the receiving states. In its declarations, the SACM has reproduced and even amplified these principles. The SACM has recognized the ‘right to free movement of the migrant person’ and defined it as a regional objective in the South American Plan for Human Development of Migration (see infra.) (SACM 2010) and the Declaration of Migration Principles of the SACM of 2010.

In the SACM, the Ministries of Foreign Affairs are the main actors and decision-makers. This explains the importance that the Declarations give to multilateralism and respect for international law and human rights treaties. It also explains the regionalization of common criteria on migration that led to the adoption of a unified regional position on the subject that was sustained by all South American countries in international forums and in the negotiations for the Global Compact for Migration. The Latin American common position for the GCM (see ECLAC 2017) was centered on the respect of migrants’ human rights, access to justice for migrants, non-criminalization of irregular migration, the elimination of detention centers and migrants’ contributions to host societies and economies. During the negotiation phase, Latin America called for the global recognition of the ‘right to migrate’ and the ‘freedom of movement’ (Milesi 2017; official of the government of Spain, interview, 22nd May 2018), for the universality of migrants’ rights and, instead of expulsions, for migrant regularization as a solution to irregularity (SACM 2017). Some countries even called for the GCM to be a binding agreement.Footnote 19 This human rights and open border position was regarded as South America’s ‘international projection’ (SACM 2018). This progressive and high-profile stance towards migration was adopted by most of the countries in the region and has led to policy and rhetoric convergence. It built from two characteristics of ‘post hegemonic/post neoliberal’ regionalism: first, opposition to American hegemony, materialized in a Latin Americanist rhetoric and the rejection of American and European ‘restrictive’ policies (Brumat and Acosta 2019). Second, Argentina and Brazil’s foreign policy orientation that sought to increase South America’s positioning in the international scene.

6.2 The Andean Community in the 2000s

The Cartagena Act (CAN 1999) introduced the completion of the Andean Common Market as an objective for 2005 (which has not been accomplished yet). Consequently, the free movement of persons became a regional objective with a specific deadline.

The Andean Charter for the Promotion and Protection of Human Rights of 2002, a nonbinding agreement, was the first regional norm that seeks to acknowledge migrants in a more comprehensive manner, and not only in relation to the economic or work activity that they perform. The adoption of the Andean Charter is closely related to the participation of the member states in the SACM (Stang 2009; Official of the CAN, Interview, 12 January 2012). It defines migrants’ basic rights and it includes the free movement and transit of migrants and their families as a priority (art. 51) regardless of their economic situation.

In 2003, the Andean Instrument for Labor Migration (Decision 116) was reformulated (CAN 2003a, Decision 545). Its objective is to achieve the free movement of ‘Andean migrant workers’ (art. 20) as a way to ‘progressively’ achieve the free movement of persons. This Instrument is less selective than the previous one, as it eliminates the categories of ‘qualified’ and ‘undocumented’ worker. It also adds new rights such as family reunification and the free movement for the workers and their families (art. 12). Another disposition that reduced the selectivity of the Instrument is the modification of the Labor Migration Offices’ responsibilities, as they will no longer have the capacity of selecting workers. These Offices also have an informative function, as they have to provide information to migrant workers (art. 17), changing their selective approach and adopting an inclusive one instead.

That same year (2003) the Decision 583 (CAN 2003b) modified the Andean Instrument for Social Security of 1977. The new Instrument goes beyond its predecessor, as it recognizes the same social security rights and obligations for any Andean resident worker (not only Andean nationals), apart from equal treatment, non-discrimination, health services and state welfare (art. 1).

The Andean Forum of Migration was created in 2008 following a proposal from Ecuador. In this occasion, Ecuador also presented a first draft of an Andean Plan for Human Development for Migration, whose name was later changed to Andean Migratory Statute. This was part of Ecuador’s strategy to ‘andinize’ its (very progressive) domestic migration policy (Official of the CAN, Interview, 2016)Footnote 20 and become a regional leader in migration issues.

The Andean Migratory Statute would be a binding instrument that seeks to be the main orientation for regional policies for the CAN and Chile (Andean Parliament 2015). It defines the freedom of movement and residence as rights, calls for a regional citizenship that includes political rights and for the ‘regional harmonization of national norms’ to guarantee the rights of migrants persons. It also promotes the ‘socialization’ of the Statute with the Mercosur, ‘to advance in regional integration’ (art. 73). The Statute would codify and deepen Andean migration Law. It was approved by the Andean Parliament in 2015, but its final approval by the Council of Ministers of Foreign Affairs, that would enforce it, is still pending.

6.3 Mercosur During the Postneoliberal-Posthegemonic Era

A turning point in Mercosur’s migration agenda was the creation of the Mercosur’s Specialized Migration Forum (FEM, for its initials in Spanish) in 2003. The FEM ‘unified’ the migration agenda and separated it from labor and security. From that moment on, Mercosur’s migration agenda was centered on updating old legal instruments, on border control and on the formulation of ‘action plans’.

FEM’s initial activities aimed at reducing the fragmentation of regional legislation on migration and at further harmonizing border control instrumentsFootnote 21 that were ‘inherited’ from the nineties.

In late 2010 the Plan of Action for the Statute of the Mercosur Citizenship (CMC 2010) was approved. It was largely the result of two factors: first, the impulse of the Lula administration that, in its last months, tried to get a relevant regional agreement that marked a milestone in regional integration approved under its Pro Tempore Presidency (official of the government of Brazil, interview, 11 August 2015), as a way of showing regional leadership. Second, the convergence of the governments’ interests on social questions in the postneoliberal/posthegemonic stage.

The Plan of Action for the Mercosur citizenship is supposed to be completely implemented by 2021 (for Mercosur’s 30th anniversary) (art. 7). This Statute does not create any new rights or institutions; it is a compilation of norms that simplifies and speeds up some procedures for Mercosur nationals. As part of the Plan, the Argentine government made some far-reaching proposals, for instance: a Unique Migration AgreementFootnote 22 (Act 4/12 in Mercosur 2015, p. 122) and the modification for the Recife Agreement—an Agreement which would have left migratory border controls to only one country (the receiving one). But countries did not agree on any of these proposals. Most of the policies that were adopted in these last years were centered, again, on the facilitation border crossings,Footnote 23 similar to the nineties.

7 Final Considerations: A New Generation?

The first generation of regional policies was highly influenced by the regional context, in which the RI model’s main objective was to achieve economic development through industrialization and intra-regional trade liberalization. The first regional policies for the movement of persons seek to regionalize certain standards for labor mobility (such as equal treatment and non-discrimination) and showed a preference for high skilled migrants. It also established the ‘nationality criteria’ for the first time. Due to their continuous reference to national legislation and the only presence of state actors in policymaking processes (particularly, the Ministries of Interior and Labor of the CAN) they never turned into a ‘real’ regional policy.

The second generation of norms was shaped by the neoliberal ideology that characterized regionalism in South America during the nineties. Open regionalism centered regional policies on economic and commercial issues, leaving the social aspects of RI behind. ‘Social’ integration had to be in line with the level of economic integration. This explains the fact that Mercosur retroceded in its original objective of achieving the free movement of persons as a way of consolidating a CM, which never happened, also explains the central role of easing border crossings (to facilitate trade) and the fact that the greatest advances in social-labor aspects were the result of pressure from a new regional actor, workers’ unions. The unions, which were organized transnationally, sought to influence decision-making in Mercosur as a way of ‘securing’ workers’ rights at the regional level, because they had lost influence towards national governments.

Second generation policies were focused on ‘technical’ aspects, such as border management, documents to enter and exit states’ territories and the simplification of bureaucratic procedures for human mobility. These apparently ‘technical’ measures had a double political objective: they worked as an instrument for economic integration and liberalization (Pellerin 1999b) while giving an initial push to the promotion of a sense of regional identity among the population, that would help to sustain the existence and development of ROs (Gómez-Mera 2005; Malamud 2010). At the same time, the adoption of regional procedures for regional mobility promoted communication among regional bureaucrats who knew each other better, and developed a sentiment of regional identity (Gómez-Mera 2005). Increased interaction proved to be an important incentive for policy convergence in the following decade.

After the 2001–2002 crisis and the failure of the neoliberal model, the regional migratory agenda underwent a deep change. The Residence Agreement reintroduced the free movement of persons in Mercosur’s agenda which was afterwards adopted by all the CAN members. The Residence Agreement is a binding norm that expands the rights of the citizens of the signatory states and modifies domestic migration policies, as it eliminates the migration categories for South Americans. This legislation eliminates the economic condition of the migrant person as a requirement for residence for a 2-year period. The Agreement was signed before the ‘neo populist/progressive’ governments took office, which shows that public officials already shared ideas related to the expansion of migrants’ rights and exchanged them. The posthegemonic/postneoliberal RI model that prevailed in the region, added to the relevance that Argentina, Brazil and Ecuador gave to relations with other South American countries helped to consolidate these ideas, gave them visibility, and also helped in implementing the regime and making policies converge.

During this period, migrant’s rights were expanded, border control norms were reformulated, the selectivity of the policies was reduced, and the migration agenda was separated from the security and labor agendas in regional institutions. Migration eventually became a central agenda that helped to reinvigorate RI after the crisis in Mercosur and promoted policy convergence between Mercosur and the CAN.

The third generation of norms can be divided into two groups. The first one is comprised of binding norms that regulate mainly labor migration and border control, most of which were ‘inherited’ from the previous generation. The second group established and expanded migrants’ rights and liberalized the residence of South American citizens in the area. It also incorporated, regionalized and promoted the adoption of international principles and basic rights. In this second group there is only one binding norm, the Residence Agreement. The rest of this legislation is non-binding: the SACM Declarations, the Declaration of Santiago, Mercosur’s Citizenship Plan of Action and the Andean Charter for the Promotion and Protection of Human Rights. These non-binding norms are the most progressive ones, and they even promote the free movement of persons as a right. The SACM played a crucial role in intensifying interaction between national officials, so it promoted the adoption of these norms and shared ideas and values on the agenda.

The third generation of norms has a dual character: on the one side it expands and promotes migrants’ rights and on the other, it improves border controls. This dual character is the result of three factors: 1- Argentina’s ‘thematic’ leadership. Argentina was recognized as a legitimate leader on migration issues by her regional partners, who accepted her proposals, mostly aimed at regularizing her irregular immigrant population; 2- foreign policy strategies of the leading countries (Argentina, Brazil and Ecuador) that prioritized relations with the region; 3- the ideologies of the governments in power in that moment, which had a strong focus on human rights and thus opposed the US and the EU’s ‘restrictive’ policies. Mercosur was the leading RO because the two countries with most aggregate power are part of it, and not the CAN. Also, the influence of certain actors changed: worker unions were more focused on the domestic scene because they had (re)gained access to national governments, who heard their demands.

The renewed interest in border and migration control is explained by 1- the greater economic, political and social state presence during this stage and 2- Argentina’s ‘thematic’ leadership. Argentina’s leadership conferred a new character to regional policies that was in line with its own migration policy: more open and oriented towards South America, while increasing control of who entered and resided in her territory. As Argentina is the main receiving country, the creation of a more transparent and predictable regional migratory regime was a central interest to her. The regime is more transparent because more information was shared between partners and is more predictable because the rules are clear, and the costs of managing the regime are more equally shared. For Brazil migration is not a central topic, as it mostly a sending country. As the country with most aggregate power, Brazil’s support of Argentina’s proposals was crucial for the approval and institutionalization of the regime.

In the Andean Community, Ecuador, who had recently adopted a very progressive migration policy (see Ramírez 2013, 2016) and whose foreign policy prioritized relations with South America, also acted as a subregional leader in this agenda. So, she proposed the adoption of the Andean Plan of Human Development for Migration which included principles that had already been embraced at the South American level, promoting policy convergence.

The political landscape in South America started changing in 2015, when Mauricio Macri, a conservative, took office in Argentina. In 2016, in Brazil, Dilma Rousseff, a neo-populist progressive, was impeached and Michel Temer, another conservative, replaced her. In January 2019, Jair Bolsonaro, a far-right politician and former military officer, became Brazil’s President. The ‘turn to the right’ of two of the most influential countries and largest economies in the region marked the start of a new stage in South American politics. These governments started to debate the possibility of reducing the Mercosur to a free trade area, which contradicts the Treaty of Asunción (Frenkel 2016, p. 21). This suggests that we could be facing the beginning of a new stage in South American RI and consequently, in the regional migration agenda.

7.1 Return to Uniaxial Integration? Possible Scenarios Facing the Return of Neoliberalism

As Comini and Tussie (2016, p. 14) point out, the leading RO, Mercosur, is facing a moment of ‘internal friction’. Its two main partners and main actors in the regional migration agenda support a RI model similar to the one that prevailed during the nineties: open regionalism, with more flexible legislation. They also reject regional models based on the postneoliberal/posthegemonic, ‘multiaxial’ projects, which they characterize as ‘ideological’ and closed to the global economy (Comini and Tussie 2016, p. 13). They also seek to rebuild strategic relations with Europe and the US (Frenkel and Azzi 2018). This could lead them to the securitization of migration policies, to satisfy European and American demands and as an imitation process (see Brumat and Acosta 2019). Also, as a way of ‘des-ideologizing’ the RI process, they could go back to adopting ‘technical’ policies. Furthermore, Brazil and Chile have recently pulled out of the GCM, breaking the regional position that had been constructed during the last two decades (see Brumat 2019).

In spite of this, there is still some continuity with the postneoliberal/posthegemonic project. So, in Argentina’s foreign policy strategy of ‘concentric circles’, the first circle is still Latin America (Comini and Tussie 2016). Argentina is pushing for an approach between Mercosur and the Pacific Alliance.Footnote 24 This new RI project comprises some CAN member states and promotes a RI model based on open regionalism and free trade. The (very few) policies that the Pacific Alliance has adopted on migration are ‘technical’ and limited to making work and tourist visas slightly more flexible. This, together with the fact that the CAN has lost some of its membersFootnote 25 and that some aspects of its migration agenda are stalled, could lead to CAN being after the PA in the order of Mercosur’s priorities.

At the same time, and due to the high level of institutionalization and domestic internalization of some regional policies, and the benefits that it generates for many statesFootnote 26 such as the Residence Agreement, second and third generation policies would be difficult –but not impossible- to eliminate.

Will all these recent developments and the prevailing ideology of the new governments lead to a change in regional migration policy? To answer this, we need first to point out that there have not yet been any new significant regional policies regarding migration. Most of the recent changes were taken at the domestic level.Footnote 27 Second, I suggest four possible scenarios for the short and medium term in the South American migration agenda:

First scenario: the second and third generation of regional migration policies will be kept + domestic restrictive measures that focus on security will be increased, in line with the ones adopted in Europe and the US. This domestic legislation would contradict regional policies. As the countries with most aggregate power (Argentina and Brazil) and other regional mid-powers, such as Chile, prioritize their relations with extra-regional (global) powers, such as the US and the EU, they will tend to adopt measures that favor these countries, giving secondary importance to South America. In the absence of a regional Court that enforces regional legislation, the contradiction between national and regional policies would have to be solved with direct negotiations and coordination in regional institutions such as the Mercosur, the CAN or the SACM.

Second scenario: the second and third generation regional migration policies will be kept + domestic restrictive measures that focus on security will be increased + worker unions will bring their focus back to the regional level as they cannot get their demands heard at the national level, pushing for the consolidation of rights for regional workers. Same to scenario one, but with the union (re)emerging as an influential actor. As happened in the neoliberal years, if worker unions do not get responses to their demands, they could return to regional institutions. At the same time, and in line with the neoliberal model of integration that prevailed during the 90s, securitist actors could (re)gain power, pushing for increasing securitist measures at the national level (Brumat et al. 2018).

Third scenario: second and third-generation regional migration policies will be kept + a return to the adoption of measures to favor the movement of highly qualified personnel and for the execution of business activities will happen + adoption of policies that could expand migrants’ rights will be suspended. As neoliberal governments tend to have closer relations with employers (see Pellerin 1999b), they could hear their demands and prioritize the migrants in highly qualified employment. At the same time, as human rights pressure groups lose influence within the states and the current governments are less likely to hear their demands, the rights-based approach could be ‘frozen’ (see Brumat 2019).

Fourth scenario: the most unlikely scenario is (partial) disintegration. If the sceptic and anti-integration constituencies within some of the most powerful governments (such as Brazil’s Minister of Economy) gain more power, they could push for a weakening of regional institutions and with them, regional migration policies such as the Residence Agreement could stop being used or even revoked.

Regional migration policies have proven to be a meeting point for the construction of shared positions, common ideas and values for South Americans. South American regionalism is in a transition moment, and the survival of the agenda proposed by the postneoliberal/posthegemonic regionalism is being put into question. In this scenario, the analysis of the evolution (or involution) of the policies developed in the regional social agenda will become crucial.