“I’m giving up on Brazil”: shifting migration aspirations and capabilities in the face of simultaneous crises

“Estou desistindo do Brasil”: mudanças nas aspirações e capacidades migratórias diante de crises simultâneas

«Me doy por vencida con Brasil»: cambios en las aspiraciones y capacidades migratorias frente a crisis simultáneas

Abstract

Based on the aspirations-capabilities framework, this paper analyses how the overlap of three shocks - a national socio-political crisis, an economic downturn, and a socio-environmental disaster - influenced households' international migration aspirations, decisions and strategies. The study uses a mixed methods approach, including a statistically representative survey and eighteen semi-structured interviews conducted with households with and without international migration experience in Governador Valadares, one of the main emigration hotspots in Brazil. It argues that, despite similar representations of international migration across socioeconomic strata, people’s aspirations and capabilities to migrate in the context of multidimensional crises vary across class positions. These external shocks shaped migration aspirations not only instrumentally, diminishing access to financial and natural resources, but also symbolically, through feelings of dismay. The working poor report living in a ‘permanent state of crisis’ and see migration as a tool for improving living conditions. For the lower middle classes, these crises have rendered visible the mismatch between their (augmented) expectations, built during the early 2000s era of optimism, and their effective chances of social mobility. Migration, therefore, appears as a solution for accessing services they could once afford. For the upper-middle classes, these crises represented a rupture, with migration now being considered as a way of maintaining their quality of life and class position.

Keywords:
Aspirations; Capabilities; Crises; Brazil; International migration

Resumo

Com base na teoria de aspirações-capacidades, este artigo analisa como a sobreposição de três choques - uma crise sociopolítica nacional, uma recessão econômica e um desastre socioambiental - influenciou as aspirações, decisões e estratégias de migração internacional das famílias. O estudo utiliza uma abordagem de métodos mistos, incluindo um survey estatisticamente representativo e 18 entrevistas semiestruturadas com famílias com e sem experiência migratória internacional em Governador Valadares, um dos principais epicentros de emigração no Brasil. Argumenta-se que, apesar das representações similares da migração internacional entre os estratos socioeconômicos, as aspirações e capacidades das pessoas de migrar no contexto de crises multidimensionais variam de acordo com sua posição de classe. Esses choques externos moldaram as aspirações de migração não apenas instrumentalmente, diminuindo o acesso a recursos financeiros e naturais, mas também simbolicamente, por meio de sentimentos de desânimo. Enquanto os trabalhadores pobres relatam viver em um “estado permanente de crise” e veem a emigração como uma ferramenta para melhorar suas condições de vida, para as classes médias baixas essas crises tornaram visível o descompasso entre suas (altas) expectativas, construídas durante a era de otimismo do início dos anos 2000, e suas reais chances de mobilidade social. A migração, portanto, aparece como uma solução para acessar serviços que outrora podiam pagar. Para as classes médias altas, essas crises representaram uma ruptura, com a emigração agora sendo considerada uma forma de manter sua qualidade de vida e posição de classe.

Palavras-chave:
Aspirações; Capacidades; Crises; Brasil; Migração internacional

Resumen

Basado en la teoría de aspiraciones-capacidades, este artículo analiza cómo la superposición de tres choques -una crisis sociopolítica nacional, una recesión económica y un desastre socioambiental- influyó en las aspiraciones, decisiones y estrategias de migración internacional de los hogares. El estudio emplea un enfoque de métodos mixtos que incluye una encuesta estadísticamente representativa y dieciocho entrevistas semiestructuradas con hogares con y sin experiencia migratoria internacional en Governador Valadares, uno de los principales epicentros de emigración de Brasil. Se argumenta que, a pesar de que las representaciones sobre migración internacional son similares entre los estratos socioeconómicos, las aspiraciones y capacidades de las personas para migrar en el contexto de crisis multidimensionales varían según su posición de clase. Estos choques externos moldearon las aspiraciones migratorias no solo de manera instrumental, disminuyendo el acceso a recursos financieros y naturales, sino también simbólicamente, a través de sentimientos de desánimo. Mientras los trabajadores pobres reportan vivir en un «estado permanente de crisis» y ven a la emigración como una herramienta para mejorar sus condiciones de vida, para las clases medias bajas, estas crisis han hecho visible la discrepancia entre sus (altas) expectativas, construidas durante la era de optimismo de principios del siglo XXI, y sus posibilidades efectivas de movilidad social. La migración, por lo tanto, aparece como una solución para acceder a servicios que alguna vez pudieron costear. Para las clases medias altas, estas crisis representaron una ruptura, considerándose ahora la emigración como una forma de mantener su calidad de vida y posición de clase.

Palabras clave:
Aspiraciones; Capacidades; Crisis; Brasil; Migración internacional

Introduction

In contrast with the nuanceless precepts of traditional push-pull models well documented in the international literature (De Haas, 2011; Van Hear, 2018), the aspirations-capabilities framework has proven effective in recognising the broader societal dynamics influencing migration trends and the lived experiences of migrants and non-migrants (De Haas, 2021). Aspirations refer to the belief that there are better opportunities 'there' than 'here', while capabilities refer to the factors that shape one's actual ability to move, including economic, social, and cultural resources (Carling; Schewel, 2018; De Haas, 2021). Concurrently, other drivers interact with specific demographic, social, economic, and political factors, shaping both aspirations and capabilities and impacting mobility at varying scales (Black et al., 2011; Carr, 2005; Van Hear et al., 2018). External shocks, such as environmental and technological disasters and epidemiological crises, exemplify forces mediating the structural-compositional links shaping migration.

Since Brazil joined the global international migration architecture in the 1980s (Martes, 1999), the Governador Valadares microregion (GV) has been one of the country’s main emigration hotspots, with long-established diasporas in the United States and Portugal (Jorgensen et al., 2019; Zapata, 2019). Based on the aspirations-capabilities framework, this paper analyses how the overlap of three shocks, a national socio-political crisis, an economic downturn, and a regional socio-environmental disaster¹ - the SAMARCO dam failure - influenced households' migration aspirations, decisions and strategies. To that end, we use an inductive mixed-methods approach, combining original survey data and eighteen semi-structured interviews conducted between 2015 and 2018 with households with and without international migration experience in Governador Valadares, Brazil.

Although previous studies have analysed migration strategies using this framework (Aslany et al., 2021; Carling; Collins, 2018; Carling; Schewel, 2018; De Haas, 2021), its usefulness in the Brazilian context remains under-researched. This case study provides a unique opportunity to explore people's aspirations and capabilities surrounding international migration in the context of multidimensional crises. We argue that these external shocks shaped migration aspirations not only instrumentally, by diminishing access to financial and natural resources, but also symbolically, through feelings of dismay. The meaning and impact of these overlapping crises vary according to households’ socioeconomic standing. While for the working poor, who report living in a “permanent state of crisis”, emigration remains a tool to improve living conditions, for the lower middle classes, these crises rendered visible the mismatch between their (augmented) expectations and chances of social mobility, with emigration emerging as an option to deal with deteriorating living conditions. For the upper-middle classes, these crises represented a rupture that threatened their future class position, with emigration now considered a way of maintaining their quality of life. Thus, the paper makes two significant contributions. First, the novel quantitative research design provides evidence of the stability of a collectively constructed entrenched culture of migration in a long-established emigration hotspot in Brazil. Second, the qualitative data adds a more nuanced understanding of how people deal with multiple crises and how these impact aspirations and capabilities to migrate, according to socioeconomic status, in a context where international migration is a “total social fact” (Sousa; Fazito, 2017).

The article is structured as follows: The first section delves into the aspirations-capabilities framework. The second section provides an overview of the case study, while the third explains the mixed methods approach to data collection. The article concludes with a discussion on how households' migration aspirations, capabilities and strategies were affected by the three overlapping crises, according to their Socioeconomic Strata (SES).

Aspirations, capabilities, and drivers: a review

For years, migration research has grappled with the question of why some people move while others stay put (e.g. Van Hear et al., 2018). Whilst most contemporary literature has challenged the basic assumptions of the neoclassical framework (De Haas, 2011; Van Hear et al., 2018), it was not until recently that migration scholars were able to answer this question using a more nuanced theoretical rationale. The aspirations and capabilities framework has been used to identify broader social processes shaping migration patterns and experiences (De Haas, 2021). Aspirations can simply be defined as the 'desire', or conviction that migrating is better than staying put, or vice versa (De Haas, 2021). Migration and staying aspirations are embedded in broader life aspirations and closely related to people's ideas of a 'good life' (De Haas, 2021; Carling; Schewel, 2018). They are also connected to individual and collective views of the past and expectations for the future in different places (Vezzoli, 2023; Schewel, 2015). On the other hand, capabilities encompass all the factors that shape one's effective ability to move, including economic, social, and cultural resources. The disjuncture or correspondence between migration aspirations and capabilities has, in turn, motivated migration scholars to elaborate a broad typology of individual inclinations to migrate or stay (such as voluntary and involuntary mobility), which has helped move beyond traditional narratives of migration as intrinsically good or bad (Carling; Schewel, 2018).

In outlining the contours of an aspirations-capabilities framework applied to mobility, de Haas (2021) argues that aspirations to move can be broadly divided into two dimensions, 'instrumental' and 'intrinsic'. Most neoclassical theories have focused exclusively on the instrumental dimension of mobility, i.e. migration as a means to achieve a better status, higher income, or adequate healthcare or security. Alternatively, recent scholarly work has emphasised that migration is not only an instrument towards an end but also an end in itself (De Haas, 2021; Carling; Schewel, 2018). This reasoning suggests that moving can be perceived as having an intrinsic, symbolic value. Intrinsic aspirations to migrate identified in the literature include the search for adventure, knowledge, and different life experiences. In places experiencing significant population exodus, migration is often seen as a key transition phase into adulthood (Kandel; Massey, 2002), and its symbolic value often transcends and overshadows typical cost-benefit calculations (Ali, 2007). As migration prevalence increases in a particular society, it also shapes its social representation as either positive or negative, which, in turn, feeds back into individual aspirations (De Rosa et al., 2021; Kadianaki; Andreouli, 2017).

Therefore, the literature on migration aspirations has become closely tied to discussions around the culture of migration and social networks, especially as scholars have realised that the perpetuation of movement could not be fully attributed to exogenous socioeconomic factors (Bachrach, 2014; Massey et al., 1993). Key here is the idea that enduring migration practices condition the repertoire of possibilities available to people from a particular region, while simultaneously facilitating further movement through the establishment of social networks (Grubanov-Boskovic et al., 2022; Vargas-Valle, 2021; Horváth, 2008; Massey et al., 1993). Although migration research rarely draws explicitly on Bourdieusian theory, this reasoning is compatible with the concepts of capital and habitus: collective experiences of migration constitute a certain habitus that predisposes people to perceive migration as a viable and desired alternative, while community and family networks and brokers help aspiring-migrants to acquire the necessary capital - economic, social and cultural - to move (Kim, 2018). Continued mobility can, thus, change perceived costs and benefits associated with other economic activities, for example, by discouraging people to invest in the local market or education, while ascribing symbolic value to staying and going (Horváth, 2008; Kandel; Massey, 2002). Even though the intrinsic and instrumental dimensions of migration aspirations can be separated for certain analytical purposes, it is important to recognise that they are also intimately intertwined.

However, not all aspirations to move, whether instrumental or intrinsic, can be readily converted into movement. Carling and Schewel (2018) define ability as the capacity to convert the wish to move into reality. Capability, on the other hand, is a broader concept that refers to “the ability of human beings to lead the kind of lives they have reason to value and to enhance the substantive choices (freedoms) they have” (Sen, 1999). As Bauman (2021) reminds us, in the postmodern era, it is common for people to express their ambitions and conception of well-being in terms of mobility, being able to choose their place of residence freely: “liberty means above all, being able to choose, and that has acquired a conspicuous spatial component” (p. 158). Notably, capabilities depend both on positive liberties, such as access to enough material resources to start a journey, and negative liberties, such as the absence of governmental barriers to movement (De Haas, 2021; Carling; Schewel, 2018). Once a person aspires to migrate, their capabilities define not only whether and when the movement will be possible, but also what migration pathways are available (De Haas, 2021). As Singer (1973), rightly pointed out from a historical-structuralist perspective, in the context of the spatial imbalances created by industrial development in Brazil, (rural-urban) migration had to be understood as a social process that was positive for the social mobility of individuals. In addition, migration was conditioned not only by individuals' social standing but also by economic push and pull factors provoked by capitalist development, as well as - or perhaps more importantly - by pre-established social and solidarity networks.

To what extent are aspirations and capabilities to move connected? Here it is helpful to draw on Appadurai's idea of the “capacity to aspire” (2004): the more resources and power one has, the more likely it is to understand, through practice, the links between desires and concrete possibilities. This idea can also be linked back to Bourdieu (1997), who noted that people, far from wanting it all, tend to adjust their desires to their chances - a process learned through continued experience or habitus. In migration research, this translates into the acknowledgement that higher levels of development, far from anchoring people to their homeland, tend to increase migration aspirations. As people come into contact with different cultures and modes of consumerism, they not only 'learn' to desire things they cannot find at home, but also 'learn' that these things can be within their reach and how they can effectively use practical knowledge to migrate.

Importantly, migration aspirations and capabilities are far from static, as they may be altered as a result of structural or contingent changes. We understand crises as drivers of migration, i.e. exogenous factors that make certain migration paths more likely than others, by affecting both desires and objective chances to migrate (Carling; Francis, 2018). Yet, echoing Van Hear and others’ (2018) notion of 'complex migration drivers', this paper reaffirms the idea that movement is rarely caused by one single driver, showing the ways in which environmental, economic, and socio-political determinants at different scales and durations are intimately entangled in Brazil.

While it may be easy to cast 'aspirations' within the realm of agency and 'capabilities' to the sphere of structure, it is crucial to recognize that both aspirations and capabilities are concurrently shaped by individual sociodemographic characteristics and the macro environment. Although movement is constrained by a series of structural factors, individuals are also able to navigate, circumvent, and change such systems of capital and resource distribution. Similarly, people are not simply born with particular desires, as these are also partly determined by broader social, cultural, and political elements, including community and family experiences (Carling; Schewel, 2018).

In this regard, emigrants' continued transnational socio-economic engagement with their families from afar has produced an array of social, economic, political and cultural transformations at the family and community levels in origin countries (Glick Schiller et al., 1992; Guarnizo; Smith, 1998; Portes, 2003). In turn, transnational migration and the social spaces that emerge from it shape - and are shaped by - the changing conditions of global capitalism and must therefore be analysed in the context of global capital-labour relations (Levitt, 2001; Portes et al., 1999; Sassen, 1988; Vertovec, 2009). In this respect, Guarnizo and Smith (1998, p. 6) have further argued that transnational practices “are embedded [in] and they sometimes even perpetuate enduring asymmetries of domination, inequality, racism, sexism, class conflict, and uneven development”. In addition, rising levels of migration and remittances (from/to the south) have effectively rendered migrant workers as key agents of the global social architecture, facilitating the exchange of money, ideas, and subjectivities within and between the Global South and North (Zapata, 2013). Thus, international migration cannot be dissociated from broader social change and development processes, as it is part and parcel of the international economic order and the (neoliberal) global mobility of capital (Castles, 2010, 2011; Sassen, 1988).

Since the 1990s, international migration has become a structural aspect of Brazilian society. The driving forces behind these population movements are inscribed in global processes of capital and labor restructuring, including recurring financial crises, slow socioeconomic development, an increasing surplus of labour as the country progresses through its demographic transition, and a lack of opportunities for social mobility, among others (Brito, 1995; Sales, 1999; Patarra, 2005). This paper examines, precisely, how families' socioeconomic background intersects with these broader structural factors in shaping aspirations and capabilities to move in Governador Valadares, Brazil.

Background to the case study

Governador Valadares (GV) is the largest municipality in the Brazilian Vale do Rio Doce mesoregion, and often regarded as a unique ‘social laboratory’ for the study of international migration in the country (Gomes; Guedes, 2020). With a population of 257,172 people in 2022, approximately 96 percent of whom live in urban areas (IBGE, 2023), GV is a key international emigration hotspot, with most families residing in the municipality directly or indirectly affected by migration (Soares, 2002; Zapata, 2019).

In 2015, GV and the Vale do Rio Doce region were hit by one of the most severe environmental disasters in Brazil’s history. The failure of a dam built by the SAMARCO mining company - owned by BHP and Vale - released 43.7 million cubic metres of iron ore mine tailings into the Rio Doce (Silva et al., 2021). This catastrophe caused the forced displacement and loss of livelihood of hundreds of people, as well as the pollution of the region's main water sources (Andrade et al., 2021; Dias et al., 2018). The disaster occurred while Brazil was undergoing a political-institutional crisis and grappling with the effects of the global financial crisis,² deteriorating the region's economic prospects even further.

The political-institutional crisis, which culminated in the impeachment of president-elect Dilma Rousseff in August 2016, began with corruption allegations involving Petrobras - the Brazilian oil giant - and the Workers’ Party.³ The trials, arrests, and impeachment divided the country’s public opinion and fueled a general mistrust in public institutions. On a deeper level, some have argued that the political crisis reflected the global economic downturn, which led to the adoption of fiscal austerity policies by Rousseff from 2014 onwards, marking the end of the ‘class conciliatory model’ that had characterised the Workers’ Party administrations since 2003 (Singer; Loureiro, 2016).

Despite this unique regional and national scenario, migration trends in GV have long been shaped by slow-onset socio-environmental dynamics (Gomes; Guedes, 2020; Jorgensen et al., 2019). Until the early the twentieth century, the forest surrounding the Rio Doce was largely unexplored, but decades of extractivist occupation have led to soil depletion and degradation, and the decline of livestock farming. Official data from the late 1980s indicates that the region had low levels of technological development, income, and employment opportunities and suffered from a scarcity of capital investment (Espíndola, 1995). Emigration then served as a safety valve for the Vale do Rio Doce population, who were experiencing increasing economic deprivation. Concurrently, the exploration of mica in the 1960s, led by American companies, contributed to the establishment of social networks between GV and the United States, particularly the metropolitan Boston region (Assis, 1999; Siqueira, 2006; Soares, 1995; 2002).

While the flows initially consisted of highly educated professionals who had been working for American companies and middle-class exchange students, emigration soon became widespread among people from lower socioeconomic backgrounds (Assis, 1999; Fusco, 2005). The solid network established among emigrants, returnees, and brokers contributed not only to lowering the costs of subsequent movements but also to turning migration into a “total social fact”, shaping the lifestyles and aspirations of the region's inhabitants (Soares, 2002; Sousa; Fazito, 2017). Over time, migration in GV became the norm, and eventually associated with success, hard work and adulthood, in contrast to the ‘laziness’ and stagnation associated with staying put (Fazito, 2010). Consequently, a significant portion of the population in the region has international migration experience or belongs to a personal or social network that includes international migrants (Gomes; Guedes, 2020; Sousa; Fazito, 2016).

In the face of more stringent migration policies and controls in the United States since the 1990s, GV has witnessed a process of diversification of destinations, with more families beginning to migrate to Europe, particularly to Portugal, a trend consistent with patterns observed in other South American countries (Zapata, 2019). Unlike Portugal, regular entry into the USA is contingent on obtaining residence or, more commonly, tourist visas. The latter, in turn, depends on the applicant's ability to prove their intention to return to Brazil, which usually involves having a stable job and sufficient income.

High emigration rates have significantly shaped the economy of the Rio Doce region. Studies have shown that the level of remittances sent to the region resulted in the creation of jobs, the establishment of small businesses and a boom in the real estate market (Soares, 1995). However, the financial crisis that hit Europe and the United States in 2008, coupled with anti-immigration policies in the latter, triggered a wave of return migration to Brazil, and to the Rio Doce Valley in particular⁴ (Siqueira, 2006). This development was particularly detrimental to GV's economy, not only because it reduced the availability of remittances, but also because it increased pressure on the region's fragile labor market. To date, there is scarce research on the extent to which these overlapping crises affected migration aspirations and capabilities, including among returnees, in a setting with a long-established migration culture.

Data and research methods⁵

This paper is based on a mixed methods approach. The quantitative data source is the first statistically representative survey designed for and conducted in GV between 2015 and 2016, based on a probabilistic sample of 1,226 urban households, with follow-up interviews conducted in 2018 and 2020.

The survey used a multi-stage sampling approach. In the first stage, neighbourhoods were grouped into clusters based on their spatial proximity and socioeconomic status. Within each cluster, the sample was further stratified by sex and age groups based on the 2012 population projection of GV provided by Freire et al. (2020) for individuals between 18 and 72 years of age. Households were then randomly selected from each stratum. The minimum required sample size was calculated to be 1,069 households, with a 5% significance level and a 3% margin of error for sample proportions. A variance estimate of 0.25 was used to determine this conservative minimum sample size. Due to budget availability, the sample size was increased to 1,226 households. We conducted face-to-face interviews, with cognitively validated questionnaires through survey pilots.

The 2015/2016 survey data was used in two ways: first, to identify the initial households to be qualitatively interviewed from the universe of survey respondents; and second, to model and analyse the collective thinking around international migration among GV urban residents using a network-based cognitive affinity measure (Duarte et al., 2022).

The quantitative approach

GV has long been exposed to a migration corridor to the United States that began with a very selective type of labour migration of professionals who had established ties with American entrepreneurs during the mica exploration in the region in the 1960s/1970s (Espindola, 1998). Since then, and following a series of economic crises in Brazil, a steady flow from GV to the USA produced a growing and more effective migration network and a large inflow of remittances. This, coupled with the influence of returnees in shaping local aspirations through their behaviour, acquired knowledge, and economic success in the local real estate market, fuelled a migration culture that permeated not only those prone to move, but also stayers who saw in the migration project an opportunity for those in the family willing to take the risk (Souza; Fazito, 2017).

Given this long and well-established migration culture in Governador Valadares, we contend that a collective representation of international migration can be traced in the thematic evocations network among GV’s urban residents. Well-designed survey instruments, innovative methodological techniques, and appropriate qualitative scrutiny of survey-based open questions can help visualise how people perceive migration collectively within this cultural environment. We use a social psychology survey instrument to study social representations and transform its content using a recently proposed cognitive affinity coefficient that converts the survey-based responses into a network that allows us to recreate socially diffused norms and ideas. This survey-based social representation is not intended to measure GV's migration culture, but rather to serve as a marker of the city's long established migration culture.

Based on our probabilistic survey, we selected all 1,226 interviewees, who self-identified as the reference person in the household and provided information about themselves and all other household members currently living in the same unit or those who had emigrated. The survey contained questions on living standards, household assets⁶ and human capital characteristics, such as household members' educational attainment.⁷ We combined the weighted sum of all assets reported with the highest level of education attained by household members to create the socioeconomic status classification, from A1 to E, as suggested by Brasil (2014). The social stratification dimension was used to understand how the social representation of migration, based on the procedure described in the following paragraphs, differs across social classes and how these differences may inform underlying differences in experiences, motivations, and aspirations regarding international migration.

We use the affinity coefficient proposed by Duarte et al. (2022), which transforms evocation data from an inductive term (typically applied to the analysis of social representations) into thematic evocation networks. Building on the Free Words Association Technique, FWAT (Abric 1994), the structured survey questionnaire of our probabilistic survey contained a module with the following questions: (1) “When I say ‘International Migration’, tell me the first five words or expressions that come to your mind, without thinking about them”; (2) “Now, I would like you to put them in order of importance, with 1 representing the most important and 5 the least important”; (3) “You mentioned the expression ‘Fill In’ was the most important for you. What does it mean to you?” [Open question]; (4) “Why is it the most important to you?”.

The expression 'International Migration' is the inductive term and works to anchor respondents’ perceived representations to a common object/idea/value, thereby reducing the likelihood of random responses. Because each person can contribute a maximum of five words or expressions (henceforth called evocations), and there is a sample of 1,226 individuals, a total of 6,130 unique evocations were observed. However, socialization and communication among individuals in a group reduce this figure, as they share not only a common dictionary (their native language) but also common meanings and perceptions (Duarte et al., 2022). Moreover, many of these unique evocations are ‘noise’ around an underlying evocation. For instance, individuals using the words 'risky' and 'dangerous' can be said to refer to the same underlying evocation of 'dangerous'.

To reduce data noise, original evocations were coded according to their underlying meaning, applying inductive reasoning (Bradley et al., 2007) when defining preliminary codes to reflect participants’ meanings. However, the categorization of qualitative data is essentially subjective. A series of steps were taken to minimise the researchers’ classification bias. Although 2,436 out of 6,130 unique words or expressions were originally evoked, the standardization (grouping them into common concepts) resulted in 87 categories of meaning. Additionally, the standardization process was conducted separately by each researcher. The resulting classifications were then discussed collectively, and, in the event of disagreement, the research team negotiated a common standardized evocation, often by reviewing answers to the particular question (3).⁸

Once evocations were standardized, they were modelled as a network of meanings in which each respondent contributes to a vector of evocations. These evocations are connected in pairs for the same person, forming a connected network. The conversion of evocation data into a thematic evocation network was based on the wordevok R library available at [site anonymized].

Four different networks were estimated: a single network based on the full sample of standardized evocations and three stratified networks - one for each SES stratum. The SES indicator was based on five class variables (A, B, C, D, E), constructed from a score formed as the sum of the weights for the educational attainment of the head of the household and a list of household assets. Class A represents the highest SES. The variable was further grouped into three levels: high (classes A and B), middle (class C), and low (classes D and E). Network visualization was performed on Gephi 0.9.2. All data analysis was based on results calibrated by the complex design of the survey.

The qualitative approach

The qualitative analysis is based on eighteen semi-structured interviews conducted in GV between June and July 2018. Interviewees were first identified through the survey and selected based on the household’s Socioeconomic Status (SES) and international migration experience.⁹ This initial stratification allowed us to contemplate the diversity of experiences by these two key socio-demographic attributes, with the sample completed through snowballing. Based on the survey, 33.8% of urban residents belonged to households classified as high SES, followed by 53.9% and 12.3% of middle and low SES, respectively. Accordingly, 9.5% of residents lived in households with previous or current international migration experience (i.e., if anyone in the household ever lived overseas, was born overseas, or if there was a household member currently living overseas). Among those with high SES, 12.9% had international migration experience, followed by 8.6% and 4.1% among middle and low SES households, respectively.

The interviews were transcribed and decoded based on the thematic network approach proposed by Attride-Stirling (2001), which comprises the following steps: decoding transcribing interviews, identifying themes, constructing thematic networks, describing and exploring the thematic networks, summarizing networks, and interpreting results. To protect the privacy and identity of interviewees, the names used here are pseudonyms. The thematic networks approach applied to the semi-structured qualitative interviews allowed us to explore the relationship between multiple crises and aspirations/capabilities to migrate.

Results and discussion

Quantitative results: thematic evocations networks in GV

We found that the well-established migration culture in GV produces little heterogeneity in the collective thinking regarding international migration, regardless of respondents' SES (Figure 1). International migration is primarily associated with improving individual and family living conditions and particularly with life improvement, financial return, job search, transnational family, and a desire to migrate/migration aspirations.¹⁰

Breaking down the overall network by a particular attribute is an indirect way to explore whether this attribute affects how individuals collectively represent migration. When the network is broken down by SES, some interesting patterns emerge. Table 1 lists the ten main evocations in the three networks (low, middle, and high SES), which are directly related to Figure 1. Regardless of SES level, economic motivations emerge at the top of the group’s representation of international migration, which reinforces the widespread migration culture in GV. The three main evocations are life improvement, financial return, and job search for all groups, with only minor changes in the order of appearance across the different strata.

FIGURE 1
Thematic evocations network for the social representations of international migration among urban residents of Governador Valadares, Brazil - 2015/2016

TABLE 1
Ten main evocations by socioeconomic status among urban residents of Governador Valadares, Brazil - 2015/2016

Ranking	High	Middle	Low
1	Life improvement	Financial return	Life improvement
2	Financial return	Life improvement	Job search
3	Job search	Job search	Financial return
4	Transnational family	Better opportunities	Change of residence
5	Better opportunities	Lack of opportunities at origin	Suffering
6	Lack of opportunities at origin	Tourism	Migration aspirations
7	Tourism	Transnational family	Transnational family
8	Change of residence	Missing loved ones	No migration aspirations
9	Migration aspirations	Change of residence	Disapproves of migration
10	Missing loved ones	Migration aspirations	Family separation

• Source: Primary survey data.

However, the quantitative thematic networks approach is limited in its ability to fully explain the nuances behind the perceptions around international migration along the socioeconomic ladder. The following section unpacks the results from the qualitative interviews to shed light on these dynamics.

Qualitative results: thematic networks

As detailed previously, research participants were divided into three groups according to their socioeconomic status. Group C, referred to as “the working poor”, consists of households whose members have low levels of education, skills, and income. Most work in labour-intensive occupations, such as rural and construction workers, taxi drivers, cleaners, supermarket cashiers, and retail sales. Of the seven households (comprising 23 residents) interviewed in this group, six had international migration experience, having returned from the USA or Portugal, after leaving Brazil around the year 2000. Group B, referred to as “the lower middle class”, primarily consists of persons with tertiary education. Participants in this group worked as dentists, social workers, teachers, and prison agents. Although all interviewees had extended family members living in other countries, mainly the USA and Portugal, only one person had previous migration experience. This group included six households, comprising 20 residents. Finally, Group A, referred to as “the upper middle class”, included five households with higher incomes, whose members hold tertiary education, comprising 14 residents in total. Four out of five families in this group had small or midsize businesses, with some members complementing their income through occupations such as psychology, law, or higher education. Except for one person, all participants had been to Europe or the USA: some travelled as tourists to visit family members, while others combined tourism with short-term business trips. In three of the households, the family business also had clear transnational linkages, particularly in the import-export sector. Except for one case, all participants in this group had travelled with tourist visas or as nationals of European countries, a topic explored further below.

Is there a crisis? Migration aspirations across the socioeconomic spectrum

For most working poor households, the incentive to migrate is a permanent feature of their lives, as they live in a “permanent state of crisis” (Lara, Group C, EM5). Thus, the global economic downturn - whose effects were acutely felt in Brazil from 2014 onwards (World Bank, 2021) - was just the latest in a series of crises compounded by other shocks, such as the adverse national socio-political context. As one interviewee put it:

We are so used to crises, that it became a common thing for us. It seems that we are always in crisis...the country improves a little, but then it gets bad…we have no purchasing power, we work and work just to survive...So, I mean, the normal thing for us is being in crisis. So, that's what influences everyone wanting to leave (Enzo, Group C, EM1).

Similarly, when asked about how the multiple crises had affected their lives, participants in the lower-middle class stressed that the economic downturn, although acute, could not be understood separately from other long-term, predisposing migration drivers. Recurrent in all interviews was the recognition that the crises only added to older problems, such as the lack of industries and investment in the city. Participants often compared the scenario in GV with that of neighbouring cities that have mining industries, stressing that the reliance on local trade made GV’s economy extremely volatile. In that sense, one participant questioned the use of the word “crisis” and its applicability to the lower middle and working classes: ‘We use the word crisis. There is no crisis. The crisis has always existed. That’s what I think. We’ve always had crises in Brazil […] Working people have always had to run after things’ (Marília, Group B, EM4).

In group B, the main factors shaping interviewees’ desire to leave the country were low-paying jobs coupled with long working hours, escalating inflation, indebtedness, and political and economic uncertainty. Participants often highlighted that the crisis affected their ability to afford services that had become more widely available to lower-middle income households in Brazil during the early 2000s, such as health insurance, private education, better nutrition, and leisure activities (Bomeny, 2011; Salata, 2015; Singer, 2015). Thus, the aspiration to leave the country seems to be partly driven by a mismatch between the expectations accrued during the decades of economic ascent and the financial hardships experienced since the mid-2010s.

While bearing in mind these long-term trends, it is also important to highlight the unique features of the current overlapping crises. In their narratives, participants often referred to the country’s political instability and the corruption scandals that emerged in the 2010s. Interestingly, interviewees highlighted how the economic and political crises are entangled, suggesting that their symbolic and objective dimensions cannot be understood separately. Complementing her previous statement, Marília added that: 'We’ve always had crises […] but before the crisis was not explicit. It is too evident now because of the scenario of corruption that we are living in' (Group B, EM4).

Notably, participants from groups B (lower middle) and C (working poor) revealed that migrants and prospective migrants are often caught in the middle of two crises: one in the USA, stemming from the 2008 recession and the tightening of immigration controls, especially during the Trump administration; and the one in Brazil. Their family members in the USA are more afraid of being deported, and those who have not migrated face more obstacles to applying for tourist visas. Nonetheless, what stands out is the perception that the crisis ‘here’ is worse than the crisis ‘there’: 'They complain, you know, that there are no jobs [in the USA]. But they haven’t come back, what I think is that the crisis there is not as bad as the crisis here' (Marília, Group B, EM4). Another interviewee contends that: 'The crisis here speaks louder. No one is afraid [of the situation in the USA]' (Humberto, Group B, EM11).

Contrary to the idea of a “permanent state of crisis”, participants from higher socioeconomic backgrounds did not report a sense of continuity. Interviewees in this group were more sensitive to the rupture brought by the financial crisis and, particularly, to the impact on their businesses:

I used to have a company that, until April, provided cashier services. We worked as banking correspondents. But the market now is very tight, very bad. I'm selling very, very little […] The market was very good by 2009/2010, when all these social housing programmes began, ‘My house, My Life’ [Minha Casa, Minha Vida]. They remained good until a couple of years ago. Then the market started retracting and I started having higher expenses than revenues, until I had to close the company (Paulo, Group A, EM10).

Interestingly, because of the transnational character of some of their businesses, participants from this group (A) were clearly aware of how the financial crash in the USA was connected to the economic downturn in Brazil, and in GV. Instead of opposing the crises ‘here’ and ‘there’, participants highlighted how the losses of migrants in the USA affected their capacity to send remittances to their families, which negatively impacted the economy as a whole. As one interviewee stated: 'during the 2008 crisis in the USA, a lot of people went through financial hardships and the level of remittances to GV diminished. This affected the investment here, affected supermarkets, affected all sectors' (Christian, Group A, EM18).

Respondents from higher SES households generally remarked having less money for leisure activities, although this did not entail cutting essential expenses such as food, education and health, as was common in the other groups. Yet, this does not mean their aspirations to migrate remained unaffected. Interviewees in this group seemed particularly troubled by the national political turmoil, and by the subjective dimension of the crises, often stating that they were hopeless in relation to the ‘future of Brazil’. They also contended that the lack of trust in institutions, especially public companies, made investments in the country appear too risky, which further fuelled the financial recession. Thus, emigration emerged as a solution to the sense of hopelessness and the country’s loss of credibility:

There is a sense of uncertainty that was transmitted to the entire world in relation to politics in Brazil. Most large companies that were considered serious, were involved in fraud, in all types of malfeasance. This brought losses to Brazil. So, all that leads to this feeling of insecurity, discomfort, and the people who can opt for a better life, will do so […] Here in GV, all mid-size business owners have been taking this decision to search for other options, to look for new paths, due to the lack of investment, public security, and alternatives. Because everyone who invested in Brazil has lost a lot of money (Christian, Group A, EM18).

I’m giving up on Brazil. You can put Jesus, or the Pope as the president of Brazil, and this country will still not work. I don’t believe in this place. This is one of the reasons why I want to leave (Marcos, Group A, EM2).

In short, the perception of rupture, following the financial and political crises in Brazil, was particularly accentuated for upper-middle class interviewees, especially because the downturn had a direct effect on family enterprises. Corruption scandals, a high tax burden and businesses' bankruptcies were all factors driving their desire to migrate, even if only for a short period. Although the economic crisis has affected Brazil as a whole, national emigration rates remain relatively low. In turn, going abroad appears as the first adaptation strategy for people in GV, founded upon its long-lasting migration culture, in which migration is part of the ‘repertoire of possibilities’ of the city’s inhabitants.

The SAMARCO environmental disaster

Whilst the socio-political-economic crisis appeared spontaneously in people’s narratives, the impact of the SAMARCO environmental disaster was more subtle. When asked about it, most participants in all three groups argued that the catastrophe did not have a direct impact on people’s motivation to move abroad. Yet, the dam collapse and subsequent pollution of the region’s main river, may have affected other migration drivers by influencing both intrinsic and instrumental aspirations to migrate indirectly. Participants remarked, for example, that the economy had become even more stagnant, as most investors were reluctant to open new businesses in the city.

With the disaster, commerce suffered a lot... The entire city life changed, not only in Valadares but all the way to the sea, to Espírito Santo. The bad image that came with it, you see a city that no longer has anything to offer... Who would want to invest in a city like this? What company would want to set up shop here? The soda factory left before it happened, but who will want it? How are you going to treat that [contaminated] water? It's very expensive...So the impact is very large across all sectors (Vander, Group C, EM17).

The catastrophe also negatively affected households’ economies, as most families would now choose to buy mineral water, given the distrust surrounding the quality of tap water. The fact that some businesses were relying on mineral water for cooking also unleashed a general rise in other services: “When it happened [the SAMARCO disaster], there was the issue of water. Everything was more expensive. Some people profited from it because the water that used to cost 6 BRL, is now 20 BRL” (Lorenzo, Group B, EM14).

Following the contamination of the Doce River, most people believed that the city’s drinking water was contaminated with heavy metals, despite authorities' claims that it was safe for consumption. Here, the subjective dimension of the crisis (mistrust in public institutions) was bound up with an objective change in behaviour. The fact that the company responsible for the disaster was not severely punished also contributed to this general feeling of impunity and the sensation that “there is no hope for Brazil”, recurrent across the interviews:

Another aspect is the impunity. Problems happen, and that’s it. SAMARCO was fined, but they did not pay the fine, they are putting off the reparation, and nobody can solve the problem. So, there is this problem of impunity, it seems that there is no hope for Brazil. I think that this augments people’s desire to leave (Ana, Group B, EM9).

Recurrent in most interviews was the perception that, while the environmental crisis was not a direct driver of migration, it had worsened the area’s historical socioeconomic problems and the (extra) burden posed by the financial crisis, killing any hope people had for the region’s economic revival while at the same time aggravating the general feeling of mistrust in public institutions.

Capabilities and modes of migration

Although the United States is the preferred destination in the GV region, because “that is where everyone goes to prosper” and where most people have tight social and family networks, participants from lower socioeconomic backgrounds had a perception that 'for people like them' - who cannot get tourist visas to enter the country through regular pathways - the previous option of migrating via Mexico has become a very costly and dangerous affair. Most were not willing to risk their lives to make the journey, although many returned migrants had used that route the first time they migrated. Thus, despite a heightened aspiration to migrate in the face of the country's simultaneous crises, most members of this social group did not possess the personal, social, or financial capability to do so. This may explain why those from lower SES tend to associate migration with suffering and general feelings of disapproval. In line with the hypothesis put forth by Bourdieu (1997), that people tend to adjust their desires to their effective chances, or their aspirations to their capabilities, over time the working poor tend to adjust their hopes of moving to the USA, a process that can either be drawn from personal experiences or from the experiences of others. As contended by some research participants:

I tell you, for me to set foot in that place via Mexico, never again. I either walk through the front door or I don't (Armando, Group C, EM7, returned migrant from the USA).

In fact, my wish was the United States. I have relatives there, but it was just difficult, complicated...you have to go through Mexico...I mean, for people like me, who don't have resources, it's almost impossible, right? And Portugal is much simpler, right? Easier. I knew I was earning less, but for my safety and... I didn't want to risk it' (Enzo, Group C, EM1, returned migrant from Portugal).

Similarly, most participants from lower-middle socioeconomic backgrounds reported increasing difficulties in obtaining tourist visas to the United States in recent years and all showed an unwillingness to cross the border irregularly through Mexico, given the overall perception that the journey had become more dangerous in the last decade. As one interviewee put it “you have to be desperate [to go through Mexico]” (Marília, Group B, EM4).

In contrast, most participants in group A already had transnational ties with the USA and Europe, and were highly mobile, even if they would generally be classified as ‘non-migrants’. Notwithstanding the fact that none of them was a holder of a US business visa (B-1), at least two reported doing business in the country and in Europe, on a regular basis while on visitor visas. Often, they do not consider ‘doing business’ as their sole motivation to go abroad. Instead, their travels were regarded as leisure trips, where they end up working on the side, usually in Brazilian small enterprises owned by friends and family members. As one participant in this group explained, these visits could eventually help them to build international networks and mark the start of an intense transnational life:

I went to Florida in 1995 and then in 1999, at the beginning of that year. In the first semester, I spent 30 days there. [...] In the middle of that year, I went to Spain, from Spain to Portugal, from Portugal to London, from London to New York, from New York to Boston. I had a cousin in Boston, and I stayed there for 3 months. I didn’t go to make money and bring it back to Brazil. I made money to finance my trip. [...] Then in 2007 I went to work with this cousin, taking care of logistics, focused on Europe, because we were starting a project with a distributor of Brazilian products in Lisbon, to supply products to Spain and Portugal. […] On this occasion, I went to the US and stayed there for 67 days (Getúlio, Group A, EM3).

Others had travelled on tourist visas for short stays with the intention of making enough money to help counterbalance the low tide affecting their business back home: As Paulo put it: “This crisis caused great damage to my company, which is practically forcing me to go there, work for 6 months to solve this” (Paulo, Group A, EM10). The aspiration for short-term migration among members of this group is obviously favoured by the fact that most of them hold US visitor visas, an asset that derives from their privileged class positions in Brazil. At the micro level, our interviews reveal how those of privileged class positions, who are more likely to obtain tourist visas, tend to opt for irregular short-term working stints over irregular permanent settlement. In this case, they may use mobility as a mechanism to compensate for the effects of economic downturns in Brazil.

Another important advantage that can be readily converted into higher capabilities to migrate among Latin American upper classes is the possibility of obtaining dual citizenship through European ancestry. Two out of five people interviewed in this group held European, passports, specifically Italian and French. Although there is not enough data on the prevalence of dual citizenship and its distribution across socioeconomic strata in Brazil, research has shown how dual nationality in developing countries is highly associated with privileged class positions (Harpaz; Mateos, 2019). For both dual-citizen families interviewed in this group, the European passport ensured greater flexibility in migration plans and opened more viable options in times of tightening immigration controls in most developed countries. Importantly, it also meant that they did not need a tourist visa to enter the United States - the longstanding preferred destination - which once more favors their circulation within the Brazil-USA corridor.

Contrary to the other two groups, mobility aspirations could, among the economically advantaged, be more readily converted into movement. US tourist visas, dual European citizenship, previous travel experiences, and transnational linkages afforded interviewees possibilities of circulation that were not available to other potential migrants.

Concluding remarks

Our analysis applies the aspirations-capabilities framework in a context of social, political, economic and environmental crises. The analysis suggests the social representation of international migration is rather similar across SES, supporting existing literature documenting the well-established culture of migration in Governador Valadares, the result of half a century of emigration to the USA. However, people's lived experiences, aspirations, and capabilities to migrate in the context of multidimensional crises vary across class positions. The working poor report living in a ‘permanent state of crisis’, with their livelihoods being constantly affected by the ups and downs of business cycles. Despite the acute economic and socio-political effects of the 2014/2016 recession, this is regarded as just the latest economic downturn to affect their means of subsistence, yet not intrinsically different from other crises. In the face of this perpetual state of few opportunities, emigration has always been considered a tool for improving living conditions. For the lower middle classes, in turn, the crises have rendered visible the mismatch between their (augmented) expectations, built during the early 2000s era of optimism, and their effective chances of social mobility. Migration, therefore, appears as a solution against deteriorating living conditions, but, most importantly, as a way of accessing services they could once afford in Brazil, such as private education and health insurance. Those with upper-middle class backgrounds, who had not previously considered migrating, now see emigration as a way of maintaining their quality of life. In addition, the 2014 crisis represented a rupture: one that did not necessarily hinder their access to basic items, but which threatened their businesses and future class position.

This analysis has also demonstrated how crises simultaneously affect intrinsic and instrumental aspirations to move. Although much has been said about the positive intrinsic values attributed to migration, fewer studies have discussed how exit can also be a way of demonstrating grievances towards a particular location or political situation. Several scholarly works have demonstrated that transnational migrants use 'exit' as a way of voicing their concerns and influencing their countries' political situation from abroad (Fomina, 2021; Hoffmann, 2010; Okamoto; Wilkes, 2008). In contrast, this paper explored how people frame their aspirations to move as the result of feelings of disappointment towards the Brazilian state, even when the voice hypothesis (domestic or transnational) has never been at stake - a sentiment epitomised by the sentence “I am giving up on Brazil”. In this sense, this paper has argued that common migration drivers, such as crisis, corruption, and man-made environmental disasters, can shape migration aspirations not only instrumentally by diminishing access to financial and natural resources but also symbolically through feelings of dismay. Although the SAMARCO dam disaster did not have a direct impact on international migration aspirations, it heightened the effects of other longstanding crises by simultaneously boosting feelings of disillusion towards Brazilian politics and obliterating any hope that the city’s residents had for the region’s economic revival.

With regard to capabilities, results show that middle and upper-middle-class households were particularly averse to migrating to the United States via Mexico, contending that the journey has become too dangerous in recent years. Rather, most maintain that they would only enter the country ‘legally’, a common way of referring to tourist visa overstaying (Jorgensen et al., 2019). In contrast, many of those belonging to lower-income households had previous experiences of crossing the US-Mexico border irregularly. The risks of the journey and the protracted periods of family separation that irregular migration entails could, in fact, explain why individuals of low SES tend to associate migration with suffering much more than their more privileged counterparts. In fact, for the most advantaged group, migration is often associated with circularity, in contrast to permanent settlement. This can be explained by their ability to get around restrictive migration policies by making use of U.S. tourist visas, transnational linkages, or ancestry-based European citizenship. Combined with socio-political, economic, and demographic factors, the culture of migration in GV renders international mobility as the preferred mechanism to deal either with a 'permanent state of crisis’ or with the confluence of newly emerged multiple shocks for households across the entire socioeconomic spectrum. These dynamics are, in turn, inscribed in broader global processes of capital and labour restructuring, development and social change.

These results advance socially embedded analyses of mobility through a mixed methods approach combining a quantitative novel research design that provides evidence on the stability of a collectively constructed entrenched culture of migration in a long-established emigration hotspot in Brazil. In turn, the qualitative analysis allows for a more nuanced understanding on how people deal with multiple crises and how they impact aspirations and capabilities to migrate, according to socioeconomic status, in a context where international migration is a “total social fact”.

Although we did not find relevant differences in social representations of international migration according to SES, other experiences or attributes such as gender, age and migration history may influence the way people signify this type of mobility. This type of analysis requires new stratifications and decomposition methods that can be tested in future studies. Similarly, future studies could explore how SES intersects with other social hierarchies to shape aspirations and capabilities.

References

• ABEP - Associação Brasileira de Empresas de Pesquisa. Critério de classificação econômica Brasil. São Paulo: ABEP, 2014.
• ALI, S. ‘Go west young man’: the culture of migration among muslims in Hyderabad, India. Journal of Ethnic & Migration Studies, v. 33, n. 1, p. 37-58, 2007.
• ANDRADE, M. V.; NORONHA, K. V. M. de S.; SANTOS, A. S.; DE SOUZA, A.; GUEDES, G. R.; CAMPOLINA, B.; CAVALCANTE, A.; MAGALHÃES, A. S.; DUARTE, D.; KIND, P. Estimation of health-related quality of life losses owing to a technological disaster in Brazil using EQ-5D-3L: a cross-sectional study. Value in Health Regional Issues, v. 26, p. 66-74, 2021.
• APPADURAI, A. The capacity to aspire: culture & the terms of recognition. In: VIJAYENDRA, R.; WALTON, M. (Ed.). Culture & Public Action. Stanford: Stanford University Press, 2004.
• ASLANY, M.; CARLING, J.; MJELVA, M. B.; SOMMERFELT, T. Systematic review of determinants of migration aspirations. QuantMig Project Deliverable. D2.2. Southampton: University of Southampton, 2021.
• ASSIS, G. Estar aqui... Estar lá: uma cartografia da emigração valadarense para os Estados Unidos. In: REIS, R.; SALES, T. (Ed.). Cenas do Brasil migrante. São Paulo: Boitempo, 1999.
• ATTRIDE-STIRLING, J. Thematic networks: an analytic tool for qualitative research. Qualitative Research, v. 1, n. 3, p. 385-405, 2001.
• BACHRACH, C. A. Culture & demography: from reluctant bedfellows to committed partners. Demography, v. 51, n. 1, p. 3-25, 2014.
• BAUMAN, Z. La globalización: consecuencias humanas. Ciudad de México: Fondo de Cultura Económica, 2001.
• BLACK, R.; ADGER, W. N.; ARNELL N. W.; DERCON, S.; GEDDES, A.; THOMAS, D. The effect of environmental change on human migration. Global Environmental Change, v. 21, suppl. n. 1, p. S3-S11, 2011.
• BOMENY, H. Do frango ao avião ou o que é possível dizer sobre a nova classe média brasileira? Notas exploratórias. São Paulo: FVG, 2011.
• BOURDIEU, P. Pascalian meditations. Stanford: Stanford University Press, 1997.
• BRADLEY, E. H.; CURRY, L. A.; DEVERS, K. J. Qualitative data analysis for health services research: developing taxonomy, themes, & theory. Health Services Research, v. 42, n. 4, p. 1758-1772, 2007.
• BRITO, F. Ensaios sobre as migrações internacionais no desenvolvimento do capitalismo. Revista Brasileira de Estudos de População, v. 12, n. 1/2, p. 21-34, 1995.
• CARLING, J.; COLLINS, F. Aspiration, desire and drivers of migration. Journal of Ethnic & Migration Studies, v. 44, n. 6, p. 909-926, 2018.
• CARLING, J.; SCHEWEL, K. Revisiting aspiration and ability in international migration. Journal of Ethnic & Migration Studies, v. 44, n. 6, p. 945-963, 2018.
• CARR, E. R. Placing the environment in migration: environment, economy, & power in Ghana’s central region. Environment & Planning, v. 37, n. 5, p. 925-946, 2005.
• CASTLES, S. Understanding global migration: a social transformation perspective. Journal of Ethnic and Migration Studies, v. 36, n. 10, p. 1565-1586, 2010.
• CASTLES, S. Migration, crisis, and the global labour market. Globalizations, v. 8, n. 3, p. 311-324, 2011.
• DIAS, C. A.; COSTA, A. S. V.; GUEDES, G. R.; UMBELINO, G. J.; SOUZA, L. G.; ALVES, J. H.; SILVA, T. G. M. Impactos do rompimento da barragem de Mariana na qualidade da água do rio Doce. Revista Espinhaço, v. 12, p. 21-35, 2018.
• DUARTE, D.; GUEDES, G. R.; PEREIRA, W. H.; RIBEIRO, R. B. Representing collective thinking through cognitive networks. Journal of Complex Networks, v. 10, n. 6, cnac053, 2022.
• ESPINDOLA, H. S. Perfil sumário da macrorregião VIII do Rio Doce. 1995. Mimeo.
• ESPINDOLA, H. S. A história de uma formação socioeconômica urbana: Governador Valadares. Dimensões, n. 6, 1998.
• FAZITO, D. Análise de redes sociais e migração: dois aspectos fundamentais do ‘retorno’. Revista Brasileira de Ciências Sociais, v. 25, n. 72, p. 89-101, 2010.
• FOMINA, J. Voice, exit & voice again: democratic remittances by recent Russian emigrants to the EU. Journal of Ethnic & Migration Studies, v. 47, n. 11, p. 2439-2458, 2021.
• FREIRE, F. H. M. de A.; GONZAGA, M. R.; GOMES, M. M. F. Projeções populacionais por sexo e idade para pequenas áreas no Brasil. Revista Latinoamericana de Población, v. 13, n. 26, p. 124-149, 2020.
• FUSCO, W. Capital cordial: a reciprocidade entre os imigrantes brasileiros nos Estados Unidos. Tese (Doutorado) -Universidade Federal de Minas Gerais, Belo Horizonte, 2005.
• GLICK SCHILLER, N.; BASCH, L.; BLANC-SZANTON, C. Transnationalism: a new analytical framework for understanding migration. In: GLICK SCHILLER, N.; BASCH, L.; BLANC-SZANTON, C. (Ed.). Towards a transnational perspective on migration. New York: New York Academy of Sciences, 1992. p. 1-24.
• GOMES, C.; GUEDES, G. R. Mobilidade e vulnerabilidade socioambiental: um estudo de caso para Governador Valadares, Minas Gerais, Brasil. Revista Espinhaço, v. 9, n. 2, p. 2-15, 2020.
• GRUBANOV-BOSKOVIC, S.; KALANTARYAN, S.; MIGALI, S.; SCIPIONI, M. The impact of the Internet on migration aspirations & intentions. Migration Studies, v. 9, n. 4, p. 1807-1822, 2022.
• GUARNIZO, L. E.; SMITH, M. P. The locations of transnationalism. In: SMITH, M. P.; GUARNIZO, L. E. (Ed.). Transnationalism from Below. New Brunswick: Transaction Publishers, 1998.
• DE HAAS, H. A theory of migration: the aspirations-capabilities framework. Comparative Migration Studies, v. 9, n. 8, 2021.
• DE HAAS, H. The determinants of international migration. Oxford: International Migration Institute, 2011. (IMI Working Papers, n. 32).
• HARPAZ, Y.; MATEOS, P. Strategic citizenship: negotiating membership in the age of dual nationality. Journal of Ethnic & Migration Studies, v. 45, n. 6, p. 843-857, 2019.
• VAN HEAR, N.; BAKEWELL O.; LONG, K. Push-pull plus: reconsidering the drivers of migration. Journal of Ethnic & Migration Studies, v. 44, n. 6, p. 927-944, 2018.
• HOFFMANN, B. Bringing Hirschman back in: ‘exit’, ‘voice’, and ‘loyalty’ in the politics of transnational migration. The Latin Americanist, v. 54, n. 2, p. 57-73, 2010.
• HORVÁTH, I. The culture of migration of rural Romanian youth. Journal of Ethnic & Migration Studies, v. 34, n. 5, p. 771-786, 2008.
• JORGENSEN, N. V.; BARBIERI, A. F.; GUEDES, G. R.; ZAPATA, G. P. International migration and household living arrangements among transnational families in Brazil. Journal of Ethnic & Migration Studies, v. 47, n. 19, p. 4386-4404, 2021.
• KADIANAKI, I.; ANDREOULI, E. Essentialism in social representations of citizenship: an analysis of Greeks’ and migrants’ discourse. Political Psychology, v. 38, n. 5, p. 833-848, 2017.
• KANDEL, W.; MASSEY, D. S. The culture of Mexican migration: a theoretical and empirical analysis. Social Forces, v. 80, n. 3, p. 981-1004, 2002.
• KIM, J. Migration-facilitating capital: a bourdieusian theory of international migration. Sociological Theory, v. 36, n. 3, p. 262-288, 2018.
• LEVITT, P. Transnational migration: taking stock and future directions, Global Networks, v. 1, n. 3, p. 195-216, 2001.
• MARTES, A. C. B. Brasileiros nos Estados Unidos: um estudo sobre os imigrantes em Massachusetts. São Paulo: Paz e Terra, 1999.
• MASSEY, D. S.; ARANGO, J.; HUGO, G.; KOUAOUCI, A.; PELLEGRINO, A.; TAYLOR, J. E. Theories of international migration: a review and appraisal. Population & Development Review, v. 19. n. 3, p. 431-466, 1993.
• NEWMAN, M. E. J.; GIRVAN, M. Finding and evaluating community structure in networks. Physical Review E, v. 69, n. 2, 2004.
• OKAMOTO, D.; WILKES, R. The opportunities and costs of voice and exit: modelling ethnic group rebellion and emigration. Journal of Ethnic & Migration Studies, v. 34, n. 3, p. 347-369, 2008.
• PATARRA, N. L. Migrações internacionais de e para o Brasil contemporâneo: volumes, fluxos, significados e políticas. São Paulo em Perspectiva, v. 19, n. 3, p. 23-33, 2005.
• PORTES, A. Conclusion: theoretical convergencies and empirical evidence in the study of immigrant transnationalism. International Migration Review, v. 37, n. 3, p. 874-892, 2003.
• PORTES, A.; GUARNIZO, L. E.; LANDOLT, P. The study of transnationalism: pitfalls and promise of an emergent research field. Ethnic and Racial Studies, v. 22, n. 2, p. 217-237, 1999.
• ROSA, A. S.; BOCCI, E.; BONITO, M.; SALVATI, M. Twitter as social media arena for polarised social representations about the (im)migration: the controversial discourse in the Italian and international political frame. Migration Studies, v. 9, n. 3, p. 1167-94, 2021.
• SALATA, A. R. Quem é classe média no Brasil? Um estudo sobre identidades de classe. Dados, v. 58, n. 1, p. 111-149, 2015.
• SALES, T. Brasileiros longe de casa. São Paulo: Cortez, 1999.
• SASSEN, S. The mobility of labor and capital: a study in international investment and labor flow. Cambridge University Press, 1998.
• SCHEWEL, K. Understanding the aspiration to stay: a case study of young adults in Senegal. Oxford: International Migration Institute (IMI), 2015. (IMI Working Papers, n. 107).
• SEN, A. Development as freedom. New York: Alfred Knopf, 1999.
• SILVA, T. G. M. S.; RODRIGUES, S. M.; GUEDES, G. R.; DIAS, C. A. Análises do desastre de Mariana identificadas na literatura técnica e científica. Revista de Economia Política e História Econômica, v. 16, p. 81-104, 2021.
• SINGER, A. Quatro notas sobre as classes sociais nos dez anos do lulismo. Psicologia USP, v. 26, n. 1, p. 7-14, 2015.
• SINGER, A.; LOUREIRO, I. As contradições do lulismo: a que ponto chegamos? São Paulo: Boitempo, 2016.
• SINGER, P. Migrações internas: considerações teóricas sobre o seu estudo. In: SINGER, P. Economia política da urbanização. 3.ed. São Paulo: Brasiliense, 1976.
• SIQUEIRA, S. Migrantes e empreendimentos na microrregião de Governador Valadares: sonhos e frustrações no retorno. Tese (Doutorado) - Universidade Federal de Minas Gerais, Belo Horizonte, 2006.
• SOARES, W. Emigrantes e investidores: redefinindo a dinâmica imobiliária na economia valadarense. Dissertação (Mestrado) - Universidade Federal de Minas Gerais, Belo Horizonte, 1995.
• SOARES, W. Da metáfora à substância: redes sociais, redes migratórias e migração nacional e internacional em Valadares e Ipatinga. Tese (Doutorado) - Universidade Federal de Minas Gerais, Belo Horizonte, 2002.
• SOUSA, L.; FAZITO, D. Cultura migratória no município de Governador Valadares: uma análise da rede de significados e seus impactos nos fluxos migratórios internacionais. Revista Espinhaço, v. 6, n. 2, p. 47-64, 2017.
• VARGAS-VALLE, E. D.; GLICK, J. E. Educational and migration aspirations among children of Mexican migrant returnees in a border context. Migration Studies, v. 9, n. 3, p. 677-701, 2021.
• VERTOVEC, S. Transnationalism. Abingdon: Routledge, 2009.
• VEZZOLI, S. Understanding aspirations to stay: relative endowment within a time-space perspective. Migration Studies, v. 11, n. 2, p. 259-285, 2023.
• ZAPATA, G. P. Epicentros de emigración: un análisis comparativo de la evolución de sus dinámicas socioeconómicas y demográficas en Colombia y el Brasil. Notas de Población, v. 108, p. 133-165, 2019.
• ZAPATA, G. P. The migration-development nexus: rendering migrants as transnational financial subjects through housing. Geoforum, v. 47, p. 93-102, 2013.

• Funding:
  This study was funded by the Brazilian Council for Scientific and Technological Development (CNPq), grants 483714/2012-7, 431872/2016-3 and (PQs) 314392/2018-1, 317328/2021-2, 307833/2019-4. Funding was also obtained from the Minas Gerais State Agency for Research and Development (FAPEMIG), grants APQ-00244-12 and APQ-01553-16; and the Brazilian Research Network on Global Climate Change (Rede Clima) / FINEP, grant 01.13.0353.02.
• Ethical Approval:
  The authors certify that their work used information and data from projects submitted and duly approved by the Research Ethics Committee of the Federal University of Minas Gerais (UFMG), CAAE 12650413.0.0000.5149 and CAAE 55007116.7.1001.5149.
• Availability of data and material:
  The software developed for creating cognitive networks is publicly available on GitHub: https://github.com/epopea/wordevok. Qualitative data may include sensitive information (original interviews) and must be carefully anonymized to avoid risks to the people interviewed. The quantitative data on evocations, anonymized, as presented in the article, are available upon justified request. The complete data of the two samples can only be made available, upon justified request, in synthetic sample format, once the names and other characteristics that can identify the research subjects have been removed, and with controlled spatial stochastic perturbation, in order to avoid the risk of identification by spatial characterization.

• 1
  Although we recognise that technically the disaster is ‘technological’, we will refer to it as a socio-environmental disaster given its pervasive consequences on the environment and society (Gill; Ritchie, 2018).
• 2
  In the 2015-2016 period, GDP and households’ consumption decreased around 7 per cent, the worst recession ever recorded in Brazil (IBGE, 2018).
• 3
  The impartiality of the anti-corruption task force investigating the scheme - Operation Car Wash - was later questioned and, in 2021, the Supreme Court annulled most of the convictions that emerged from the investigation.
• 4
  Data from the 2010 Census suggests that Brazilian return migrants accounted for 65% of all recent immigration to Brazil (268,486), 10.3% of which returned to the state of Minas Gerais (Oliveira, 2013).
• 5
  The research received approval by the Ethics Committee of the Federal University of Minas Gerais (UFMG) (CAAE: 55007116.7.1001.5149).
• 6
  The question was "Number of household assets: colour TV, radio, bathroom, car, maid, washing machine, VHS or DVD, fridge, freezer unit"
• 7
  The survey questions were: A) What is the highest level of education obtained by you (respondent) and by members of the household? 1) Illiterate; 2) Incomplete elementary school; 3) Elementary, but incomplete high school; 4) High school, but incomplete college; 5) College or higher education.
• 8
  Original and standardised evocations can be made available as Supplementary Materials.
• 9
  These were defined as families who had or currently have a household member residing abroad.
• 10
  This category includes evocations such as 'I would like to migrate', 'it is my dream', 'a dream shared by many Brazilians', and 'personal ambition'.

Edited by

• Editors:
  Cassio Turra and Igor Cavallini Johansen

Data availability

The software developed for creating cognitive networks is publicly available on GitHub: https://github.com/epopea/wordevok. Qualitative data may include sensitive information (original interviews) and must be carefully anonymized to avoid risks to the people interviewed. The quantitative data on evocations, anonymized, as presented in the article, are available upon justified request. The complete data of the two samples can only be made available, upon justified request, in synthetic sample format, once the names and other characteristics that can identify the research subjects have been removed, and with controlled spatial stochastic perturbation, in order to avoid the risk of identification by spatial characterization.

Publication Dates

• Publication in this collection
  10 Jan 2025
• Date of issue
  2024

History

• Received
  28 Mar 2024
• Accepted
  07 Oct 2024