Introduction

Migration is a phenomenon present in all countries, with a significant impact on the population. The World Migration Report 2018 notes that international migration is a complex phenomenon associated with multiple economic, political, social, and security aspects affecting daily life in an increasingly interconnected world. Current global estimates indicate that, by 2015, there were approximately 244 million international immigrants worldwide, most of whom resided in Europe, Asia, and North America (^{International Organization for Migration [IOM], 2018}).

In the case of Ecuador, according to the Statistical Registry of International Entries and Departures of the National Institute of Statistics and Censuses (^{INEC, n.d.}), for the period 1997-2018, 1 497 680 international departures were registered in 2018: 37.2% to the United States and 7.4% to Spain. Migration is not new in Ecuador. International migration in this country was initially a regional phenomenon, concentrated in the south, due to a crisis in the main productive activity in the area: tequila straw. As a result of the revitalization of migrant networks, it expanded in subsequent decades (^{Serrano, 2008}). Until the late 1990s, the Ecuadorian migratory phenomenon was primarily concentrated in the provinces of Azuay and Cañar, the United States being its main destination (Serrano, 2008).

The first migration from Ecuador to the United States occurred in 1930, according to Ecuadorian census data, but it was not until the early 1960s that it intensified and diversified to destinations such as Venezuela, Canada, and Spain (^{Herrera, 2008}). Over the years, migration to the United States has maintained an upward trend, despite the multiple risks and vulnerabilities entailed by migration to that country, which is largely irregular. By 2010, it was estimated that 435 209 Ecuadorians lived in the United States (^{Datosmacro.com, n.d.}). The American Community Survey (ACS) subsequently estimated that by 2013, the number of Ecuadorians in the United States had risen to 428 500, 58% of whom resided in New York-New Jersey, constituting the third largest group of Latin American immigrants in this metropolitan area, and the ninth largest nationwide (^{Jokisch, 2014}).

According to ^{Jokisch (2014}), between 2000 and 2013, approximately 10 700 Ecuadorians acquired legal permanent residence in the United States annually. The majority (80%) of this group achieved this status through the family, either through sponsors or immediate family member, with sponsored being the preferred admission categories of U.S. citizens. This author notes that in 2012, approximately 170 000 unauthorized Ecuadorians were found in the United States. In regard to the profile of Ecuadorian immigrants, until 2013, they had lower educational attainment than the U.S. population, and were mostly married men, aged between 18 and 30, with irregular status (^{Herrera et al., 2005}; ^{Herrera et al., 2012}; Jokisch, 2014).

Previous studies indicate that migration by Ecuadorians to Spain intensified in 1999, due to the serious financial crisis in the country, resulting in the depletion of the monetary reserves and a sharp increase in the fiscal deficit, and eventually leading to devaluation of the national currency and the dollarization of the economy (^{Herrera, 2008}; ^{Ramírez & Ramírez Gallegos, 2005}; ^{Serrano, 2008}). It was then that the immigration of Ecuadorians to Spain, which, having begun in the 1960s, reached record levels (^{Herrera et al., 2005}). The situation in Ecuador, coupled with the growing demand for labor in the Global North, triggered what has been dubbed by some authors a migratory stampede (Ramírez & Ramírez Gallegos, 2005). Thus, in Spain, while there had been 8 973 entries of Ecuadorian migrants in 1999, by 2000, there were 91 120, equivalent to a 915% increase (^{Herrera, 2008}).

The Ecuadorian population in Spain achieved a peak of 487 239 in 2005 (^{Jokisch, 2014}). In 2010, the number of Ecuadorians in Spain totaled 496 666 (^{Datosmacro.com, n.d.}). In 2013, data from the Spanish National Institute of Statistics showed that approximately 456 233 Ecuadorian immigrants resided in Spain, approximately one third of whom lived in or near Madrid, followed by those living in Barcelona, Valencia, and Murcia (Jokisch, 2014). The migratory flow to Spain was far more geographically and socioeconomically diverse. Immigrants were drawn from all provinces, were more urban and better educated, younger, with an age range of 16 to 44, and legal status. Some even had dual nationality thanks to the regularization law (Royal Decree 2393 of 2004) enacted in 2005 (Jokisch, 2014; ^{Iglesias Martínez et al., 2015}). In addition, between 2008 and 2013, Spain’s family reunification policy enabled nearly 157 000 Ecuadorians to join their family members in the country, facilitating legal residence and therefore Spanish nationality (Jokisch, 2014). Spain is now one of the main migratory destinations for Ecuadorians, second only to the United States.

One of the particularities of the migratory process-not exclusive to Ecuadorians-has been the creation of networks and chains for the flow of people, money, ideas, objects, information, and images that cross borders, boosting links between communities of origin and destination, an issue addressed by Douglas ^{Massey in the 1990}s through the theory of social networks (^{Massey, 1990a}; ^{Massey, 1990b}; ^{Massey et al., 1991}; ^{Massey et al., 1993}; ^{Massey & Espinosa, 1997}). This theory, focusing on social actors, postulates that “Migrants are creating in the receiving societies a set of ties with friends and relatives in receiving societies that established a successive migratory flows to be indefinitely established” (^{Herrera, 2006}, p. 191). It also proposes the category of social capital defined by Bourdieu as “the sum of the current and virtual resources available to an individual or a group for possessing a lasting network of relationships of mutual reciprocal links with others and of mutual recognition” (Bourdieu, 1980, as cited in ^{Aparicio, 2006}, p. 154).

Within this framework, networks promote an increase in migratory flows by reducing the costs and risks of displacement, in addition to facilitating the insertion of immigrants into the country of destination (^{Herrera, 2006}). According to this theoretical approach, migrations are construed as the result of familial and social rather than individual decisions (^{Devoto, 1991}; ^{Durand, 1994}; ^{Durand & Massey, 2003}; ^{Massey et al., 1993}; ^{Pedone, 2006}; ^{Ramírez & Ramírez Gallegos, 2005}).

Following the network theory, each actor, together with others, creates a network in time and space which, as the migration dynamic consolidates, becomes transnational. Transnational migrant networks thereby constitute socio-spatial microstructures with their own dynamics based on a set of relationships and links, strengthening population movements (Pedone, 2002). Transnational migratory networks comprise various actors: the migrant and their family, friends, acquaintances, social organizations, and other associations as well as illegal actors, such as money lenders, pseudo travel agencies, and coyotes.⁵

The functioning of these networks can largely be attributed to the expansion of new information and communication technologies, whereby migrants maintain links with microlocal and local contexts, such as neighborhoods and cities in the country of origin, while simultaneously maintaining links with neighborhoods and cities in the destination country. These links constitute the source of the translocal and transnational nature of migration networks (Pedone, 2002). In fact, what distinguishes current networks from those that existed 50 years ago is the existence of new information and communication technologies and the ease of international travel (^{Davis, 2000}). For ^{Solé et al. (2007}), one of the most significant types of transnational practice, with the greatest impact on the lives of migrants and their families, is long-distance communication. ^{Ramírez (2007}) notes that the possibility of establishing contact in real time and everyday life online has transformed the lives of immigrants and their families.

The theory of networks as a theoretical input enables one to observe how access to social protection tends to rely on a network comprising formal and informal actors, which not only includes links or ties in the country of origin, but also goes beyond borders. Indeed, crossing borders has become a strategy for seeking social protection. In this respect, it is essential to carefully review the meaning of transnational. Transnationalism emerged in the early 1990s, based on the combined approach of various authors who highlighted the dual links migrants simultaneously maintain in the countries of origin and destination, such that their everyday lives depend on constant, multiple interconnections between borders (^{Kearney, 1991}; ^{Rouse, 1991}; ^{Goldring, 1992}; ^{Glick Schiller et al., 1992}).

Transnationalism can therefore be defined as a process whereby transmigrants forge and sustain multi-situated social and relational fields, thereby both maintaining and establishing relationships that cut across geographic, cultural, political, and economic lines (^{Portes et al., 2003}). Thus, beyond sending remittances, what occurs is an exchange of ideas, practices, social capital, and identities circulating between the countries of origin and destination, which has been called “social remittances” (^{Lacroix et al., 2016}, p. 1). According to ^{Cano et al. (2006}), the novelty of the transnational phenomenon is that “new communication and transportation technologies have allowed immigrants, for the first time, to interact almost simultaneously in several places at the same time and have significantly contributed to the expansion of the phenomenon in recent times” (p. 14).

In this sense, to understand the new transnational social dynamics, such as transnational social protection, a broader theoretical-methodological approach is required to grasp the complexity of the phenomenon. This complexity lies on the one hand, in the fact that social protection is provided by several sources apart from the state. On the other hand, the fact that this phenomenon is transnational forces one to examine how sources of social protection operate within and beyond national borders. At the methodological level, it involves moving from a single-site study to an “analysis of relationships between several sites” (Marcus, 2001, cited in ^{Rivero, 2017}, p. 329). The challenge is precisely for researchers to be able to create multi-sited designs that include the relational analysis of sites (Rivero, 2017). In this regard, Rivero (2017) pointed out “it is not the diversity of sites and locations where the researcher is located that contributes to the methodological level, but the construction of a study object focused on the relationships, associations and links between these places” (p. 329).

Transnational social protection is understood as the policies, programs, people, organizations, and institutions that transnationally protect people in areas such as health, housing, education, employment, childcare, and cash transfer (^{Levitt et al., 2017}). Under this premise, four potential sources of social protection have been established: state, market, civil society organization or the third sector (including non-governmental organizations [NGO], church groups and unions) and individual ties (such as family, friends, neighbors, coworkers, and other community members) (Levitt et al., 2017). The present study was conducted on the basis of these sources.

Social protection with a transnational focus has previously been studied from various perspectives. For instance, sources of transnational social protection have been used to examine how they provide support in different areas, not only within nation states, but also by extending their protective arm towards the territory of others (^{Levitt et al., 2017}). It has been shown that immigrants not only meet their own protection needs, but also those of their families-in the countries of origin and destination. This is true of health protection that uses four cross-border strategies: workers’ insurance, mobility, individual and collective remittances, and diasporic health policies (^{Lafleur & Romero, 2018}).

Likewise, other authors have turned their gaze towards the dynamics of transnational social protection networks based on sociodemographic characteristics. This line includes the works of ^{Amelina et al. (2012}) as they investigate how gender and class constitute relevant markers of heterogeneity in the use of informal social protection. Likewise, the research by ^{Huete (2011}) linking transnational social protection with variables such as origin, educational attainment, and length of stay, shows that certain characteristics of migrants contribute to differentiated use based on formal or informal networks, and finds a relationship between educational attainment and access to formal or informal sources.

However, there are also studies on social protection without a transnational perspective, focusing on the aid migrants receive from sources located solely within destination countries. It is striking that within the issue of social protection, this is the most widely adopted approach. Many of these studies have attempted to contrast the differences in access to social protection by sociodemographic characteristics, primarily migratory status (^{Hoang, 2011}; ^{Maldonado Valera et al., 2018}; ^{Sigona, 2012}; ^{Toma, 2012}). Thus, for example, it has been found that men tend to be connected to relatively more extensive networks, whereas women tend to be more linked to family networks that provide information and practical support, but also social protection (^{Hoang, 2011}).

Additionally, it has been reported that women, irregular and married migrants, have greater difficulty accessing social protection (^{Maldonado Valera et al., 2018}; ^{Rubio, 2001}). Likewise, it has been observed that irregular or undocumented status impacts the support immigrants can receive from community organizations and support agencies (^{Sigona, 2012}). A handful of studies have focused on documenting the similarity of dynamics in access to social protection by categories of material and non-material protection (^{Rubio, 2001}; ^{Vasta, 2004}), types of sources of protection (^{Martínez et al., 2001}; ^{Torres, 2013}), and in the state’s social protection policies, in the case of Spain (^{Moreno et al., 2006}) towards the migrant population.

In this study, analyses of social protection with a transnational focus in the Ecuadorian context are used as a starting point, because they constitute a key source of information. Most of these studies have focused on the social protection of Ecuadorians in Spain, while research on the social protection of Ecuadorians in the United States is scant. The following studies are important: ^{Pedone (2004}), who reconstructs the socio-spatial trajectory of Ecuadorian families in Spain, and ^{Ramírez and Ramírez Gallegos (2005}), who model the transnational migratory circuit of settlers from neighborhoods in Quito and Guayaquil to European and American cities.

^{Setién et al. (2011}) focus their research on the role of the transnational networks of Ecuadorian immigrants in Spain, while ^{Torres (2013}) explores the social protection of Ecuadorian migrants in Valencia. Further development is required in this regard, not only because the world is becoming increasingly transnational, but also because Ecuador is a country with a high emigration rate and migratory profiles indicating that these periods of transnational migration could last for prolonged periods. This points to the need to understand the status of transnational social protection.

Given that international migration in Ecuador is drawn mainly from the southern region, this study seeks to analyze the access to social protection of Ecuadorian migrants in two communities in the province of Azuay and two communities in the province of Loja, ideal because of their high levels of migration to the United States and Spain, as evidenced by the Latin American Migration Project (^{LAMP-ECU4, n.d.}) when examining data from the VII Population Census and VI Housing Census, 2010. In particular, this study assesses the extent to which Ecuadorian migrants have access to social protection in terms of employment, housing, and money loans; it seeks to identify which is the main source of support and whether access to this support depends on the sociodemographic profile of the migrant. Based on the theory of migration networks and from a localized perspective in the receiving countries, it is analyzed the access to social protection of Ecuadorian migrants, their sources of social protection by category, and the relationship between social protection and the sociodemographic profile of the migrant. This knowledge is essential for designing new and better public policies to benefit this population, particularly since there are no other sources of information. It is also important to contribute to filling in the knowledge gap in the area of social protection, as this could lead to further research.

Methodology

This study focused on the social protection migrants from four Ecuadorian communities have received in their destination countries, the United States or Spain, either at the time of their departure or once they had settled there. Data are taken from the 2012 Ecuador Latin American Migration Project⁶ (^{LAMP-ECU4, n.d.}), designed to gather social, economic, and demographic information on Ecuadorian migration. LAMP-ECU4 surveys have a high level of representativeness at the community level, since samples are taken in from at least 200 households, guaranteeing a level of confidence of 95% and a margin of error of approximately 6%, considering finite populations. Households were selected through simple random sampling based on a sampling frame listing the dwellings in each community.

The information corresponds to four communities in southern Ecuador, two in Azuay province with high migration to the United States (one urban and one rural), and two in Loja province with high migration to Spain (one urban and one rural). ^{LAMP-ECU4 (n.d.)} conducted 803 household surveys, obtaining 135 with migratory experience. Within the latter, 243 members of the respondents’ households were identified who had migrated: 143 to the United States and 100 to Spain.

It should be noted that, although the data are more than a decade old, due to the lack of information, it is important because it provides information on a specific aspect, which will be useful for public policy design, since migration patterns have not changed in recent years, with the United States and Spain continuing to be the main destination countries for Ecuadorians (^{Datosmacro.com, n.d.}). The sources of social protection for migrants have probably not changed either.

For the statistical analysis, the sample was adjusted by a weighting factor of each community considered essential for the correct calculation of the descriptive statistics of the population under study, when more specific groups of the population (minorities, unemployed persons, and others) were surveyed. Weights were calculated as the inverse of the sample fraction. Using the weightings, the number of household members with migratory experience to the United States and Spain amounted to 853 cases, 492 and 361 respectively. Table 1 describes the specific weights of the sample for each community in Ecuador.

A descriptive, inferential analysis was conducted of social protection, with an emphasis on sending countries. Social protection was constructed based on the answers to the questions:⁷Who provided you with housing when you first arrived? When you needed money, who did you go to? How did you get your job? Three categories of social protection were subsequently defined: employment, housing, and money loans. In employment, support for or help finding a job was included. In housing, housing assistance was considered, and in money loans, the support or help of cash transfers was included. In regard to the sources of social protection, each question had a series of responses, which were recoded considering the four sources of social protection established by ^{Levitt et al. (2017}): state, market, civil society organizations, and individual ties.

In this sense, state comprised the protection of the state or its affiliated entities from the highest to the lowest levels. Market included individuals or entities from the private sector such as employers, banks, financial entities, and employment agencies. Civil society organizations encompassed NGOs, shelters, church groups, and unions. Individual ties comprised the nuclear and extended family, friends, neighbors, and coworkers. It should be noted that there were responses in the categories employment, housing, and money lending that were not included within social protection labels. In these cases, responses included the following: “I did not need help, I was looking on my own and no-one offered help.” These responses were therefore assumed to signify the absence of social protection.

Within this framework, the analysis was conducted in three stages. In the first stage, access to social protection was analyzed based on the percentages of the presence and absence of protection, disaggregated by country of destination and categories of social protection: employment, housing, and money loans. In the second, an analysis was undertaken based on the sources of social protection by the country of destination and protection categories: state, market, civil society organizations, and individual ties (family, friends, neighbors, and community) defined by ^{Levitt et al. (2017}). Percentages of access by source of protection were compared by country of destination and protection categories.

Finally, the relationship between social protection (a dichotomous variable with a value of 1 for presence and 0 for absence of protection) and the sociodemographic profile (gender, age, marital status at the time of migration, years of education, migratory status, place of origin, number of trips abroad, and number of months abroad) was analyzed. The relationship was determined by statistical tests. Pearson’s chi square test (χ²) was used to determine the relationship between nominal categorical variables (social protection in regard to gender, marital status at the time of migration, migratory condition, and place of origin). The Mann-Whitney U test (U) for independent samples that did not follow a normal distribution was used to determine the association between a nominal categorical variable (social protection) and a discrete quantitative variable (age, years of education, number of trips abroad, and number of months abroad). In addition, conditioning probabilities were calculated as a complement to the associations found.

Results

The results are presented in three sections: 1) access to social protection by migrants in each category and country of destination; 2) access to social protection by source and category; and 3) relationship between social protection and the sociodemographic profile of the migrant.

Access of Ecuadorian Immigrants from Four Communities to Social Protection in the United States and Spain

Analyses of the data revealed that a large percentage of Ecuadorian migrants from the four communities had access to social protection in both countries of destination, as borne out by the three categories of social protection studied, although the social protection of Ecuadorian migrants in the United States and in the category of housing were more evident (Table 2).

Regarding the absence of social protection, one particular feature emerged in the analysis. Not all subjects without social protection were equal, in the sense that some were proactive in response to the limited social protection from the state, market, third sector or individual ties, as their ability to find work for themselves (response categories: looking for work and standing at a corner) distinguished them from the other totally unprotected migrants (migrants who gave nobody as an answer).

Social protection showed a significant association with the country of destination in the categories of social protection in employment (χ² = 21.86; p = 0.000) and housing (χ² = 39.94; p= 0.000). However, no such association was observed in the category of money loans, for which access was similar in both countries of destination. Table 2 shows that the probability of migrants obtaining social protection for employment in the United States (0.728) was higher than in Spain (0.569). Likewise, in the housing category, the probability of obtaining social protection was 0.996 in the United States, compared to 0.905 in Spain.

An analysis of social protection based on the set of categories (employment, housing, and money loans) showed that there is compensation for social protection (ranging from 0 to 3), meaning that the absence of protection in one category was offset by social protection in one or more other categories. However, there were differences by country of destination in the compensation of social protection. These differences were statistically significant (χ² = 30.06; p= 0.000), indicating that the number of types of social protection varies according to the context. Table 3 shows that for every 100 migrants, three did not receive social protection in the United States, whereas 10 failed to do so in Spain.

An analysis of social protection by source revealed that for the majority of the migrants studied did not receive support from the State, but did receive support from the market, civil society organizations and individual ties. However, the traditional model based on family, neighbors, community or individual ties was the main source of support. This observation was a common denominator in the three categories of social protection and the two countries of destination. It should be noted that support received through a family ties was more prevalent in the United States than in Spain. In fact, it was statistically determined that there is an association between sources of social protection in employment and the country of destination (χ² = 11.71; p= 0.001), as well as between sources of housing support and context (χ² = 47.66; p= 0.000), while no association was found in the category of money lending (χ² = 4.42; p= 0.109). These findings suggest that social protection in housing and employment depends on the country of destination, comparatively, more support is received in the United States than in Spain. However, in regard to money loans, protection was similar in both countries. Table 4 shows that more than 90% of the observed migrants obtained support in the three study categories thanks to their individual ties.

Discussion

The present quantitative study analyzes access to social protection in destination countries, sources of social protection, and the association between social protection and sociodemographic profile in a sample of Ecuadorian migrants from four communities in southern Ecuador.

The results show that Ecuadorian migrants originating from the four communities mainly received social protection in employment, housing, and money loans. In terms of protection categories, support was therefore greater in the housing category, whereas among destination countries, support was greater in the United States. For these migrants, individual ties were the main if not the only providers of social protection on their arrival in the destination countries. The social protection model can be said to remain traditional, based on the family, friends, neighbors, and community. This tendency may not have changed, largely because of the crucial role individual ties play in the lives of migrants. Their role could even be said to have intensified in the context of the pandemic and post-pandemic, which warrants investigation. On the other hand, it was found that social protection is associated with certain variables of the sociodemographic profile. This varies, depending on the categories of social protection and destination countries.

Indeed, access to social protection of the majority of migrants in this study is the result of the provision of support or help through individual ties. In the set of sources of protection, the fact that individual ties are the main source of supply may be due to immigration status, since as other research has shown (^{Maldonado Valera et al., 2018}; ^{Rubio, 2001}; ^{Sigona, 2012}; ^{Vasta, 2004}), irregularity limits access to social protection from the state or market. In these conditions it is common for support from civil society and/or family, and community ties to prevail, since it seems that “the more vulnerable condition makes them more dependent on these aids” (Rubio, 2001, p. 14). From this perspective, differences in access to social protection by destination countries make more sense, since Ecuadorian migration to the United States occurs primarily in irregular conditions (^{Serrano, 2008}).

At the same time, the concept of absence of protection as opposed to access to protection is relevant, since migrants displayed differences in the way they dealt with this situation. Many of these migrants reported having used themselves to obtain protection, which has been called self-protection, although this nuance is lost when it is grouped within the category of absence of protection. A change in the perspective of study would contribute a great deal to academia, especially given the dearth of studies focusing on the absence or social lack of protection.

Regarding the source of social protection from “individual ties,” it should be noted that, as in other studies of Ecuadorian migrants such as those by ^{Ramírez and Ramírez Gallegos (2005}), ^{Pedone (2004}), ^{Setién et al. (2011}), ^{Torres (2013}), ^{Vasta (2004}), ^{Huete (2011}), ^{Hoang (2011}), ^{Toma (2012}), ^{Levitt et al. (2017}), and ^{Maldonado Valera et al. (2018}), the findings of this study show how individual ties play a crucial role in the lives of migrants, particularly support from the nuclear family, which increases on their arrival in destination countries, and undoubtedly constitutes an invaluable resource (^{Sigona, 2012}).

Differences in access to social protection between categories may be linked to migrants prioritizing their needs on their arrival in destination countries, foremost among which is having a place to stay. Likewise, ^{Setién et al. (2011}) state that in a new country [obtaining housing] is the main issue. This would also explain the existence of migrant diasporas, since it has been shown that the choice of a destination city depends more on the networks or links they have there than on any other factor (^{Ramírez and Ramírez Gallegos, 2005}). Support for categories such as housing or employment was probably the result of a negotiation prior to their arrival, with individual ties, with a view to guaranteeing support. In addition to housing, it may also have included being met at the airport, as shown in the research by ^{Solé et al. (2007}) and more clearly in Ramírez & Ramírez Gallegos (2005). Along these same lines, ^{Vasta (2004}) argues that in contexts where everyone knows everyone else, community norms and reciprocity abound.

The results of this study revealed the relationship between access to social protection and the variables in the sociodemographic profiles of migrants. This confirms the findings of ^{Amelina et al. (2012}), who found that gender and class affect access to informal social protection, and those of ^{Huete (2011}), who demonstrated that the origin, educational attainment, and length of stay of migrants determined a differential use based on formal or informal networks. ^{Hoang (2011}) and ^{Toma (2012}) also found a similar pattern to that found in this study. These authors reported that men tend to be connected to relatively more extensive networks, whereas women are more likely to be linked to family networks. Likewise, ^{Sigona (2012}) found that illegal status is a distinguishing factor limiting access to services and their aspirations. ^{Maldonado Valera et al. (2018}) observed that social protection is affected by gender and immigration status. In their study, women and irregular migrants have less access to formal sources, as well as less access to goods and human rights restrictions.

In the case of this study, the association between social protection and the sociodemographic profile of the migrant varies according to the country of destination and the category of protection. In the United States, markers of heterogeneity in access to social protection included marital status when they migrated, immigration status, age, and number of months abroad. In this context, access to social protection was more likely for married, undocumented migrants. In Spain, these markers included sex, marital status at the time of emigration, migratory status, place of origin, years of education, number of months abroad, and number of trips abroad. Thus, access to social protection was most likely for immigrants who were male, married, documented and from the province of Loja. Although the characteristics constituting markers of heterogeneity in access to social protection vary between destination countries, migratory status, and gender are the main characteristics helping or hindering access to protection in various destination countries for migrants, according to various sources.

The differences between these two countries lie in the characteristics of migration. Migration to the United States is a long-standing tradition that has been predominantly male. The majority of migrants to the United States were young, married, and male, traveling under irregular conditions, and mainly drawn from the southern region of Ecuador (^{Herrera et al., 2005}; ^{Herrera et al., 2012}; ^{Jokisch, 2014}). Conversely, Ecuadorian migration to Spain is characterized by being much more geographically and socioeconomically diverse. Migrants came from all provinces, including Loja, were urban, better educated, and young, with regular migratory status and in some cases, held dual nationality.

Although the study data do not make it possible to generalize about the access to social protection of Ecuadorian migrants in the main destination countries, they do provide interesting information pointing to the need for more extensive studies of both networks, categories, and sources of protection, as well as research on social vulnerability due to the different behaviors found among unprotected migrants. In addition, they confirm the findings of previous research. Based on this, decision makers could focus their actions and/or justify the formulation or reformulation of policies for Ecuadorian migrants. Looking to the future, it would be interesting to explore whether different sources within the resource environment interact with access to employment, housing, or money lending, and determine the contribution or weight of each of the sources in the social protection of immigrants. This would contribute to science and society because it would allow for a greater understanding of new phenomena and inform public policy for the benefit of the immigrant population.

Conclusion

This study contributes to the knowledge of transnational social protection by showing that a high percentage of Ecuadorian migrants from four communities in southern Ecuador have access to social protection, their main source of social protection being individual ties. Finally, access to social protection was associated with the sociodemographic profile of the migrants. This information contributes to the construction of an explanatory model of social protection and suggests topics for future research.

Among the four types of social protection, individual ties played a major role. This does not mean that the state, market, or civil society organizations were absent, but rather that they reflect the need to undertake research processes focused on the role of each of these sources of protection in the lives of migrants.

Although this study refers to the access to social protection migrants receive on their arrival in the country of destination based on information from more than a decade ago, the results constitute a powerful input for the academic community because they shed light on complementary, new, and better research. They also serve as an input for the design of public policies in the absence of other more current research and information sources. This study also showed the importance of conducting research on the absence of social protection.

Finally, this paper emphasizes the need for up-to-date information for decision-making in terms of social protection for Ecuadorian immigrants, who were among the most vulnerable populations during the pandemic and warrant more attention. In this respect, cooperation between academia, government, and society will provide mutual benefits, greater knowledge, and better policies and conditions for this population.