Aging, Fertility and South-North and South-South Migration ³

As an eminently demographic phenomenon, international migrations have had and still have very important effects on the population structure. They also have an impact on aging, a trend that is currently of great concern in countries that have reached a posttransition phase or a second demographic transition. The increase in life expectancy in the elderly and the decrease in fertility rates (which in some cases is below the level of intergenerational replacement) accentuate the aging process, causing a shortage of young people entering the labor market which along with changes in the population structure give rise to a demand for immigrant workers to compensate for this population imbalance ( ^{Van de Kaa, 1987} , ²⁰⁰² ; ^{Bongaarts, 2001} ). These trends, which characterize the second demographic transition, have been more evident since the late 1990s, with significant social and economic implications ( ^{Lee and Mason, 2011} ).

In this regard, there are two aspects that are of particular concern. On one hand, the direct and indirect contribution made by the immigrant population in the societies of destination ( ^{Canales, 2015} ). Despite the significant effect on the population structure, migratory movements do not fully compensate for the negative effects of aging and the decrease in the relative weight of the workforce, as shown by research for the case of south-north migrations in Europe ( ^{Bijak, Kupiszewska, Kupiszewski, Saczuk, and Kicinger, 2007} ; ^{Castro and Rosero-Bixby, 2011} ) or for south-south migrations in Costa Rica ( ^{Reboiras, 2015} ). On the other hand, in the countries of origin, migration causes a transfer of the demographic bonus, given the impossibility of taking advantage of the window of opportunity represented by a population structure with a heavier active and reproductive age population and a smaller dependent population. However, migration is driven by both a demographic and social complementarity, according to the labor market segmentation theory of destination countries ( ^{Piore, 1979} ; ^{Massey, Arango, Hugo, Kouaouci and Pellegrino, 1998} ).

The objective of this article is to study the interrelationship between fertility and migration by analyzing the reproductive patterns of Peruvian migrants in Chile and Spain, defined as migrants because their place of birth is different from the country of destination, regardless of whether or not they have attained nationality. In other words, the aim is to compare the behavior of the south-south and south-north flows of Peruvian women. One of the main features of this approach is that, despite its importance in quantitative terms, south-south migration has not yet received the necessary attention from the social sciences, and to a lesser extent the comparison with south-north migration. This situation is associated, among other reasons, with the lower availability and quality of existing databases ( ^{Ratha and Shaw, 2007} ; ^{Hujo and Piper, 2007} ).

To carry out this objective, the country of birth (Peru) is first used as the defining criteria for immigrant status. Secondly, since both migratory flows are, to a large extent, of an economic nature, a certain homogeneity in the immigrants' objectives is guaranteed. In addition, both Chile and Spain, although with different intensities and stages in the demographic transition, share a clear trend towards population aging and a notable increase in economic immigration since the end of the twentieth century, so there is a similarity between the contexts of reception that allows the proposed comparison. At the same time, the main heterogeneous aspect continues to be two of the main destinations of Peruvian migrants during the last decades: together with Argentina, Chile is considered one of the main destinations of the south-south movements of Peruvian migrants while Spain stands out as a new destination since the 90s of south-north flows, despite the fact that the United States continues to be the first objective of the north for Peruvians. Along with the destination, which entails different migration pathways and selectivity processes, key differentiating features of immigrant women that can influence the selectivity of flows and their reproductive patterns are analyzed: socioeconomic origin (educational level), legal integration (acquisition of nationality at destination) and integration into the labor market (labor situation). The South-South migration boom of Peruvians in Chile has been called the “fifth migration wave” ( ^{Navarrete, 2007} ). According to data from the Department of Foreigners and Migration ( ^{DEM, 2016} ) and the National Statistics Institute of Chile ( ^{INE-Chile, 2016} ), immigrants have risen from less than 200,000 in 2001, 300,000 in 2010 — about 2% of the total population— to more than half a million in 2015, representing 3% of the total number of residents. Peruvians are by far the main immigrant group in Chile: more than a third of all immigrants are Peruvians, which means around 2% of the total population and a total of 119,366 Peruvians in 2012 ( ^{INE-Chile, 2012} ).

In the case of Spain, according to the National Statistics Institute ( ^{INE-España, 2017} ), in 1998 foreign-born residents represented only 2.9% of the population; a decade later in 2008, Spain had more than 6 million immigrants, 13.1% of the total population. In this immigration boom, Latin Americans have represented the main region of origin; however, the flow of Peruvians to Spain has been much less than that of other Andean countries such as Ecuador, Colombia or Bolivia. Even so, the number of Peruvians living in Spain went from 47,304 in 2001 to 162,425 in 2008, representing slightly less than 3% of the total number of immigrants in that country ( ^{INE-España, 2017} ). The impact of the serious economic crisis that began in 2008 and its effects on the labor market have caused a radical change in the context of welcoming immigrants in Spain ( ^{Esteban, 2011} ; ^{Grande, Paniagua and Del Rey, 2016} ). This impact has led to significant flows of returns or re-migration to third countries, particularly among Latin American immigrants ( ^{Recaño and Jáuregui, 2014} ). 66,687 Peruvians were living in Spain during 2006, which is 58% less than at the time prior to the outbreak of the aforementioned crisis ( ^{INE-España, 2017} ). As a result, for the purposes of this article, the fertility of Peruvian women in Spain in 2007 is studied, which enables the study of the comparison between south-north and south-south migration without the distortion caused subsequently by the outbreak of the economic crisis.

After this introduction, the article is structured in four other paragraphs. In the second section, a review of the main hypotheses explaining the reproductive behavior of the immigrant population is presented as a theoretical framework. The third section explains the data sources used and the methodology used. The fourth section presents the results of the research. Finally, the fifth section summarizes the main findings in a miscellaneous way and proposes, based on them, the discussions and lines of research that open from the analysis.

Population Structure

Before focusing on the reproductive patterns, it is important to analyze the impact of migratory flows on the structure of the population of the destination countries. Both in the case of south-south migrations from Peru to Chile and south-north migrations from Peru to Spain, a young structure is observed with the stock of immigrants of both sexes concentrated in the active age (graph 1). This is the consequence of an international movement, mostly of an economic nature. It should be noted that in the selected years 70% of Peruvian women living in Chile and almost three quarters of those living in Spain are active in the labor market, that is, they work or are looking for work (table 1).

Source: Prepared by the authors based on Chile's 2012 Population Census (^{INE-Chile, 2012}) for the native Chilean population and Peruvian immigrants in Chile, and Spain's 2012 Continuous Municipal Register (^{INE-España, 2017}) for the native Spanish population and Peruvian immigrants in Spain.

Graph 1. Comparative population pyramids of Peruvian immigrants and natives in Chile 2012 and Spain 2012L  

The Peruvian immigrant population has a double effect on the aged structure of the native population that can be seen in the population pyramids of both destination countries, although with a more notable aging in Spain than in Chile. On the one hand, as has already been said, immigration is an injection of active population that slightly reduces the dependency ratio of host societies. On the other hand, it means an increase in the number of women of reproductive age: between 15 and 49 years old. The relative proportion of Peruvian immigrants is much higher than that of native women in both cases, as shown in Graph 1. In other words, in addition to the direct contribution of people of economically active age, through new births immigration can contribute to a rejuvenation of the population in the medium and long term.

Reproductive Guidelines of Migrant Women

The reproductive guidelines studied show important differences between the population in the country of origin and the emigrated population, between natives and immigrants, and between south-south and south-north Peruvian migrations.

In the first place, a higher fertility is observed among Peruvian migrants than among those native to the countries of destination. Table 2 shows these differences in both TFR and CFR. In the case of Chile, native women had an average of 1.94 children, reflecting a mature phase of the demographic transition, while the number of children of Peruvian women residing in Chile was 2.36 children. A similar relationship is observed in the case of Spain. At that time the natives had an average of 1.25 children and the Peruvian immigrants, 1.51. The CFR reaffirms this difference between natives and immigrants, but also shows a clear trend of a progressive decrease in the final fertility according to the age they had at the time of migration and, therefore, of greater similarity with the natives; that is to say, that the average decreases depending on the younger one arrived at the destination country.

On the other hand, if the reproductive calendar is observed through the ASFR (graph 2), it is concluded that the natives present in the two study cases a more delayed calendar than the Peruvian immigrants. In both cases, up to the age of 29, the fertility of the immigrants is much higher than that of the native women, especially for the Peruvian women in Chile, who have their highest ASFR in that age range. Between the ages of 30 and 34, the differences between native women and immigrants in Chile are not significant; although the latter maintain a higher ASFR, a much higher fertility is observed in that age range among native women compared to immigrants in Spain, although these also reach their highest ASFR at that age, although with a much less concentrated calendar than that of local women. From the age of 35 onwards, in both cases there is a higher fertility of immigrants compared to native women, as a consequence of the need to “recover” interrupted fertility due to the effect of the migratory movement itself.

Source: Prepared by the authors based on Population Census 2012 of Chile (^{INE-Chile, 2012}), Population and Housing Census of Peru (^{INEI-Perú, 2007}), Natural Movement of the Population (^{INE-España, 2008a}) and Continuous Municipal Register 2012 of Spain (^{INE-España, 2017}).

Graph 2. Comparison of Specific Fertility Rates by Age in Peru, Chile and Spain according to native population and Peruvian immigrants  

Second, unlike non-migrant Peruvians, female migrants have lower fertility and a more delayed reproductive calendar (Table 1 and Graph 2). In this sense, at first sight the evidence seems to reinforce the hypothesis of selectivity, given that Peruvian immigrants have lower fertility and a more delayed reproductive calendar than their compatriots in Peru. Although there are significant differences between those who live in Spain and those who live in Chile, the latter's behavior is more similar to that of women in Peru.

From these first two comparative exercises it can be concluded that the fertility of Peruvian immigrant women follows a kind of double movement. The first confirms the hypothesis of socialization, which would confirm that, compared to native women, immigrants maintain to some extent a reproductive behavior more similar to that of origin, or at least a complete assimilation of the reproductive patterns of destination is not observed. Moreover, such assimilation occurs to a greater extent when the host society has reached an earlier age (in the case of both Chile and Spain); therefore, reproductive socialization has taken place not only in the country of departure, but also in the host society.

At the same time, following the hypothesis of assimilation and selectivity, there is a second movement, which causes that compared to the Peruvians who have not migrated, the immigrants adopt a less intense reproductive behavior and with a more delayed calendar, thus approaching the guidelines of the natives. Between both movements (which push and determine the fertility of the immigrants) placing it between that of the population of origin and that of the natives, the effect of the interrelation of events between fertility and migratory movement comes into play.

Thirdly, the results show that reproductive behavior is significantly different between south-south and south-north migrants, with TFR 2.36 and 1.51 respectively (Table 2), and with a delayed schedule for those who migrated to the European country as opposed to those who migrated to the neighbor country of the Southern Cone (Graph 2). On the one hand, the Peruvians that migrated to Spain show a reproductive behavior closer to the patterns of the Spanish natives, due to the important selectivity of the south-north flow and the greater weight of the adaptation factor. On the other hand, those residing in Chile show a less differentiated fertility level with respect to the country of origin, that is to say, more determined by socialization at origin than by assimilation, although with a calendar notably more delayed than that of the Peruvians who have not migrated.

These differences between south-south and south-north migrations are explained, on the one hand, by the late fertility and the lower number of children of Spanish natives, which generates a context that influences the patterns of assimilation more than does the context of Chilean reception. As it has been demonstrated for the case of Peruvian women in Argentina ( ^{Rosas, Cerezo, Cipponeri and Gurioli, 2008} ) the context and characteristics of the destination population determine, directly and indirectly, fertility levels and the differences between natives and immigrants.

On the other hand, given the scarce antiquity of migrations in both cases, due to certain patterns of selectivity of migrations towards the north and towards the south, which would represent a part of that difference that is not due only to assimilation. Peruvian migrants in Spain, compared to those who have gone to Chile (as observed in Table 1) have a higher level of education and greater economic capital, since the investment for migration to Europe is significantly higher, so it is possible to assert counterfactually that these women would have had fewer children than the average even without having undertaken migration. In order to investigate the possible patterns of selectivity, the TFR and ASFR are analyzed according to the different socioeconomic characteristics selected.

Fertility According to Level of Studies

Table 1, which shows the distribution of the socioeconomic characteristics of the target populations, reflected strong differences in the educational level of Peruvian women of childbearing age in Peru, Chile and Spain. These data allow us to affirm that there is a clear positive selection in both migratory movements, but much more intense in the south-north migrations than in the south-south ones. In Peru, there is a greater presence of fertile women with basic education (primary or less). But among south-south migrants in Chile there are three quarters who have secondary education, far above the percentage among the population at origin. Among south-north migrants in Spain more than a third have higher education, while only 14.2% of Peruvians in Chile and 16.2% of those who have not migrated reach that level. How does this selectivity, according to educational level, affect reproductive behavior?

The ASFR by age and last grade of studies (graph 2) confirm the selectivity hypothesis. In other words, the lower FER among south-north Peruvian migrants is largely due to the lower fertility of women with higher education, who are over-represented in this migratory flow. However, significant differences continue to be observed at the same educational level. On the one hand, in the south-south migration to Chile, Peruvians present a slightly higher fertility and a calendar similar to the population at origin with secondary and higher education levels, which would reinforce the importance of socialization patterns. This is not the case for Peruvians who live in Chile and have a lower level of education, probably with their fertility marked by difficulties in integrating into the labor market and economic conditions that encourage motherhood.

On the other hand, in the case of the south-north movements, Peruvians in Spain show significantly less intense fertility and a more delayed calendar than the population of origin and their compatriots in Chile. In other words, at the same level of studies, the fertility of Peruvian women in Spain shows a tendency similar to the patterns of the natives of the destination society.

In the explanation of these differences according to type of migration, both the interrelation of events as well as the possible greater delay in the calendar due to the migratory movement in the south-north migration than in the south-south, and the processes of integration and labor insertion in destination (which are different according to level of studies; for example, see the higher fertility of Peruvian women in Spain who have a lower level of education, such as the reasons for migration itself) the type of transnational contact with the society of origin and the regrouping processes that determine reproductive patterns throughout the life cycle and which articulate themselves in different ways in south-south and south-north migrations.

Source: Prepared by the authors based on Population Census 2012 of Chile (^{INE-Chile, 2012}), Population and Housing Census of Peru (^{INEI-Perú, 2007}), Natural Movement of the Population (^{INE-España, 2008a}) and Continuous Municipal Register 2012 of Spain (^{INE-España, 2008b}).

Graph 3. Comparison of Specific Fertility Rates by Age of Peruvian Immigrants in Chile and Spain according to level of studies  

Fertility According Nationality Acquisition in the Country of Destination

To test the hypothesis of legality and the effect of integration processes in the host society, graph 4 represents the ASFR by age, comparing migrants who have obtained the nationality of the country of destination with those who have not acquired it (even if they are residing legally). It should be remembered that access to nationality is very different in both cases of study. One-third of Peruvians of reproductive age residing in Spain in 2007 had acquired Spanish nationality; however, only 7.4% of Peruvians living in Chile have Chilean nationality (Table 1). These differences are caused by the different criteria established by States for the acquisition of nationality and by the migrants' own decisions according to their needs related to the value of the legal status in destination or to the future plans of the migrants.

Source: Prepared by the authors based on Population Census 2012 of Chile (^{INE-Chile, 2012}), Population and Housing Census of Peru (^{INEI-Perú, 2007}), Natural Movement of the Population (^{INE-España, 2008a}) and Continuous Municipal Register 2012 of Spain (^{INE-España, 2017}).

Graph 4. Comparison of Specific Fertility Rates by Age of Peruvian immigrants in Chile and in Spain according to acquisition or not of the nationality of the country of destination  

In the case of south-south Peruvian migrants, the acquisition of the nationality of destination (although it is a minority) is associated with a lower level of fertility, and above all, with a much delayed reproductive calendar compared to those who have not acquired the nationality, who show a very similar behavior to that of their compatriots who have not emigrated. In the case of south-north migrants, the relationship is inverse. Peruvians in Spain who have acquired nationality have a more delayed reproductive calendar and a higher TFR than those who do not have nationality (1.59 as opposed to 1.37 children per woman). The delay in the reproductive calendar is explained by the interruption hypothesis, due to the postponement of child raising during the first years after migration, which ends up delaying the age at which children are born and, at the same time, it encourages the length of time needed to acquire nationality.

The delay in fertility shown by migrants who acquire nationality presents evidence in favor of the assimilation hypothesis, showing a calendar more similar to that of native women among those who have a better legal insertion. However, in the case of south-north migrants, the higher fertility among those having nationality contradicts the postulates of the hypothesis of assimilation of the patterns of the natives, that is, a better legal integration does not imply a reproductive behavior more similar to that of the natives as it does occur in south-south migrants. For Peruvians in Spain, the initial economic objective of migration, together with the significant weight of pioneering women, means that economic and legal integration is prioritized to a large extent initially after arrival, and then the objectives of family reunification and integration are addressed.

Fertility According to Employment Situation

Finally, the reproductive behavior of migrant women also shows notable differences with respect to their employment situation, comparing economically active women (working or actively seeking work) and inactive women. The prevailing economic motive in both migratory flows marks a starting-point difference: compared to only 42% of Peruvian women of reproductive age who are economically active in Peru, the percentage rises to 70.6% among Peruvian women in Chile and up to 74.9% in those who migrated to Spain. Looking at the ASFRs by age for active and inactive women, a very similar reproductive behavior is observed between the population at origin and south-south migrants in Chile. The most significant differences appear again in the case of south-north migrations to Spain, which, although maintaining the same pattern of much higher fertility among inactive women, in both labor situations they have a lower TFR than those of Peruvians in Chile and Peru.

Source: Prepared by the authors based on Population Census 2012 of Chile (^{INE-Chile, 2012}), Population and Housing Census of Peru (^{INEI-Perú, 2007}), Natural Movement of the Population (^{INE-España, 2008a}) and Continuous Municipal Register 2012 of Spain (^{INE-España, 2017}) and National Inmigrant Survey (^{INE-España 2008b}).

Graph 5. Comparison of Specific Fertility Rates by Age of Peruvian Immigrants in Chile (2012) and in Spain (2005-2007) according to employment situation  

In short, the pattern of selectivity of migratory movements and their economic motivation, which imply the greater incorporation of women into the labor market, can explain the lower fertility of immigrants, if compared with the total population. Given these significant differences between economically active and inactive women, it can be concluded that occupational and reproductive decisions are joint for immigrant women. Thus, their fertility depends to a large extent on the family or economic objective of migration, and on the subsequent work strategies of women that are closely linked to the labor market situation of their partners. In this sense, the literature underlines the sustainability of patriarchal family models among south-north Latin American migrants ( ^{Sana and Massey, 2005} ), explaining the exit of immigrant women from the labor market in favor of child care and custody ( ^{Del Rey and Grande, 2015} ). For example, the progressive increase in female labor participation and motherhood among economically active women have been two important achievements in behalf of gender equality linked to the development of the welfare state during the last decades in Spain ( ^{Miller-Moya, 2004} ), which in the medium term could be assimilated by the migrants who integrate in this context of reception.