|Home | About | Journals | Submit | Contact Us | Français|
Because Mexico–U.S. migration represents the largest sustained migratory flow between two nations worldwide, much of the theoretical and empirical work on migration in the Americas has focused on this single case. Yet in the past few decades, migration has emerged as a critical issue across all nations in Latin America and the Caribbean. Indeed, over the past fifteen years, this region has changed its historical position from a net migrant-receiving region to one of the leading sending areas of the world.
In this volume, we offer the first systematic assessment of Latin American migration patterns using ongoing research on the Mexican case as a basis for comparison. We include work by leading scholars of migration who draw on a common source of comparable data. Our specific purpose is to determine whether and how Mexican migration is similar to or different from migration in other countries of the hemisphere. The analyses are comparative and based on data from the Mexican Migration Project (MMP) and the Latin American Migration Project (LAMP), which together offer the most comprehensive and reliable source of data on migration from Latin America and the Caribbean.
Each chapter examines specific propositions or findings derived from the Mexican case that have not yet been tested for other Latin American or Caribbean nations. Work from Mexico has now produced a fairly conventional account of migration and settlement in the United States, but we know very little about how Mexican patterns generalize to other migratory flows in the region. A major shortcoming of prior research is its overreliance on data from just one country.
To the extent that other countries have been studied, most of the research to date has been conducted on a case-by-case basis rather than comparatively. As a result, conclusions about migration trends and patterns in Latin America are derived from a diverse array of studies that make use of divergent data, methods, and theoretical models. In this volume, we seek to remedy this lack of coherence by systematically comparing Mexico to other source countries in the Americas using a common framework of data, methods, and theories. In doing so, we hope to situate findings about Mexico–U.S. migration in the larger context of migration in the Americas and to discover how country-specific characteristics affect patterns and processes of emigration.
Such a comparative approach to the study of migration represents a unique and innovative contribution to scholarship on international migration—a topic of considerable interest in the twenty-first century. The chapters derive from papers originally presented at a conference held at Vanderbilt University in May 2008. With support from Vanderbilt’s Center for the Americas, the conference brought together an interdisciplinary set of scholars from universities throughout the Americas who were instructed to base their analyses, at least in part, on data collected by the MMP and the LAMP in 118 Mexican communities and 35 communities from eight other nations of Latin America and the Caribbean, including Costa Rica, Nicaragua, Peru, Paraguay, Argentina, Puerto Rico, Haiti, and the Dominican Republic. In their comparative breadth, the resulting chapters offer a new perspective on the causes and consequences of migration in the Western Hemisphere. In this introduction, we underscore some of the salient challenges and rewards of doing comparative work on migration.
We begin by describing the structure and organization of the MMP and LAMP, highlighting their origins, developments, and strategies of data collection. These projects, and the important gaps in data they were designed to overcome, explain why research done to date has generally not been comparative and why it has failed to discern which causes and consequences of migration are unique to Mexico and which apply as well to other countries and regions. We conclude by looking to the future to consider how truly comparative investigations might better inform research on migration in the Americas.
Several decades ago, little was known about emigration from Mexico and other nations in the Western Hemisphere; what was known was typically based either on small, nonrandom samples that offered rich data but limited generalizability or on large, national-level probability samples that offered substantial generalizability but few measures of migration, its causes, or its consequences. The MMP was founded in 1982 to remedy this situation. In that year it applied random sampling to survey households in four communities located in the Mexican states of Jalisco and Michoacán, following up shortly thereafter with snowball samples of out-migrants from the same communities located in U.S. destination areas. Since 1987 it has used the same methods to gather data from households in an expanding array of locations and to date has surveyed more than 120 communities in nineteen Mexican states (for a list of communities and states, see http://mmp.opr.princeton.edu, Documentation, Appendix A).
The MMP pioneered the development of a method of data collection known as the ethnosurvey, which combines qualitative interviewing and ethnographic fieldwork with representative survey sampling to generate information on Mexico–U.S. migration that is historically grounded, qualitatively reliable, quantitatively accurate, and interpretively rooted in the context of both receiving and sending areas (Massey 1987). The method uses a semistructured interview schedule that is flexible and unobtrusive, requiring the same information to be gathered from each respondent but leaving question wording and ordering to the judgment of the interviewer, depending on circumstances. Data from the interview are then cross-checked against reports from local informants to ensure their validity, and a separate ethnographic study of the community is conducted to generate an independent base of qualitative information. The end result is high-quality information on the characteristics and behavior of both documented and undocumented Mexican migrants to the United States.
The process of selecting communities for the MMP has traditionally relied on anthropological methods. The first communities selected were located in the states of west-central Mexico, the traditional source region for migration to the United States. From the earliest days of Mexico–U.S. migration, five states in this region have accounted for at least half of all migrants to the United States (Durand, Massey, and Zenteno 2001). In addition to Jalisco and Michoacán, these states include Guanajuato, San Luís Potosí, and Zacatecas. Other states in western Mexico, such as Aguascalientes, Colima, and Nayarit, also have long histories and high rates of U.S. migration, but given their small populations, they do not contribute so many migrants to the outflow.
In choosing communities, investigators did not deliberately seek places known to contain large numbers of U.S. migrants. Nonetheless, the fact that the earliest samples were concentrated in core sending states meant that they generally displayed significant rates of out-migration to the United States. Within each state, at least four communities were selected to represent four levels of urbanization: ranchos (fewer than twenty-five hundred inhabitants), pueblos (twenty-five hundred to ten thousand), cities (ten to one hundred thousand), and metropolitan areas (more than one hundred thousand). In ranchos, pueblos, and smaller cities, a census of dwellings was conducted to generate a frame for random sampling; whereas in larger cities and metropolitan areas, a specific neighborhood was demarcated, and all dwellings within it were enumerated to generate the sampling frame. In all cases, neighborhoods had to include at least twelve hundred enumerated dwellings, from which a random sample of two hundred was taken.
In addition to covering a range of urbanism, investigators also sought to represent diverse patterns of social and economic organization by selecting a range of community types, including mining towns, fishing villages, industrial communities, tourist centers, Amerindian villages, commercial towns, diverse metropolitan service economies, and a variety of agrarian communities. As coverage of Mexico’s migratory heartland grew more complete, the selection of communities was broadened to include newer sending states with more recent histories of out-migration to the United States, including interior states such as Oaxaca, Puebla, and Guerrero as well as border states such as Baja California, Chihuahua, and Nuevo León.
As U.S. migration has spread throughout Mexico over time, even newer sending states, such as Veracruz and Hidalgo, have been added to the sample. The MMP continues to use anthropological criteria for selecting communities in new sending states, but the choices are guided by an index of migration intensity developed for municipalities by Mexico’s National Population Council (CONAPO) using the 2000 census. This index provides a reliable indicator of the level of U.S. migration prevailing at the municipal level and is useful in identifying new communities of origin for U.S. migrants in states where little information has heretofore been available.
These methods yield data with a high degree of representativeness at the community level. Indeed, in several rural ranchos and pueblos, all households were surveyed to eliminate sampling error entirely. The MMP thus offers data from a wide variety of communities to provide a strong platform for comparative study and generalization, with a broad range of sizes, regions, ethnic compositions, and economic bases. Seasonal U.S. migrants generally return home in the winter months, so at least four communities are surveyed each December and January, with a target sample size of two hundred households, except in communities with fewer than five hundred residents, where a smaller sample may be taken. In very small communities, the smaller sample may constitute a 100 percent survey of households. If initial fieldwork reveals that a significant share of migrants come back during months other than December and January, interviewers return to the community during those months to gather a portion of the two hundred interviews.
The community surveys yield solid information about where migrants go in the United States, and during ensuing summer months interviewers travel to U.S. destination areas to gather nonrandom snowball samples of ten to twenty out-migrant households from each community who have settled permanently in the United States and no longer return to Mexico with much frequency. On average, they represent approximately 10 percent of the number surveyed in Mexico. Currently the MMP data set contains 124 communities, with information on more than 140,000 persons.
The LAMP was established as an extension of the MMP to enhance understanding of international migration from other countries in Latin America and the Caribbean. The LAMP began in 1998 with a round of surveys in Puerto Rico and subsequently expanded to the Dominican Republic, Nicaragua, Costa Rica, Haiti, Peru, and Guatemala. In 2007, fieldwork for the study was also conducted in El Salvador and Columbia; and most recently, a modified version of the LAMP survey was implemented in Paraguay to study migration from that country to Argentina. In all cases, LAMP personnel work closely with local researchers and institutions to develop an accurate understanding of local conditions and circumstances and thereby ensure data quality. (For detailed information about the methods used in each country, see the project Web site at http://lamp.opr.princeton.edu.)
For our purposes, the key feature of the two data sets is that they share the same methodology and thus can support rigorous comparative analyses. To the extent possible, sampling strategies, questionnaire design, interviewing, and fieldwork in the LAMP were structured to be the same as in the MMP, though details were often tailored to account for unique circumstances in each setting. In Puerto Rico, for example, questions about legal status were eliminated; and outside of Mexico, information on access to ejidos, that country’s unique system of communal farming and land tenure, was dropped. Despite such minor differences, the two data sets offer social scientists detailed, comparable information that can be used to compare patterns and processes of international migration directly across countries.
Nonetheless, although together the two projects cover eight countries in Latin America and the Caribbean, there are many more communities and households in the MMP than in the LAMP. Whereas no country in the LAMP has more than 10 community surveys, Mexico has 118. As a result, the LAMP offers a more limited basis for analysis and generalization, and smaller sample sizes place certain limits on the accuracy of statistical estimates and the ability to model migratory dynamics across the hemisphere. Despite these inevitable limitations, the MMP and LAMP data nevertheless allow researchers for the first time to build an explicitly comparative focus into their studies, at times with surprising results.
For example, the existence of comparable data on migration across a variety of countries permits a more systematic analysis of how gender influences processes of international migration. Within any single country investigators can do little more than compare the migratory behavior of males and females, but across countries they can study how different gender and family systems condition the migratory decisions of men and women. Latin America offers a wide diversity of gender regimes, ranging from Mexico’s patriarchal family structure to the Dominican Republic’s matrifocal pattern of family organization.
As a result, migration decisions in Mexico are typically made by males with little or no input from women, and the resulting movement is almost always male-led, with women following later for purposes of family reunification (Cerrutti and Massey 2001; Aysa and Massey 2004). In contrast, migration decisions in the Dominican Republic are made by women independently of men, and movement is typically for economic reasons rather than family reunification (Massey, Fischer, and Capoferro 2006). Similarly, Sana and Massey (2005) found that in Mexico, where patriarchal families are cohesive and male migration is often temporary, families receive remittances from migrants abroad and often devote them to productive activities. In the Dominican Republic, however, family instability makes the sending of remittances more hazardous; and when they are received, they are used more for household survival and maintenance than for capital accumulation or business formation.
Although the foregoing results illustrate differences stemming from cross-cultural variation, sometimes migrants from very different countries respond in the same way to common causes. Donato, Aguilera, and Wakabayashi (2005) report, for example, that immigrants from Mexico, the Dominican Republic, and Nicaragua all reported a similar deterioration in U.S. labor market conditions after 1986, when a major piece of restrictive immigration legislation was signed into law. Among immigrants from all three nations, hourly wages fell, cash payments increased, and the likelihood of earning a subminimum wage rose after 1986 compared to the years before.
These findings illustrate what is possible when researchers are able to draw on a common source of comparable data to estimate the same statistical models to study migrant decision-making and migratory outcomes across diverse origin countries. In addition to the MMP and the LAMP, ethnosurvey methods have increasingly been employed in other regions to create comparable data sets on international migration from China (Liang 2001; Liang and Morooka 2004; Liang and Zhang 2004), Senegal (Beauchemin, Diagne, and Lessault 2007), and Poland (Massey, Kalter, and Pren 2008). We strongly believe that further theoretical and substantive progress in understanding international migration must make greater use of comparative research, and we offer this volume as a model and a first step toward advancing that larger agenda.
The chapters in this volume offer a representative sampling of current research in the field of international migration. Broadly understood, this work focuses on four fundamental questions related to the movement of people across borders. The first concerns the individual determinants and basic processes of movement: what individual circumstances lead people to make such a significant decision? Embedded in this question is a related concern about who leaves and who remains behind. Are migrants more or less educated than nonmigrants? Are they more likely to be women? Are they employed or unemployed, professionals or laborers, married or single? Answers to these and other questions are critical to furthering our understanding of the motivations, characteristics, and behavior of migrants.
A second line of inquiry pursued by many scholars entails the related objective of identifying and understanding the larger structural causes that ultimately underlie individual and household decisions to move. Combining data on individual, family, and community-level determinants of migration, this effort seeks to model how macro-level changes in socioeconomic and political organization affect micro-level decision-making and determine the composition of migratory flows around the world.
A third agenda being actively pursued by contemporary migration scholars focuses on the salient consequences that migration has on individuals, households, and communities in sending and receiving nations and, more broadly, how migration can be expected to reshape culture and society at origin and destination. Research under this rubric includes analyses of migrant occupational outcomes, attitudinal changes among both immigrants and natives, the socioeconomic costs and benefits of migration for different classes of people at points of destination, and the likely socioeconomic consequences of migration for those left behind.
A final area of inquiry is the realm of social policy, particularly the effect that governmental attempts to control the quantity and quality of immigrants have on the actual size and composition of the resulting international flows. As the United States is the leading magnet for immigration in the Americas, U.S. policies have had a disproportionate influence on historical patterns of movement throughout the hemisphere. Before 1965, there were no numerical limits on immigration from Latin America to the United States; however, since that date increasingly stringent numerical quotas and qualitative criteria have been applied and more militant border policies adopted. Whatever the intent of these restrictive measures, an open empirical question is their practical influence on the number, origins, and characteristics of immigrants.
We employ these broad areas of inquiry as the organizing framework for this volume. In the first section, each chapter addresses a distinct dimension of the processes that underlie migration patterns across the Americas. Durand and Massey begin by offering a comprehensive historical overview of the region’s major migratory trends, with a particular focus on the watershed transformation of countries in Latin America and the Caribbean into sources rather destinations of international migrants. The authors identify three critical patterns of movement over the past fifty years: south–north migration to the United States and Canada, interregional migration within Latin America, and transoceanic migration to Europe and Japan.
The analysis of Durand and Massey offers a point of departure for the subsequent work of Lindstrom and López Ramírez, who develop a comparative framework to identify the characteristics of the pioneer migrants—those who initiate a migratory process within a community by being the first ones to leave. In the absence of information or experience, the first people to migrate face high costs and unknown risks without much of a social support network. Their analysis suggests that these conditions select for ambitious, risk-tolerant individuals who are less constrained by economic or family obligations than those who migrate later. Pioneer migrants are more likely to move as free and independent agents rather than as followers, a tendency that appears common across countries, whether Mexico, Guatemala, Nicaragua, Costa Rica, or the Dominican Republic.
The subsequent papers by Donato and Cerrutti and Gaudio explore how migrant profiles vary across countries characterized by different socioeconomic and historical contexts. They show how comparative research may be used to identify how and why characteristics such as gender’s effect on migration vary across time and space. Donato examines how the lifetime chances of migration vary by national origin and highlights differences by legal status related to the dramatic growth in the size of the U.S. undocumented population since the late 1980s. Her findings reveal more diversity in contemporary gender-related processes than in the past. For example, the well-known pattern of male-led undocumented migration from Mexico differs sharply from both the pattern of female-led documented migration from the Dominican Republic and the gender-neutral pattern of movement between the island of Puerto Rico and the U.S. mainland.
Cerrutti and Gaudio also compare two migration streams with distinct gender profiles: the male-dominant Mexican migration to the United States and the female-dominant Paraguayan migration to Argentina. Underlying the differences are conditions related to the socioeconomic roles of women in the two sending nations, which in turn affect the extent to which independent moves by women are accepted and promoted in society. In the case of Paraguay, women have played a central role in the economy, and as the peasant economy has declined, more of them have been pushed to migrate both internally and internationally. In addition, conditions in Argentina—especially its strong labor demand for personal service workers—help to explain women’s higher propensity to migrate from Paraguay compared with Mexico.
Taken together, these three chapters not only offer strong confirmation of basic theoretical and empirical findings derived from research on Mexico, but they also highlight how nation-specific factors can shape common underlying processes to yield different patterns and processes of international migration in specific cases. The next section of the volume offers analyses of both the macro- and micro-level causes of recent migration from the Americas and how they determine the socioeconomic and demographic characteristics of the region’s new migrants.
The chapters by Hiskey and Orces and Alvarado and Massey focus on the role that political shocks and violence can play in promoting emigration during times of political transition and economic transformation. These papers highlight the value of comparative research by showing the influence of variables that have largely been absent from research on the determinants of migration from Mexico. Abrupt shifts in a nation’s political economy can trigger migration through economic, political, or social channels; by transforming employment and wage levels; by affecting the quality and quantity of public goods provided by the state; or by influencing the level of violence and insecurity that prevails in daily life.
Fussell further demonstrates the utility of the comparative approach by examining the process of cumulative causation, now well established as a principal feature of Mexico–U.S. migration, across the countries of Nicaragua, the Dominican Republic, Costa Rica, and Puerto Rico as well as Mexico. She finds strong evidence of cumulative causation but with an important addendum: that the contours of cumulative processes are shaped in important and predictable ways by particular country characteristics, including migratory history and type of relationship with the main country of destination.
Finally, Takenaka and Pren examine differences in the education profiles of emigrants from two Latin American countries that lie at different ends of the migrant selectivity continuum: Peru and Mexico. Their analysis reveals large educational differences between Mexican and Peruvian immigrants to the United States, differences that appear to follow from the different stock and distribution of social capital in the two countries, which stem from very different histories of network formation. Widespread access to migrant networks yields flows of poorly educated Mexicans who migrate temporarily as part of a strategy to finance remittances and thereby improve living conditions back home, whereas more limited access to migrant networks in Peru yields an outflow of well-educated Peruvians who mostly emigrate abroad permanently in search of higher incomes and better career prospects.
The third section of the volume considers the consequences of migration for various countries in Latin America and the Caribbean. Four chapters offer important insights into the consequences of migration for sending and receiving societies. Within receiving countries such as the United States, a critical issue is the relative integration of migrants into socioeconomic structures. Of particular importance is the labor market, which ultimately provides the economic resources that immigrants use to finance their mobility strategies at home and abroad. In her chapter, Flores documents sharp differences in the ability of different groups to translate educational achievements at home into occupational achievements in the United States, yielding very different prospects for the long-term integration of Guatemalans, Mexicans, Costa Ricans, and Nicaraguans.
The next three papers consider the effects of the dollar remittances sent home by migrants as well as migrants’ own economic prospects once they return home. Duany analyzes cross-national differences in the amount of money sent home by migrants and highlights how the extent and nature of ties between source and home countries condition the size of the remittance flows as well as the uses to which these “migradollars” are put by the families that receive them.
Amuedo-Dorantes, Georges, and Pozo, for their part, assess the important influence of remittances on human capital formation in sending regions. They examine educational outcomes within migrant and nonmigrant households that receive and do not receive remittances and demonstrate a complex set of compensating effects. Although families left behind may receive money to help to fund children’s schooling, they also typically suffer socially from the absence of a key household member. These consequences are brought to the fore by the comparison of remittance effects on nonmigrant and migrant households in Haiti.
Cobo, Giorguli, and Alba focus on what happens to migrants after they return home to their sending nations, considering their occupational prospects in local labor markets. They find that not all migratory experiences lead to occupational mobility upon return and uncover substantial variation with respect to national origin and age in patterns of mobility. This variability is intriguing but cannot be conclusively characterized without additional research on the mechanisms by which occupational outcomes are achieved.
The last topic taken up by the volume is the effect of U.S. immigration policies on the behavior and characteristics of immigrants. Riosmena offers a revealing analysis of how the distinct legal frameworks faced by migrants from different countries determined the status of later generations of U.S. migrants and ultimately conditioned their economic prospects and social well-being in American society. In their analysis, Massey and Riosmena survey the panoply of restrictive policies undertaken by the United States over the past several decades and document their failure to limit the growth of Latin American immigrant populations. Given widespread access to social capital throughout the Americas, American efforts to limit the number of legal immigrants and increase the costs and risks of entering the United States without authorization have not worked, but have back-fired. Instead of reducing the inflow of immigrants, they perversely have lowered the outflow of migrants back to their countries, leading to an increase in the net rate of immigration. The authors propose basing immigration policies on the new economics of labor migration and social capital theory rather than neoclassical economics.
On the whole, the chapters in this volume clearly show that origin context matters greatly in determining patterns and processes of international migration in the Americas. The volume represents the first step in a much larger research agenda to specify how country and regional characteristics affect migratory decision-making, the characteristics and motivations of migrants, and the composition of the resulting outflows. The advances reported here are only suggestive, however, and to move forward researchers must generate and analyze more data that are comparable across nations. With additional data from communities in Haiti, for example, more elaborate analyses could be carried out to determine more precisely the differences between households that receive and do not receive remittances.
At the same time, we also need to broaden the sample to include more countries with comparable data, especially nations that only recently have begun to participate in the international migratory system emerging in the Americas. Some of these newer sending nations, such as Bolivia and Ecuador, send substantial numbers of migrants to the European Union while also sending workers to wealthier nations in the region, such as Argentina and Chile; and in many cases, these flows are substantially female. Understanding the reasons for cross-country differences in the feminization of migrants and modeling other compositional differences, particularly in comparison with the well-studied case of Mexico, potentially offer scholars new analytic leverage to understand more fully the dynamics of out-migration.
In the end, this volume makes clear that comparative analysis is critical to building an accurate theoretical and substantive understanding of migration. Although cumulative causation is crucial to explaining out-migration from a number of nations in the Americas, the specifics of the cumulative processes often vary sharply by country of origin characteristics. Nonetheless, the characteristics of pioneer migrants appear to be quite similar across national origins. Moreover, despite the fact that the United States ostensibly sponsors an immigration policy that is neutral with respect to national origin, in practice Dominican and Nicaraguan migrants have been much more able to obtain and enter with legal visas than those from other nations in the Americas, raising provocative questions about political and other kinds of biases in U.S. policy. We hope that these and other findings established in this volume will inspire other social scientists to push forward energetically and carry out more comparative research using the MMP, the LAMP, and other available data sets and thereby lead us ultimately to a more robust and generalizable theory of migration.
Katharine M. Donato is a professor and chair of sociology at Vanderbilt University and an editor of American Sociological Review. Her research interests include international migration between Mexico and the United States, social determinants of health, immigrants in the U.S. economy, and ethnic and gender stratification.
Jonathan Hiskey is an associate professor of political science at Vanderbilt University. His recent work focuses on the political impact of migration on sending communities across Latin America. He is the author of numerous articles on these topics in such journals as the American Journal of Political Science, the Latin American Research Review, and the British Journal of Political Science.
Douglas S. Massey is the Henry G. Bryant Professor of Sociology and Public Affairs at Princeton University and the president of the American Academy of Political and Social Science. His research focuses on international migration, race and housing, discrimination, education, urban poverty, and Latin America, especially Mexico. His most recent books are New Faces in New Places: The Changing Geography of American Immigration (Russell Sage Foundation 2008) and Categorically Unequal: The American Stratification System (Russell Sage Foundation 2007).
Jorge Durand is a professor and senior investigator at the Universidad de Guadalajara and adjunct professor at the Universidad Iberoamericana in Mexico City. He has been elected as a foreign associate of the National Academy of Sciences, the American Academy of Arts and Sciences, and the American Philosophical Society. His most recent books are Detras de la Trama: Politicas Migratorias Entre Mexico y Estados Unidos (with Douglas Massey and Nolan Malone; Porrúa 2009) and Mexicanos en Chicago: Diario de campo de Robert Redfield 1924–1925 (with Patricia Arias; Porrúa 2008).