https://dh.scu.edu/exhibits/files/original/90/2935/Mexican_Migration_to_the_United_States_A_Critical_Review.pdf e958007cbc35149ef39b31edfc5966d9 PDF Text Text Mexican Migration to the United States: A Critical Review Author(s): Jorge Durand and Douglas S. Massey Source: Latin American Research Review , 1992, Vol. 27, No. 2 (1992), pp. 3-42 Published by: The Latin American Studies Association Stable URL: https://www.jstor.org/stable/2503748 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms The Latin American Studies Association is collaborating with JSTOR to digitize, preserve and extend access to Latin American Research Review This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE UNITED STATES: A Critical Review Jorge Durand, Universidad de Guadalajara Douglas S. Massey, University of Chicago Although social scientists usually do not speak in terms of laws, they believe they are at least able to make valid empirical generalizations. In studies of Mexican migration to the United States, for example, generalizations drawn from the research literature abound. Thus the problem is not that generalizations are lacking but that they are frequently inconsis- tent and contradictory. Often such inferences are simply invalid because they are based less on evidence than on the investigator's own preconcep- tions. As a result, the field of Mexican migration studies has been plagued by a fragmented debate that seems to go on and on without resolution. The intractability of this debate has been magnified by the inher- ently political nature of migration between two countries with very different cultures, histories, and levels of wealth and also by the fact that much of the movement is clandestine and therefore unobservable. Adding to these problems are the inevitably divergent interests, perceptions, and languages of researchers from north and south of the border, a combination that makes the lack of a clear consensus hardly surprising. We believe nevertheless that the inconsistency of research findings on Mexican migration to the United States is more apparent than real. In an effort to bring some order to this disparate body of research findings, we have undertaken a critical review and synthesis of research carried out in Mexico and the United States. We will begin by examining two particu- larly salient issues-the number of Mexican migrants to the United States and the quantity of their monetary remittances to Mexico-and will then suggest that once rhetoric is separated from fact and analysis from opinion, the various estimates are actually relatively consistent. We will also examine the growing number of case studies of Mexican communities that send migrants to the United States. These investi- gations have yielded apparently contradictory generalizations on a vari- ety of topics: the class composition of U.S. migration, the legal and demographic background of U.S. migrants, the economic effects of emi3 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review gration on the community, the effects of Mexican agrarian reform and agricultural modernization on U.S. migration, and the prevalence of dif- ferent strategies of migration. Most of the generalizations suffer the basic limitation of being based on isolated case studies. We will argue that such apparently inconsistent generalizations about Mexico-U. S. migration are not necessarily contradictory when they are examined in comparative perspective. Rather, diverse outcomes occur in various communities when common processes of migration are shaped and differentiated by structural variables operating at the community level. Because these structural variables exert their influence between communities rather than within them, such variables cannot be studied effectively through individual case studies. They must instead be observed and analyzed comparatively across communities. We have therefore undertaken a systematic and critical review of all the community studies that we uncovered in a survey of the research literature. Our goal is to identify the community-level factors that are most important in determining the expression of basic migratory outcomes. Our review suggests that only a few community factors account for the diversity of conclusions in different case studies: the age of the migration stream; the degree to which productive resources are equitably distributed; the quality of local resources, especially land; the niche in the U.S. industrial structure where the community's migrants first became established; and the geographic, political, and economic position of the com- munity within Mexico. In addition, our review suggests that life-cycle factors-particularly the relative numbers of workers and dependentsexplain much of the heterogeneity in outcomes at the household level. In undertaking this critical review, we are not recommending that community studies should be avoided-quite the contrary. We neverthe- less suggest that researchers should be sensitive to the peculiarities of the communities they are studying, particularly to the influence of the community-level factors identified above, before generalizing from a single case study. Ultimately, however, we argue for a research design that would incorporate the study of many different communities into a common analytic framework. Such a design would be capable of analyzing simultaneously the operation of variables at three levels-the individual, the household, and the community. It would also permit analyzing how community social and economic structures affect migration and how migration affects the social and economic organization of communities. THE NUMBERS GAME: HOW MANY MIGRANTS The debate over how many persons migrate between Mexico and the United States is obviously a key issue, but figures advanced by investigators on opposite sides of the border have always differed sub4 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. stantially. As early as 1930, Manuel Gamio noted the profound disparity between estimates prepared in Mexico and the United States. He sought to rebut the view widely held in the United States that more than a million Mexicans were living in the United States by 1929. He argued that it was impossible to determine the number of Mexican immigrants from the U. S. Census because it was then taken in July, when the number of Mexican migrants was at its peak. Because Mexican migration is highly seasonal, the number of Mexicans enumerated would have been substantially lower if the census had been taken in December (Gamio 1930a; 1930b). Gamio showed that monthly data on remittances declined sharply during the winter months, as the number of U.S. exit forms filed by Mexicans rose (Gamio 1930b). Gamio's contemporary in the United States reached a different conclusion, however. Paul Taylor argued that Mexican migration was largely "permanent" and that the winter decline in remittances was explained by the fact that there was less work during this season and migrants did not have the extra income to send home (1929, 237). Taylor examined national census figures, school statistics, and mar- riage registrations and also undertook special surveys in source and destination areas (P. Taylor 1930, 302-8; Taylor 1931, 48-56; Taylor 1933). This early "numbers debate" was never resolved. It was still being argued when the Great Depression caused the mass deportation of Mexicans from the United States. This event relegated the counting of immigrants to the background, as scholars began to worry instead about the counting of deportees, setting off yet another debate. According to Mercedes Carreras, 311,717 Mexicans were deported from the United States between 1930 and 1933 (Carreras 1974, 145). Ralph Guzman, however, believes that the figure was at least half a million, a number he considers conservative (Guzmain 1979). Abraham Hoffman put the figure at 319,673 between 1930 and 1933 and at 458,039 for the period from 1929 to 1937 (Hoffman 1974). The deportations also called attention to the large number of women migrants. Carreras (1974) reported that among Mexicans deported between 1931 and 1933, some two-thirds were women, a finding that questioned the prevailing stereotype of Mexican migrants as single males traveling alone. After a long hiatus during the depression years of the 1930s, Mexico-U.S. migration was revived in 1942 by the Bracero Program, a temporary worker program initiated by the U.S. government. Although it was begun as a temporary wartime measure, the program was successively extended by the U.S. Congress and did not end until 1964 (see Craig 1971). In contrast to the disputed estimates of the number of immigrants and deportees, widespread agreement exists as to the number of braceros. Because each bracero contract generated its own administrative record, official statistics are generally considered to be reliable by Mexicans as well as U.S. researchers. During the twenty-two years that the 5 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review program lasted, some 4.6 million braceros entered the United States (Morales 1981, 148). During the Bracero Program, however, another international flow arose that was much harder to measure and far more intractable as a subject of study: undocumented migration. From 1942 to 1964, official statistics indicate, nearly 5 million Mexicans were apprehended and deported from the United States, a fact widely reported on both sides of the border (see Samora 1971; Cornelius 1978; Morales 1981). Since 1964 an additional 17 million Mexicans have been arrested and expelled from the United State (U.S. INS 1989). But although observers may agree on the number of apprehensions, they disagree profoundly about the relationship between this figure and the number of illegal migrants. Because undocumented migration is by definition clandestine, it cannot be measured directly, a situation that has forced researchers to rely on indirect measures. The resulting estimates can be characterized as either analytic or speculative (Passel 1986). Analytic estimates are based on quantitative procedures whose assumptions can be scrutinized and whose biases can be systematically evaluated, whereas speculative estimates are based on little more than opinion or conjecture. In general, speculative procedures yield estimates that are much larger than those developed by using analytic methods, and they have therefore received far more attention from the U. S. press and from U. S. politicians. Speculative estimates generally put the number of undocumented Mexicans at 5 million or more. For example, Lesko Associates (1975) as- sembled a panel of 'experts" to guess the number of undocumented Mexicans in the United States in 1975 and averaged panel members' guesses to derive an estimate of 5.2 million undocumented Mexicans. In 1976 the Commissioner of the U.S. Immigration and Naturalization Service (INS), Leonard Chapman, estimated that 10 percent of Mexico's entire population (about 6 million persons) were living in the United States illegally (Chapman 1976). He arrived at this number by adding up "lestimates" provided by INS district offices. Arthur Corwin put the figure at 7 million in 1981 (Corwin 1982), and Edwin Reubens estimated the total net flow of undocumented migrants to be 600,000 in 1978 (Reubens 1980). None of these figures were derived from analytic procedures, however, and they have been widely criticized on both sides of the border (Bustamante 1979; Cornelius 1979; Massey 1981; Passel 1986). These estimates are significantly larger than the ones derived by U. S. demogra- phers using quantitative methods. For example, Frank Bean, Allan King, and Jeffrey Passel estimated that the maximum number of Mexicans that could possibly be in the United States in 1980 was 3.8 million, and most of their estimates ranged from 1.5 to 2.5 million (Bean, King, and Passel 1983). Passel and his colleagues have demonstrated that 1.1 million undocumented Mexicans were counted in the 1980 U.S. Census taken on 6 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. 1 April (Passel and Woodrow 1984; Warren and Passel 1987). Subsequent research has suggested that only one-half to two-thirds of all undocumented Mexicans were enumerated in the 1980 census (Heer and Passel 1985), yielding a range of 1.7 to 2.2 million for the total undocumented Mexican population in the United States in April 1980. Passel's reanalysis of final data from the 1980 Mexican Census yielded a best estimate for 1980 of 1.9 million, far fewer than the 5 to 7 million put forth by more speculative observers (Passel 1985). Analytic methods have also yielded estimates of net flow that are well below the speculative figures of 500,000 to 600,000 per month. For example, Jeffrey Passel and Karen Woodrow estimate that during the early 1980s, the net flow from Mexico was about 200,000 persons per year, a rate that would yield a total U.S. population of 3.1 million undocumented Mexicans by 1986 (Passel and Woodrow 1987). An opportunity to confirm this number independently occurred with the passage of the Immigration Reform and Control Act (IRCA) in 1986. It offered legalization to all undocumented migrants who had entered the United States before 1 January 1982 and a special amnesty for those who had worked for ninety days in agriculture during the year preceding 1 May 1986. In all, some 1.2 million Mexicans applied for the general amnesty, and another 1.1 million sought the special amnesty (Bean, Vernez, and Keely 1989). The total of 2.3 million applicants represents 75 percent of the estimated total popula- tion of undocumented Mexicans in 1986, suggesting that Passel's analytic estimates were close to the mark and that more speculative estimates were far too large. Undocumented migration has also become a salient issue in Mexico, and Mexican investigators have made their own attempts to measure the phenomenon. As in the United States, movement across the international border is not captured well by official statistics (Garcia y Griego 1987), and Mexican investigators, like their U.S. counterparts, have had to rely on indirect methods of estimation. Juan Diez Canedo, for example, tabulated the total number of dollars sent to persons with Spanish surnames in Mexico via checks or money orders drawn on U. S. banks (Diez Canedo 1984). By assuming an average size of remittances (taken from North and Houstoun 1976), Diez Canedo estimated the number of undocumented Mexicans in the United States to be 481,000 in 1975. Other investigators have employed the Mexican Census in attempting to assess the magnitude of undocumented migration. By using data from a census question about foreign residence during the past year and making assumptions about the sex ratio among undocumented migrants, Rodolfo Corona derived a statistical model that permitted estimating the number of migrants who had returned from the United States (Corona 1987). He calculated that 64,000 to 83,000 undocumented Mexicans had lived in the United States for at least six months by June 1979 but had 7 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review returned to Mexico by May 1980. The rather narrow target population limits the utility of these estimates, however. Gustavo Lopez and Sergio Zendejas have concluded that the 1980 Mexican Census offers little basis for accurate estimation (Lopez and Zendejas 1988). Jorge Bustamante approached the problem from a different angle in his 1990 study. Three times a day, he took a photograph of undocumented migrants gathered at two staging points on the international border near Tijuana. He then counted the number of migrants captured on film and computed a daily total of gross undocumented migration that was subsequently tabulated by month. This approach, however, cannot measure the extent of return migration nor can it capture the flow along the entire border, limitations that undermine its utility in specifying the size or rate of change in the undocumented Mexican population. The best source of Mexican data on undocumented migration is a survey carried out during 1978-79 by CENIET (Centro Nacional de Informacion y Estadisticas del Trabajo), a branch of Mexico's Secretaria de Trabajo y Prevision Social. Called the Encuesta Nacional de Emigracion a la Frontera Norte y a los Estados Unidos (ENEFNEU), the survey was commissioned to obtain "basic information that would be representative of the phenomenon in its entirety" (Bustamante 1979, 30). This survey administered a short questionnaire to a stratified prob- ability sample of 62,500 households in 115 localities during December 1978 and January 1979. The questionnaire asked about absent household members age fifteen and over who were working in the United States and about persons age fifteen and over who had worked there during the past five years but had returned to Mexico. The results indicated that 519,300 Mexicans were working in the United States as of January 1979 and another 471,400 had worked there between January 1974 and January 1979 but had returned before the survey was taken. Some 91 percent of the returned migrants reported that they had entered the United States without documents, and the combined population of absentee and returned migrants was 84 percent male (CENIET 1982; Garcia y Griego 1987). These percentages suggest that 472,000 undocumented Mexican workers were residing in the United States in 1978, of whom 443,700 were male. Although these figures appear to conflict with the 1.1 million undocumented Mexicans counted in the 1980 U.S. Census, they are not necessarily contradictory because the ENEFNEU sample excluded two important segments of the undocumented population covered by the U.S. Census estimates: persons under the age of fifteen and adult nonworkers, the latter category being composed primarily of married women. Robert Warren and Jeffrey Passel (1987) estimate that only 44 per- cent of the undocumented Mexicans counted in the U.S. Census were males over the age of fifteen, but the ENEFNEU survey yielded an un8 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. documented population that is 94 percent male, all over the age of fifteen. Moreover, the Mexican survey was carried out in December and January, when the undocumented population is at its minimum, whereas the U. S. Census took place in April, when seasonal migrants begin to return to the United States. Finally, the Mexican survey provided estimates for 1978, whereas the U. S. Census was carried out in 1980. Thus the ENEFNEU survey estimates the number of undocumented Mexican males residing in the United States as of January 1978, while the U.S. Census counts the entire undocumented Mexican population in the United States during April 1980. When these basic differences are taken into account, the two estimates turn out to be fairly close. The ENEFNEU survey indicates that there were 444,000 undocumented Mexican males age fifteen or over living in the United States in 1978, whereas Warren and Passel (1987) estimate that 484,000 such men were counted in the U.S. Census in 1980 (.44 times 1.1 million). Adjusting for underenumeration yields a total estimate of 836,000 undocumented Mexican men age fifteen and over in 1980 (.44 times 1.9 million). Depending on one's assumptions about net undocumented migration between 1978 and 1980, the two estimates either agree closely or differ modestly, but they are not wildly contradictory. If net undocumented Mexican male migration is assumed to be about 200,000 men per year, then the CENIET figure is slightly larger than the U. S. Census figure (844,000 versus 836,000). If net male migration was only 100,000 between 1978 and 1980, then the CENIET estimate is somewhat smaller (644,000 versus 836,000). Manuel Garcia y Griego and Francisco Giner de los Rios updated the CENIET estimates and suggest that by April 1984 there were 758,000 undocumented workers in the United States (1985, 230), a figure again within the range of estimates developed by Passel and his colleagues (assuming that "undocumented workers" refers mainly to men over the age of fifteen). Thus analytic estimates prepared in Mexico and the United States do not contradict one another and are consistently smaller than the speculative estimates put forth in the U.S. media. The U.S. estimates have one weakness not shared by the ENEFNEU data, however. Although Passel and his colleagues could estimate the composition of the undocumented population by age and sex, their indirect methodology did not permit analyzing other migrant characteristics. For this purpose, the CENIET data are considerably more useful because that survey asked returned U.S. migrants a detailed set of questions that enable describing them along a variety of dimensions. Several studies have taken advantage of these data to describe various aspects: the characteristics of undocumented workers (Ranney and Kossoudji 1983), the traits of undocumented women (Kossoudji and Ranney 1984), the process of border crossing (Kossoudji 1990), and the 9 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review process of labor-force adjustment (Kossoudji and Ranney 1986). Lamentably, however, data from the ENEFNEU survey have not been analyzed in depth within Mexico (see the final report, CENIET 1982). Apparently, a change in presidential administrations impeded the project's continuation within the Secretaria de Trabajo y Prevision Social. Yet ten years later, the survey remains the most reliable Mexican source on undocumented migration to the United States (Garcia y Griego 1988, 134). Despite the progress achieved in estimating the number of undocumented migrants, the "numbers game" has once again become a public issue as a result of the 1986 Immigration Reform and Control Act (IRCA). The new debate is focusing on whether IRCA's enforcement provisions have reduced the net flow of undocumented Mexican migrants to the United States. On 15 January 1988, the New York Times published a report by INS Director Alan Nelson, who stated that the number of undocumented entrants had been lowered as a direct consequence of IRCA.1 Three days later, Jorge Bustamante questioned this conclusion in Excelsior because it was based on INS apprehensions during December, when the flow of migrants traditionally diminishes.2 A subsequent analysis by Bean et al. (1990) attempted to control for the seasonality of undocumented migration, economic conditions in Mexico and the United States, and the ongoing legalization process. This study concluded that IRCA had reduced the flow of undocumented migrants, but it has been contradicted by other researchers' data indicating that IRCA has had little or no effect (Cornelius 1989; Massey, Donato, and Liang 1990; Bustamante 1990). Thus the debate on the number of Mexican migrants has continued in one form or another for fifty years, often using the arguments and counterarguments first broached by Manuel Gamio and Paul Taylor in the 1930s. The debate reveals the limitations of social science in studying a subject that is by definition unobservable, and it also underscores the political nature of a process linking two countries with different interests, perceptions, and histories. Indeed, the debate frequently has had little to do with facts. As Frank Bean, Edward Telles, and Lindsay Lowell have noted, "the heated discussions during the [U.S.] Congressional debates of the last few years on immigration legislation have shown that the exagger- ated figures are often still taken seriously" (Bean, Telles, and Lowell 1987, 674). Likewise, voting by U.S. Congressional representatives on immigration issues is determined more by party politics than by a rational assess- ment of interests and effects (Lowell, Bean, and de la Garza 1986). Thus the 'numbers game" often resembles a dialogue among the deaf rather than a reasoned debate about facts. 1. "Administration Calls Law on Aliens Effective," New York Times, 15 Jan. 1988, p. A-32. 2. Jorge Bustamante, "Manipulacion de la Migra," Excelsior, 18 Jan. 1988, pp. A6-7 10 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. "MIGRADOLARS: REMITTANCES TO MEXICO Since Mexican migration to the United States began, investigators have attempted to measure the size and effect of migrants' remittances. At the turn of the twentieth century, money orders coming from the United States attracted the attention of Mexican journalists, who reported in various Mexican newspapers the arrival of large quantities of dollars in rural communities (see Durand 1986, 1988). The first systematic study of remittances was carried out by Manuel Gamio, who compiled information on the origins, destinations, and size of postal money orders sent to Mexico from the United States (Gamio 1930a). These data provided Gamio with a reliable indicator of the main destination areas in the United States and the major Mexican sending communities. Gamio also attempted to infer the number of migrants from the size and distribution of remittances, an undertaking requiring a number of tenuous assumptions that were strongly criticized by his contemporary, Enrique Santibaniez. The latter stated that "we do not doubt the existence of these postal money orders; but it is another thing to attribute them to the savings of Mexican workers, when they could have had other origins" (Santibaniez 1930, 98). Paul Taylor also studied remittances but focused his attention locally, on the town of Arandas in Los Altos, a region of Jalisco, where he counted 7,678 remittances by telegraph or postal money order between 1922 and 1931 (P. Taylor 1933). From the number and denomination of the checks, he computed the total inflow of dollars into the community and used this figure to quantify the economic effect of international migration. This estimate had one weakness, however: Taylor could not determine the number or size of remittances that arrived in private letters. He simply assumed that the remittances he counted represented the total arriving in the forms of telegraphs and money orders. As the dispute between Gamio and Taylor demonstrates, differences concerning the size of remittances frequently follow national lines. In general, Mexican researchers have tended toward sins of omission and U.S. investigators toward sins of commission. Reports from Mexico have systematically understated the quantity of U.S. remittances, and the Mexican government frequently ignores the subject entirely (apparently the current policy). This official reticence has not always been the case, however. In their final reports, Presidents Ruiz Cortines and Miguel Aleman both offered specific estimates of the foreign exchange contrib- uted by migrants (Morales 1981). In the United States, meanwhile, researchers have tended to exaggerate remittances by multiplying a large number of undocumented migrants (often derived by extrapolating ap- prehension figures) by an average remittance derived from some field study (which may or may not be representative). 11 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 0001 Jan 1976 12:34:56 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review A recent example of the ongoing debate is that between Wayne Cornelius and his former student Juan Diez Canedo. Cornelius (1978) estimated that remittances to Mexico totaled more than 3 billion dollars in 1975, but Diez Canedo (1984) put the figure at only 317 million in the same year. Cornelius computed an average size of remittances based on his survey of migrants in Jalisco and then multiplied it by the number of Mexicans he assumed were working seasonally in the United States. Diez Canedo, in contrast, totaled checks or bank orders made out to persons with Spanish surnames in Mexico drawn on U.S. banks. Both estimates are open to criticism: that of Cornelius was made essentially by assumption whereas Diez Canedo's ignored remittances in the form of cash and goods. A more careful but heavily qualified estimate was prepared by Garcia y Griego and Giner de los Rios using data from both countries and information from the CENIET survey (1985, 236). They proposed a figure of 1.8 billion dollars in remittances for 1984. The analysis was replicated in a later article (Garcia y Griego 1988), which stated that "even though the estimates are not exact," they provide 'adequate" information to counter alarmist claims of millions of undocumented migrants entering the United States and billions of dollars leaving the country. Charles Keely and Bao Nga Tran used data tabulated by the International Monetary Fund on "unrequited private transfers" (IMF 1987) to examine trends in remittances for various countries from 1970 to 1985 (Keely and Tran 1989). They argue that these figures understate the funds remitted by migrant workers because they include only money trans- ferred through monitored channels and do not incorporate noncash remittances. The IMF data show that in 1975, 59 million dollars were privately transferred from the United States to Mexico, far less than the amounts reported earlier by Cornelius or Diez Canedo. But the amounts increased substantially during the late 1970s and again after the onset of the economic crisis in 1982. By 1984, private transfers totaled 2.3 billion dollars, exceeding the estimate of 1.8 billion put forth by Garcia y Griego and Giner de los Rios (1985). It is thus clear that migrant remittances represent a significant source of foreign exchange for Mexico, some 2 billion dollars in 1984. COMMUNITY STUDIES: FROM THE SPECIFIC TO THE GENERAL Given the problems encountered in estimating the number of migrants and the size of their remittances, researchers have turned to studying particular Mexican communities as a means of analyzing the social and economic processes that underlie the aggregate statistics. These basic processes are difficult to analyze using large-scale surveys for several reasons. First, processes of migration are developmental and longitu12 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. dinal, and surveys taken at a single point in time cannot capture this inherent dynamism. Second, migration is more than an individual or family decision: it is also a response to larger structural conditions in sending and receiving societies, factors that cannot be studied easily by using surveys of individuals or households. Finally, migrants and their families inevitably make decisions within a local setting. National, state, and regional conditions are important, but their effects are mediated through local social and economic institutions that are difficult to study by using survey data alone. Community studies provide a tool for analyzing the migration process in a way that is historical, developmental, and sensitive to the effects of local conditions as well as to those of the national political economy. Community studies suffer from one serious problem, however: it is difficult to generalize about broader patterns and processes of migration from a single isolated case. Yet this inherent weakness has not stopped most researchers, who find the temptation to generalize too great to resist. Thus the field is replete with general statements about the nature of Mexico-U. S. migration based on the experience of a single community. For example, Joshua Reichert considered his case of Guadalupe, Michoacan, sufficient to infer a "migrant syndrome" characteristic of the entire Mexican central plateau (1981). Similarly, although Cornelius points out that he refers only to his own study (1979), he frequently contrasts his findings with those of other investigators in order to make broader statements. Yet Cornelius does not provide the information needed to know whether the comparisons are appropriate: his sampling methodology, which communities he studied, or the overall sample size. He states only that the survey was carried out in Los Altos, Jalisco. At this juncture, a variety of community studies have been com- pleted and their findings introduced into the research literature. Table 1 lists seventeen case studies carried out within specific Mexican communities, and table 2 lists eight other comparative studies that have con- trasted several communities at once, yielding information on twenty additional settings. Together these studies provide data of varying detail and quality on thirty-seven separate Mexican migrant communities. The authors of these studies have put forth a host of generalizations about the nature of Mexico-U.S. migration. But most investigators have made few attempts to consider the peculiarities of their settings or to situate their studies within the broader research literature. As a result, inconsistent conclusions have proliferated on a variety of topics: which social classes migrate, the demographic and legal composition of the flow, the economic effects of emigration, the effects of Mexico's agrarian reform and agricultural modernization, and the relative importance of different strategies of migration. In the ensuing sections of this article, we will review findings on these issues and will argue that discrepancies between 13 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review TA B L E 1 Community Studies of Mexican Migration to the United States, 1929-1990 Community State Investigator Acuitzio del Canje Michoacan Wiest (1973, 1979, 1984) Aguililla Michoacan Rouse (1989) Arrandas Jalisco P. Taylor (1929, 1931, 1932, 1933) Alvaro Obregon Michoacan Trigueros and Rodriguez (1988) Copandaro Michoacan Rionda (1986, 1988) Gomez Farias Michoacan Lopez (1986a, 1986b, 1988) Guadalajara Jalisco Escobar, Gonzalez, and Roberts (1987), Escobar and Martinez (1990) Guadalupe Michoacan Reichert (1979, 1981, 1982), Reichert and Massey (1979, 1980) Jalostotitlan Jalisco Gonzalez and Escobar (1990) Jaripo Michoacan Fonseca (1988), Fonseca and Moreno (1988) Las Animas Zacatecas Mines (1981, 1984), Mines and de Janvry (1982) Los Altos villages Jalisco Cornelius (1976a, 1976b, 1978) Patzcuaro villages Michoacan Taylor (1986, 1987), Adelman, Taylor, and Vogel (1988), Stark and Taylor (1989) San Jeronimo Oaxaca Stuart and Kearney (1981) San Francisco del Rincon Guanajuato Durand (1989b, 1990) Santa Ines (Municipio Michoacan Fernandez (1988) de Villamar) Villa Guerrero Jalisco Shadow (1979) studies occur frequently because investigators fail to consider the impact of basic structural factors operating at the community level. Which Classes Migrate One of the most widely studied aspects of Mexican migration to the United States is the socioeconomic selectivity of migration-that is, the question of which social classes migrate. Various studies have reported on this issue, and the prevailing wisdom is that migrants come from the lower-middle segments of the income or wealth distribution (see Portes and Rumbaut 1990, 10-11). The rationale for this observation is that the rich have little incentive to migrate whereas the poor lack the resources to cover the costs and risks of a trip to the United States. For example, Gustavo Lopez concluded, based on his study of Gomez Farias, Michoacan, that "Imigration is most common among those owning irrigated land, in comparison with those who possess rain-fed fields; and among those in the first category, the more land they own, the 14 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. TA B L E 2 Comparative Community Studies of Mexican Migration to the United States Community State Investigator Jaral del Progreso Guanajuato Grindle (1988) Tepoztlan Morelos Union de San Antonio Jalisco Villamar Michoacan Gomez Farias Michoacan Cornelius (1990a, 1990b) Las Animas Zacatecas Tlacuitapa Jalisco El Bajio Guanajuato K. Roberts (1982, 1984) Las Huastecas San Luis Potosi Mixteca Baja Oaxaca Valsequillo Puebla Altamira Jalisco Massey, Alarc6n, Durand, Chamitlan Michoacan and Gonzailez (1987) Santiago Jalisco San Marcos (Guadalajara) Jalisco Guadalupe Michoacan Mines and Massey (1985) Las Animas Zacatecas Hecorio Michoacan Dinerman (1982) Ihuatzio Michoacan G6mez Farias Michoacan Goldring (1990) Las Animas Zacatecas Mexicali Baja California Selby and Murphy (1984) Mazatlan Sinaloa Tampico Tamaulipas Queretaro Queretaro San Luis Potosi San Luis Potosi Corralillos Jalisco Orozco(1990) El Refugio Jalisco Los Dolores Jalisco more they migrate" (Lopez 1986b, 574). His facts appear to coincide with those of Kenneth Roberts (1982), who found a connection between com- mercialized agriculture and migration, and with those of Richard Mines, who affirmed that migrants were more likely to own land than nonmigrants (Mines 1981). Likewise, Ina Dinerman reports that landowner- ship was more common among migrants in Huecorio than among nonmigrants (1982). Despite these studies confirming the prevailing wisdom, other investigators suggest considerable diversity in the class background of 15 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review U.S. migrants. Cornelius reports that in Los Altos, Jalisco, "those who migrate illegally are among the poorest residents of the community- especially the sons of landless workers and ejidatarios" (Cornelius 1976b, 7). He nevertheless allows that "those at the very bottom of the local income distribution are not likely to migrate because they lack . . . the resources to cover the costs of transportation and the fees charged by the coyotes... ." Likewise, although Raymond Wiest found that the very poorest households were unlikely to support a migrant, he could never- theless "state unequivocally from domestic group histories that nearly all the households involved in migration to the United States had incomes well below the median before sending members to the United States" (Wiest 1973, 197). Other researchers have located U.S. migration even more decisively among the poor and landless. In Guadalupe, Michoacan, Reichert (1979) argues, migration arose among the landless precisely because they had no other option but to migrate. James Stuart and Michael Kearney called migration "the only alternative to starvation" in San Jeronimo, Oaxaca, where 90 percent of the families did not possess enough land to support a family (Stuart and Kearney 1981). In two other rural communities (Altamira, Jalisco, and Chamitlan, Michoacan), U.S. migration was most prevalent among landless day laborers (Massey et al. 1987). Just to complicate matters further, one study reported finding no relationship between either income or wealth and the likelihood of U.S. migration among villagers living near Patzcuaro, Michoacan (E. Taylor 1986). Relatively few studies have examined international migration emanating from urban areas, and their results are also inconsistent. Agustin Escobar and Maria Martinez indicate that U. S. migration is quite common among manual workers in Guadalajara: 17 percent of those in their sample had worked in the United States, but the percentage was higher (up to 25 percent) among workers in small firms, suggesting that emigration was most common among the poorest and least stable segments of the working class (Escobar and Martinez 1990). In contrast, Henry Selby and Arthur Murphy surveyed five Mexican cities and found that migrants consistently came from households that were better off than others (Selby and Murphy 1984). Two other surveys carried out in Guadalajara and Santiago (a nearby factory town) found that U.S. migrants were concentrated among skilled manual and service workers rather than among unskilled laborers (Massey et al. 1987). Thus depending on which study one consults, U.S. migrants are drawn either from the landless or the landed, from the skilled working class or unskilled laborers, from the stable middle class or the poorest segments of society. One study has even suggested that class plays no role at all. It is therefore difficult to make sense of these diverse findings without knowing more about the communities involved because factors 16 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. operating at the community level play a large role in determining which classes migrate and when. Several of the studies listed in table 2 have undertaken a comparative analysis of communities in an effort to determine which factors affect the socioeconomic selectivity of international migration. These studies sug- gest that class background is not a fixed characteristic of migrants that invites simple generalization. Rather, the class composition of migration at any point in time is a variable determined by two key community characteristics: the age of the migration stream and the degree of inequality in the distribution of productive resources. The first factor recognizes that migration is a dynamic, developmental process in which decisions made by migrants at one point affect the course and selectivity of migration in later periods. For example, in their comparative study of four migrant communities, Massey et al. (1987) documented that U. S. migration generally begins in the middle segments of the local occupational or wealth distribution (consistent with the common wisdom) but that over time it becomes progressively less selective and eventually incorporates all segments of the socioeconomic hierarchy Ultimately, the flow becomes dominated by the poorest class of landless campesinos and unskilled workers rather than by the middle class of small property owners or skilled laborers. A similar conclusion was reached by Mines and Massey (1985) in comparing Guadalupe, Michoacan, with Las Animas, Zacatecas. The changing selectivity of migration results from the growth and elaboration of migrant networks, which are composed of ties of kinship, friendship, and paisanaje (shared community origin) between migrants and nonmigrants located in the United States and Mexico. The first migrants who leave for the United States have no social ties to draw on, and for them migration is very costly and risky, especially if they have no legal documents. As Cornelius (1976b) and others have argued, these costs and risks tend to exclude the poorest segments of society. After the first migrants have arrived in the United States, however, the costs of migration are substantially lowered for friends and relatives living in the same community of origin. Because of the nature of kinship and friendship structures, each new migrant creates a set of individuals with social ties to the United States and its labor market. Migrants are invariably linked to nonmigrants through bonds of kinship and friendship, and the latter draw on obligations implicit in these relationships to gain access to employment and assistance at the point of destination, thus reducing their costs substantially. Once the number of network connections in an origin area reaches a critical threshold, migration becomes self-perpetuating in creating the social structure needed to sustain it. Every new migrant reduces the costs of subsequent migration for a set of friends and relatives. Some of those 17 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review left behind are induced to migrate, which further expands the set of persons with ties abroad, which in turn reduces the costs and risks for a new set, causing some of them to migrate, and so on. Over time, migration becomes progressively less class-selective and ultimately a mass movement. Thus depending on when migration began and how developed the networks have become, the class composition of migration from a community may be dominated by the middle class, the poor, the landless, or the landed. That is to say, the effect of migrant networks is quite powerful and ultimately comes to dominate class factors in predicting migration, as has been demonstrated by quantitative studies carried out at community and national levels (Massey 1987; Massey and Garcia EspaAa 1987). The reason that Edward Taylor found no relationship between economic class and migration in his 1986 study was because he controlled for the effect of networks, which dominated all other variables in his statistical models. The age of the migration stream is important because it indicates the maturity of migrant networks but also because widespread migration has the power to transform the class distribution itself. In rural com- munities, money earned through U.S. wage labor can finance acquiring farmland, thus enabling migrant families to move from the class of landless workers to small landowners. If enough families acquire land through migration, it may appear at some point that migration results from landownership. It is therefore crucial to sort out the temporal sequence of migration and class membership, given that U.S. migration is a primary avenue of social mobility in Mexican sending communities. Unfortunately, however, when reporting a positive association between landownership and U.S. migration, many studies do not undertake this task. For example, in the two Michoacan communities of Gomez Farias and Huecorio, Lopez (1986a) and Dinerman (1982) report an apparent correlation between landownership and U. S. migration and conclude that the former causes the latter. It is nonetheless likely that the order of causality is reversed. Studies done in two neighboring Michoacan communities, Guadalupe and Chamitlan, showed that migration often preceded landownership, the likelihood of landownership rose sharply as migrant experience increased, and it was tied directly to U.S. earnings (Reichert 1981, 1982; Massey et al. 1987). The association between migration and landownership is also affected by the degree to which land is available for purchase. This linkage illustrates the effect of a second community-level factor: the structure and distribution of local productive resources. In communities where land is unequally distributed and held in large tracts by a few landowners, the possibilities for acquiring farmland are limited, reducing the likelihood of class mobility through migration (Reichert 1981; Grindle 1988; Gonzalez and Escobar 1990). Moreover, even at the outset of migration, class selec18 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. tivity is strongly shaped by land-tenure arrangements. For example, Reichert's town of Guadalupe was made up entirely of landless families when recruitment for the Bracero Program began in 1942, thereby guaranteeing that all migrants were landless day laborers (Reichert 1979). Luin Goldring has compared Las Animas, Zacatecas, and Gomez Farias, Michoacan, to illustrate the important role played by land tenure in determining the class composition of migrants (Goldring 1990). In Las Animas, land from a nearby hacienda was gradually sold to townspeople beginning at an early date, creating a property distribution dominated by small landowners. In Gomez Farias, however, land remained under the control of the hacienda of Guaracha until the late 1930s. Although the agrarian reform movement eventually provided small plots to a minority of households, they were generally insufficient to support families and were quickly leased to absentee landlords in nearby cities (Cornelius 1990a). Thus when the United States began recruiting migrants during the 1940s, these two Mexican communities sent very different classes of workers, not because processes of migration differed in the two communities but because common processes of recruitment drew on divergent class structures. The Demographic and Legal Composition of Migration Another common topic of generalization concerns the age, sex, and legal composition of Mexican migrants to the United States. Studies carried out during the 1920s revealed that migrants were demographically diverse, incorporating not only single men but women, children, and frequently entire families (Gamio 1930a, 1931; P. Taylor 1932; Carreras 1974). This heterogeneous pattern was effaced by the Great Depression and the mass deportations that followed. When the migration stream began again under the aegis of the Bracero Program, it was renewed principally as a male phenomenon. Very few braceros were female (Hancock 1959; Galarza 1964; Samora 1971). The prevailing view about contemporary patterns was established in the mid-1970s by Cornelius (1976a, 1976b), based on his study of rural communities in Los Altos, Jalisco. He found that U.S. migrants were overwhelmingly males of working age. Although most were single when they migrated for the first time, the migrant population generally was made up mostly of married men who traveled without their wives and children. The vast majority were undocumented. This view of Mexican migration has apparently been confirmed by studies of migrants apprehended by the INS, which reveal a great preponderance of men among Mexicans (Samora 1971; North and Houstoun 1976). This profile persisted remarkably among politicians, researchers, and the public. Thus when CENIET designed its 1978 national survey of 19 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review emigration, it did not ask households about persons in the United States but about workers age fifteen and over (CENIET 1982). The survey was therefore guaranteed to confirm the stereotype of Mexican migrants as males of working age, despite the fact that the undocumented population counted in the 1980 U.S. Census was 45 percent female and 21 percent under the age of fifteen (Warren and Passel 1987). Community studies reveal great diversity in the demographic and legal background of migrants, although enough individuals adhere to the classic stereotype of undocumented male workers to confirm its basis in fact. Field studies carried out in several rural communities (Acuitzio, Michoacan; San Jeronimo, Oaxaca; Huecorio, Michoacan; and Alvaro Obregon, Michoacan) report that virtually all U.S. migrants are males of working age who go without documents (see Wiest 1973; Stuart and Kearney 1981; Dinerman 1982; Trigueros and Rodriguez 1988). Yet other investigations of different communities (Guadalupe, Michoacan; Las Animas, Zacatecas; Altamira, Jalisco; Chamitlan, Michoacan; Santiago, Jalisco; Santa Ines, Michoacan; Tlacuitapa, Jalisco; Jalostotitlan, Jalisco; and Gomez Farias, Michoacan) report substantial participation by women and children and large numbers of legal migrants (see Reichert and Massey 1979; Mines 1981; Massey et al. 1987; Fernandez 1988; Cornelius 1990a; Gonzalez and Escobar 1990; Goldring 1990). This contradiction stems from the effect of two key factors: the age of the community's migration stream (which determines the maturity of the networks and the particular U.S. immigration policies to which migrants were exposed) and the niche in the U.S. occupational structure in which the community's migrants first became established. The operation of these two community-level factors create from a common process of labor migration a multiplicity of demographic and legal compositions. Communities sending large numbers of women and children in- variably have long histories of migration. When migrants are grouped according to the date they left on their first U.S. trip, the participation of women and children typically grows over time (see Reichert and Massey 1980; Massey, Donato, and Liang 1990; Fonseca and Moreno 1988; Gon- zalez and Escobar 1990). The first migrants are invariably men, and for a long time the flow continues to be dominated by males visiting the United States temporarily for limited periods of wage labor. At this stage, migration is dominated by economic motivations and imperatives. Over time, however, social processes come into play. As migrants acquire increasing experience in the United States, gain familiarity with U.S. employers and settings, and expand their network connections to other migrants, the costs and risks of migration progressively drop. Al- though men are at first reluctant to expose their wives and children to the hazards and hardships of migration, the life of a solitary migrant worker eventually becomes difficult to sustain. As the costs and risks drop with 20 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. the expansion of the networks, men increasingly bring their wives and children into the migratory process and the demographic base of migration broadens. The incorporation of women and children often happens rapidly. In Guadalupe, Michoacan, for example, virtually no women or children participated in U.S. migration during the twenty-five years from 1940 to 1965. But in the next fifteen years, the flow came to be dominated by women and children, and in the most recent period (1975-1978), 53 percent of new migrants were women and 67 percent were younger than fifteen (Reichert and Massey 1980). Thus in communities with a long migratory tradition, the average age of migrants tends to fall over time and the proportion of women rises (Massey, Donato, and Liang 1990; Gonzalez and Escobar 1990). The emergence of family migration is facilitated by another process in the development of migrant networks: the formation of branch communities in the United States. As migrants increasingly specialize in U.S. labor to the exclusion of other economic pursuits, the constant shuttling back and forth becomes more difficult to sustain, and some migrants ultimately settle in the United States, typically in urban areas. The settlement of other migrants from the same Mexican town greatly strengthens the networks and focuses migration increasingly on specific destinations, making subsequent settlement by additional townspeople more likely and considerably easier. This process provides a secure, Spanish-speaking environment in which dependents can adjust and adapt to life in the United States. The age of the migration stream also determines one more factor that influences the demographic and legal composition of the flow: the rules and regulations governing the acquisition of legal papers in the United States. In general, U.S. immigration policy has evolved in the direction of making it increasingly difficult for Mexicans to acquire legal status in the United States (Jasso and Rosenzweig 1990). Prior to 1965, Mexican immigration was not restricted numerically, and until 1976, Mexico had no specific quota, being subject only to the overall hemispheric ceiling of 120,000 immigrants. After 1976 all nations in the Western Hemisphere, including Mexico, were placed under a quota of 20,000 immigrants each (exempting immediate relatives of U.S. citizens). Then in 1978, the age at which a child could sponsor the immigration of his or her parents was raised to twenty-one, and hemispheric quotas were abolished in favor of a single worldwide ceiling of 290,000. In 1980 the ceiling was reduced to 270,000. All these changes reduced the number of U.S. immigrant visas available to Mexicans. The increasing restrictiveness of U. S. immigration policy means that communities with older migration streams are more likely to be dominated by legal migrants because applications for visas were made 21 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review under more liberal policies. In general, communities that began sending migrants during the bracero years (1942-1964) are characterized by flows that now contain a high proportion of legal migrants (Reichert 1979; Mines 1981; L6pez 1986a; Massey et al. 1987). When it became clear that the Bracero Program was going to end, U.S. employers began to actively sponsor legalization of their workers under the liberal provisions prevailing in U.S. immigration before 1965 (Reichert 1979; Mines 1982). During the late 1960s and early 1970s, much of the legal migration that occurred consisted of wives and children of former braceros and undocumented migrants who had received their documents in the 1950s and early 1960s. Until the IRCA amnesty was declared in 1986, flows from communities that began to send migrants during the 1970s were domi- nated by undocumented migrants, and the later a community's entry into the migration process, the more its flow was dominated by working-age men. Early investigations indicate, however, that the IRCA legalization program may have far-reaching effects in changing this situation. During a brief period, IRCA legalized the status of more than 2 million formerly undocumented migrants (Bean, Vernez, and Keely 1989). One community study indicates that this amnesty caused a pronounced shift in the legal composition of the flow to the United States (Cornelius 1989), transforming it from a stream dominated by illegals into one composed mainly of legals and rodinos, as those receiving amnesty are popularly called (being beneficiaries of the Simpson-Rodino Law cosponsored by Senator Alan Simpson of Montana and Representative Peter Rodino of New Jersey) (Durand 1989a). Although it is not known how many women and children were legalized under the terms of the amnesty, another field study suggests that rodinos have been overwhelmingly male (Goldring 1990). But even if they are mainly men, the amnesty will inevitably be followed by a wave of subsequent legalizations as wives, children, and other relatives use the family reunification provi- sions of U.S. immigration law to obtain their own documents. Thus the composition of migration to the United States is critically determined by the age of the migration stream. Communities that began sending migrants early now contain large numbers of women and children and a high proportion of legal migrants. Yet even among communities that began participating in U.S. migration at the same time, flows can display substantial differences in demographic and legal composition. For example, although migration began at about the same time in Las Animas, Gomez Farias, Guadalupe, and Tlacuitapa, only 31 percent of Animenos between the ages of fifteen and sixty-four are legal migrants whereas 42 percent of Guadalupenos possess documents (Mines and Massey 1985). Similarly, although 35 percent of all migrants are female in Gomez Farias, only 23 percent of U. S. migrants in Las Animas are women (Goldring 1990). And whereas 57 percent of all women between the ages 22 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. of twenty-five and twenty-nine have migrated to the United States in Gomez Farias, only 38 percent have done so in Tlacuitapa (Cornelius 1990a). These differences among communities with similar histories of U.S. migration suggest that other community-level factors come into play. Goldring suggests, based on her comparative analysis of Las Animas and Gomez Farias (1990), that one of the most important is the location of migrants within the U.S. occupational-industrial structure. The centrality of this factor is frequently overlooked by field researchers, who tend to focus on conditions in Mexico and ignore those in the United States (see, for example, the studies by Wiest 1973, 1979; Stuart and Kearney 1981; Dinerman 1982; E. Taylor 1987, 1988; Trigueros and Rodriguez 1988; Fernandez 1988; Escobar and Martinez 1990). In contrast, the earliest studies of Mexico-U.S. migration clearly recognized the importance of U.S. social and economic conditions and gathered data on both sides of the border (see P. Taylor 1929, 1930; Gamio 1930a, 1930b; Fabila 1929). Santibaniez stated, "To treat with any seriousness the issue of Mexican migration to the United States, it is absolutely necessary to consider it in its bilateral condition, since we Mexicans cannot understand it unless we study it in context and take into account U.S. interests. Likewise, the United States cannot determine anything more than what is convenient to its interests if it does not take account of the circumstances that have fatally determined the human current that leaves Mexico . . ." (Santibaniez 1930, 16). In considering the effect of U.S. employment on the demographic and legal status of migrants, the crucial distinction lies between agri- cultural and urban occupations, with the former leading to legalization and participation by women and children and the latter being associated with undocumented migration by working-age men. To a large degree, this choice is a matter of chance-of being in the right place at the right time. But once a few migrants from a community become established in a particular occupational-industrial niche, they acquire the ability to obtain jobs for friends, relatives, and other townspeople. Over time, therefore, migration from particular communities tends to focus not only on specific geographic destinations but on particular niches in the U.S. occupationalindustrial structure: busboys and dishwashers in a certain restaurant; janitors in a certain building; operatives in a particular factory; workers in a certain car wash; fruit pickers for particular growers. For example, migrants from Santiago, Jalisco, went to a specific neighborhood in Los Angeles because one townsman became the union representative in a lamp factory there. Migrants from Chamitlan, Michoacan, focused on two other California cities where a townsperson had obtained the position of foreman on a ranch and a paisano had became headwaiter in a restaurant (Massey et al. 1987). 23 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review Goldring (1990) compares the cases of Gomez Farias and Las Animas to demonstrate the role played by the U.S. occupational niche in shaping the composition of the migrant streams. Around 90 percent of migrants from Gomez Farias work in the strawberry fields in Watsonville, California, living mainly in one labor camp, because several Gomenios developed early connections with particular growers based in that city (Lopez 1986a; Cornelius 1990a, 1990b; Goldring 1990). In contrast, migrants from Las Animas go to urban destinations because one Animeino owns a construction business in Orange County and several others have acquired stable jobs and settled in South San Francisco and Los Angeles (Mines 1981; Cornelius 1990a, 1990b). As a result of these occupational differences, the migrant stream is primarily legal and family-oriented in Gomez Farias but is diverse and weighted more toward undocumented males in Las Animas (Goldring 1990). For various reasons, agricultural labor is more conducive to a pattern of migration by families than is urban employment. First, farm labor has historically provided a more secure path to legal status, which is then passed on to family members to enable their migration without risk. Over the years, growers have worked hard to ensure the continued flow of Mexican labor, often by arranging for legalization of their farm workers. Growers were responsible for initiating and perpetuating the Bracero Program, which gave migrants legal work permits. When this program ended, growers actively sought permanent resident status for their work- ers. As legal status became increasingly difficult to acquire, growers fought to maintain the "Texas Proviso," which exempted them from prosecution for hiring illegal workers. When growers lost this battle in 1986, they successfully lobbied the U.S. Congress for a "Special Agricultural Worker" amnesty that was made available to farm workers on extremely generous terms. Through the growers' efforts, IRCA also arranged for a "Replenishment Agricultural Worker" program to admit additional workers "to prevent farm labor shortages" (Martin 1990). As a result of these efforts, farm workers have consistently been more able than urban workers to obtain legal status for themselves and their families, making seasonal family migration an attractive and viable strategy The diversity of farm work also provides a range of light and heavy tasks that all family members can undertake to earn income. Moreover, growers often provide temporary housing at very low cost (sometimes even free), lowering the costs of family migration. Thus family migration for U.S. farm labor represents a rational strategy for maximiz- ing family income (Reichert 1979). In virtually all cases where a pattern of "legal shuttle" or "recurrent" migration has emerged, workers are employed primarily in agriculture (Reichert 1979; Mines 1981; Lopez 1986a). In contrast, urban employment discourages, although it does not prevent, legalization and family migration. Urban employers generally do 24 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. not act collectively to ensure their labor supply and have not made intensive efforts to secure legalization for their workers. Rather, they rely on the self-reinforcing nature of network migration to provide them with undocumented workers. In general, therefore, it has been significantly more difficult for urban-based migrants to acquire legal papers. Urban workers never had a Bracero Program, and the general amnesty program was far more restrictive in its criteria of eligibility than the special agricultural worker program. Consequently, many families who have resided for years in urban areas of the United States are ineligible for amnesty (Chavez, Flores, and Lopez-Garza 1990), while farm workers who only recently crossed the border have qualified, often with credentials of dubious validity (Martin 1990). Undocumented status, in turn, discourages family migration because of the hazards and costs associated with crossing the border and living clandestinely in the United States. Aside from such legal problems, urban employment discourages the migration of women and children in other ways. In most entry-level urban jobs, wages are low and city rents are expensive. Housing is rarely, if ever, subsidized or arranged by the employer. Moreover, in urban areas children are less able to work and contribute to family income. Thus the migration of wives and children, rather than maximizing family income, tends to lower net earnings and make it harder to accumulate savings quickly (Massey et al. 1987). In the long run, however, solitary migration is difficult to sustain, and as the length of stay and the number of trips increase, family migration eventually occurs, leading to greater diversity in the composition of migrant streams directed to urban areas. The Economic Effects of Emigration Probably the most widely studied aspect of migration to the United States is its effect on local economic development within Mexico. Community studies are remarkably unanimous in condemning international migration as a palliative that improves the material well-being of particular families without leading to sustained economic growth within migrant communities. Thus Reichert refers to migration as an illness or "syndrome" whereby migration undermines local development. Wiest (1979) calls migration an "addiction" that townspeople must satisfy to fulfill their need for consumer goods and an elevated standard of living. Stuart and Kearney (1981) consider U.S. migration to reflect a "dangerous dependence." The description of emigration and its effects in these ominous terms stems largely from the finding that U.S. earnings are spent overwhelmingly on current consumption, leaving little money for productive investment. Community studies consistently report that U. S. remittances 25 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review and savings are directed to a few nonproductive ends: family maintenance and health; the purchase, construction, or remodeling of homes; and the purchase of consumer goods. For example, in Huecorio, Michoacan, Dinerman (1982) found that 67 percent of U.S. income went to one of these ends. In Jalostotitlan, Jalisco, the figure was 93 percent (Gonzalez and Escobar 1990). Similarly, in the three communities studied by Cornelius, 92 percent of remittances and 66 percent of savings were spent in these ways (1990a, 1990b); and Lopez reports that 83 percent of U.S. earnings went to family maintenance, housing, health care, or parties (1986a). Although additional studies do not report specific figures, similar spending patterns were reported by Shadow (1979), Reichert (1981), Stuart and Kearney (1981), Mines (1984), Fernaindez (1988), and Wiest (1979, 1984). As a result, most observers have concluded with Reichert that "although out-migration has generated higher per capita income and increased rates of consumption, it has not led to the development of the town economy in ways that have stimulated production or generated new employment opportunities" (Reichert 1981, 63). Richard Mines and Alain De Janvry also point out that in rural areas, successful migrants "buy up village lands and businesses when they become available, [but] they have little incentive to invest time or money in improving their investments since a semi-skilled job in the U.S. provides much more income . (Mines and De Janvry 1982, 452). Despite the apparent unanimity of opinion about the ill effects of migration to the United States, several studies provide counter-examples that question the mainstream view. In the town of Alvaro Obregon, for example, Paz Trigueros and Javier Rodriguez report that 30 percent of U. S. earnings were spent on land, tools, or livestock (Trigueros and Rodriguez 1988). Escobar and Martinez (1990) surveyed manual workers in Guadala- jara and found that 31 percent of migrants used their savings to set up a business. In their survey of one neighborhood in Guadalajara, Massey and his colleagues found that 21 percent of migrants used their savings productively and another 10 percent spent their savings on something that could be considered productive under some circumstances (Massey et al. 1987). In addition, several community studies report that migrant earnings have been crucial in community efforts to improve local infrastruc- ture, providing a key infusion of capital to build roads, install electricity, extend sewer lines, provide potable water, and restore public buildings. For example, even as Reichert (1981) lamented the "migrant syndrome" in Guadalupe, Michoacan, he reported that contributions by legal migrants accounted for 84 percent of all funds contributed by townspeople toward six capital improvement projects. Likewise, in Chamitlan, Michoacan, migrant earnings represented a "significant portion" of private contribu26 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. tions to recent public works (Massey et al. 1987). Goldring indicates that townspeople in Las Animas and Gomez Farias have generally been in a better position to make capital improvements than inhabitants of surrounding communities because of the high concentration of U.S. migrants, who are expected to contribute dollars for community improve- ment projects (Goldring 1990). Thus although most studies find a lack of development and high consumer spending resulting from U.S. migration, a few report productive investment and local economic growth. This discrepancy suggests caution in accepting generalizations about a "migrant syndrome" of dependency without considering the characteristics of the communities involved. In fact, most places that have been studied exhibit a geographic location and structural position in the Mexican political economy that are unsuitable for productive investment. Typically, the sending communities are isolated rural villages that are removed from natural markets; they frequently lack basic infrastructure like paved roads, electricity, telephones, running water, and sewage; and they often lack ready access to an appropriate labor force. Even in the realm of agriculture, many communities are poorly suited for investment because of poor quality land, lack of water, a fragmented land tenure system, and highly unequal prop- erty distribution. Under these circumstances, productive investment is not only unlikely but unwise because the chances of business failure are extremely high. Even so, most communities contain some business enterprises, and when the effect of U.S. migration is examined from the viewpoint of businesses rather than from the perspective of migrants, U.S. earnings are found to play a much more important role. For example, among businesses that Cornelius identified in his survey of three rural communities, 63 percent were owned by migrants and 61 percent were founded with U.S. earnings (1990a, 1990b). Of 19 enterprises in Guadalajara studied in detail by Escobar and Martinez, 32 percent were owned by migrants and 10 percent were capitalized with U.S. earnings (1990). The Massey team uncovered 162 business ventures in their survey of four communities, of which 39 percent were owned by migrants and 12 percent were established with U.S. earnings. Migrant-owned businesses accounted for 39 percent of business employment in these communities (Massey et al. 1987). Rather than concluding that migration inevitably leads to depen- dency and a lack of development, it is more appropriate to ask why productive investment occurs in some communities and not in others. In general, a perusal of the above communities suggests that the highest levels of business formation and investment occur in urban communities, rural communities with access to urban markets, or rural communities with favorable agricultural conditions. The effect of such structural factors 27 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review cannot be studied in the isolated case but can only be examined across a range of different communities. Douglas Massey and Laurence Basem developed a multivariate statistical model to analyze investment in four Mexican communities (Massey and Basem n.d.). They found that one of the strongest determinants of productive investment was the community to which one belonged, although the limited number of communities prevented them from identifying which community factors were specifi- cally responsible for differences in the tendency to invest productively. In addition to neglecting the effect of structural factors operating at the community level, studies of spending patterns generally do not take into account the life-cycle stage of the households involved. Active U.S. migrants are heavily concentrated in the age group of twenty to thirtyfour years of age, a time when most people are marrying, forming families, and raising children. During this phase of the life cycle, demands for family maintenance, housing, and medical care are greatest, and it is not surprising that migrants channel most of their earnings into current consumption. As migrants progress through the life cycle, however, spending on consumer goods falls off sharply and productive investment rises (Massey et al. 1987). Moreover, as families age and their migrant experience grows, they become increasingly likely to invest in agricultural inputs that raise productivity, such as machinery, fertilizers, insecticides, and improved seeds (Massey et al. 1987). Finally, patterns of migrant spending also depend in part on historical conditions that present different opportunities for investment in different eras (Durand 1988). The purchase of land was feasible up through the 1930s but became increasingly difficult thereafter, as private land was increasingly redistributed as state-sponsored ejidos. Competition for the remaining private parcels brought about a rapid inflation of prices that precluded further investment in land for most migrants. During the 1940s and 1950s, those who had received land under the government's redistribution program had the opportunity to invest in inputs such as machinery and fertilizers to make their parcels produce, but this avenue of investment too was eventually exhausted. At last, during the 1970s and 1980s, the extension of urban infrastructure and services to rural areas and the increasing congestion of urban metropolises made industrial investment attractive in the countryside and initiated a new wave of rural industrialization (Arias 1988; Durand 1989b, 1990). Thus although most observers have concluded from Mexican com- munity studies that U.S. migration leads to dependency and a lack of development, we argue that this conclusion does not account adequately for the structural, life-cycle, and historical factors that operate to constrain individual patterns of spending and investment. A careful look at the evidence suggests not that a "migrant syndrome" or an "addiction" exists but that under structural circumstances where productive investments are 28 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. likely to be met with success and during phases of the life cycle when current needs are at a minimum, spending for production is quite likely An important task for future research is to identify more specifically those community and household characteristics that are conducive to the productive use of income earned abroad. The Effect of the Reparto Agrario and Agricultural Modernization During the regime of President Lazaro Cardenas in the 1930s, Mexico embarked on an ambitious program of land reform and redistri- bution known as the Reparto Agrario (Yates 1981, 140-43). During this period, millions of hectares formerly held by wealthy landowners in large haciendas were broken into smaller plots and given over to the campe- sinos who traditionally had worked the lands. This fundamental shift in the political and economic organization of Mexico touched virtually all rural communities, and most historical case studies have considered the effects of the agrarian reform movement on U.S. migration. In general, the prevailing view is that the Reparto Agrario dis- couraged migration for several reasons: faced with the prospect of acquiring land, potential migrants stayed to fight for their allocations; the rules of the Reparto Agrario required that land remain in production, forcing campesinos to stay and care for it; and the acquisition of land provided a means of support to rural families that obviated the need for international migration. Consistent with these reasons, several studies indicate that the Reparto Agrario indeed acted as a brake on migration abroad (Lopez 1986a; Fonseca 1988; Fonseca and Moreno 1988). A broader survey of studies, however, reveals that agrarian reform had diverse effects in different communities, depending upon local social, economic, and political conditions. In the towns of Altamira, Jalisco, and Chamitlan, Michoacan, for example, the Reparto Agrario provided an incentive to migrate rather than the reverse. In these communities, campesinos were given land but not the capital or credit they needed to acquire tools, seeds, fertilizers, and other productive inputs. The only pragmatic solution for many families was U.S. migration, a traditional source of ready cash (Massey et al. 1987). This solution was facilitated by the Bracero Program, which arranged for recruiting, transporting, and housing migrants beginning in the 1940s. In other communities, it was the absence of an agrarian reform program that led to widespread emigration. Such was the case in Guadalupe, Michoacan, where townspeople under the influence of a local priest opted for entrenching against agrarian activists. Because of their actions, the town did not receive an allocation of ejido lands from the neighboring hacienda, even though other nearby communities got generous allot- ments. Because they had no land, men from Guadalupe were forced to 29 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review migrate at an early date and soon built relationships with U.S. employers that allowed them to acquire legal status. U.S. earnings provided them with markedly higher incomes than the ejidatarios in surrounding communities, and gradually they began to acquire lands from neighboring ejidos. In the end, families from Guadalupe gained control of most of the land in the area. In Copandaro, Michoacan, an even more peculiar outcome followed the Reparto Agrario (Rionda 1986, 1990). Here a group of peasants bought parts of the hacienda before the reform began and became small landowners with an interest in preserving private property. With the advent of the agrarian reform movement, these new landowners were destroyed politically and economically by agrarian activists who came to political power under President Lazaro Cardenas. The ruined property owners found relief from their travails only through U.S. migration. Years later, however, the agrarian activists who had expropriated the lands found the small plots insufficient to support their families and were ironically forced to borrow money from the U. S. migrants whose property they had taken earlier. Thus U. S. migration became widespread in Copandaro despite an active agrarian reform program. In the hacienda of Jalpa, Guanajuato, yet another sequence of events transpired when U.S. migration preceded the Reparto Agrario and obviated the need for it. In this community, an enlightened hacienda owner launched his own private agrarian reform program and sold a large part of his property to local campesinos and to poor families in neighboring municipios. According to Paul Taylor, many migrants from neighboring Arandas bought land from the hacienda with U.S. earnings (1933, 34). An ejido was never formed, and the process of buying and selling land stopped only when U.S. remittances declined with the onset of the depression in the 1930s. In sum, it is impossible to conclude from the study of any single community how the Reparto Agrario affected the process of migration to the United States because its effects were contingent on community factors that are difficult to analyze in an isolated instance. Judging from the studies accumulated so far, these structural factors include the extent of political divisions between agrarian activists and landowners, the dis- tribution of land prior to the reform movement, and the degree to which a community participated in U.S. migration before initiation of the Reparto Agrario. During the 1960s and 1970s, another upheaval affected rural com- munities when a wave of agricultural modernization introduced commercial crops, capital-intensive production methods, fertilizers, and mechanization to many areas of Mexico (Yates 1981; Barkin and DeWalt 1988). Community studies have generally shown that this wave of modernization set in motion a series of changes that displaced farm laborers and 30 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. increased the pressures for out-migration (Lopez 1986a; Massey et al. 1987; Fonseca 1988; Goldring 1990; Grindle 1988). Specifically, agricul- tural modernization consolidated property holding, substituted machines for hand labor, introduced capital inputs such as herbicides and fertilizers, and expanded work opportunities for women in factories and canneries. A few studies, however, have examined the effects of agricultural modernization across a range of different communities and have reached a more guarded tentative conclusion. Kenneth Roberts compared four communities in the states of Guanajuato, San Luis Potosi, Oaxaca, and Puebla and found that the effects of agricultural modernization depended greatly on the distribution of property and the quality of land within the community (K. Roberts 1982, 1984). When commercial crops and capitalintensive production methods were introduced into areas with good soil, irrigated land, and an even distribution of property, farm incomes rose and the risks to household income fell, thereby reducing the need for migration. But when commercialized agriculture was introduced under unfavorable agricultural conditions and an unequal distribution of land, farm incomes fell and risks rose, leading families to diversify their sources of support through internal and international migration. These results accord with those of Jesus Arroyo (1989). His study showed that the introduction of commercial agriculture into poorly developed rural areas was strongly linked to out-migration, but its application within welldeveloped rural areas and semi-urban areas did not yield large outflows of labor (see also Arroyo, de Leon, and Valenzuela 1990). Again, a uniform process can produce very different outcomes depending on structural characteristics of the local community. Migrant Strategies One last area of debate in the literature on migration concerns the relative importance of different strategies that migrants employ while working in the United States. Investigators have generalized from the experience of diverse case studies to develop various typologies that attempt to summarize the motivations and behaviors employed by migrants in traveling and working in the United States. In the early period of U.S. migration, for example, Gamio identified two kinds of migrantspermanent and temporary (1930a). More recently, Mines used migrants' legal status, duration of trip, and commitment to U.S. work to classify migrants into four basic types: undocumented shuttles, beginner permanents, long-term permanents, and legal shuttles (Mines 1981). Similarly, based on the frequency and duration of U.S. trips, Massey et al. (1987) defined three basic strategies of migration: temporary, recurrent, and 31 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review settled. Cornelius (1978) provided the simplest typology of all, dividing migrants into sojourners and settlers. Often these strategies are treated as if they were fixed characteristics of migrants and their communities (see Cornelius 1978, 1990c; Mines 1981). Some investigators, however, have argued that migrant strategies are not reified traits but fluid patterns of behavior that shift ov time in response to changing household circumstances and community conditions. For example, Goldring (1990) links the prevalence of different strategies of migration in Las Animas and Gomez Farias to the contrasting niches that their migrants occupy within the U.S. occupational-industrial structure. Those from Gomez Farias specialize in shuttle or recurrent migration because they are overwhelmingly employed in agriculture, which is more conducive to legalization and has a built-in seasonality that encourages moving back and forth frequently. Las Animas, in contrast, is dominated by migrants employing a settled strategy because their employment in urban areas and their undocumented status are more com- patible with longer periods of residence. Consistent with these results, Massey et al. (1987) found that migrants from rural communities tend to work in U.S. agriculture and employ temporary or recurrent strategies, whereas those from urban areas worked in U.S. cities and tended to employ settled strategies. In addition, Massey and his colleagues also found that strategies varied systematically over the life cycle. Early phases of the life cycle, before marriage and family formation, were associated with a settled strategy; middle phases of the life cycle, during childbearing and childrearing, were associated with a temporary strategy; and late phases of the life cycle, after most children are grown, were correlated with either a recurrent or settled strategy Finally, Massey et al. (1987) found that the choice of strategies shifted as households and communities accumulated increasing amounts of U.S. migrant experience. During early phases of the process, just after the initiation of migration, temporary patterns were most common; but as the amount of U.S. experience grew, recurrent and settled strategies increasingly predominated. SUMMARY AND CONCLUSION In this article, we have summarized research-classic as well as recent-on a variety of salient issues related to Mexican migration to the United States. We have considered the sixty-year-old running debate about the number of Mexican migrants who enter and reside in the United States and reviewed the controversy about the size of their remittances. Disagreements on these issues stem primarily from the fact that much of the migratory flow is undocumented and hence unobservable. We have also considered results from two decades of community studies, which 32 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. provide detailed and reliable information on undocumented and legal migration to the United States. The main weakness of community stud- ies-and the source of much confusion-is that individual case studies do not readily sustain generalization to the whole of Mexico. Our review suggests that public debate over the number of undocumented migrants has been characterized by a great deal of rhetoric centered on highly speculative estimates and by considerable confusion over the terms of reference. When attention is restricted to estimates based on analytic methods and when one understands that some studies focus on undocumented workers while others consider undocumented migrants, the range of estimates is actually quite small. It leads to figures of under 2 million undocumented Mexicans in the United States in 1980 and just over 3 million in 1986. Similarly, migrant remittances appear to have totaled some 2 billion dollars in 1984. Community studies have attempted to analyze the social and economic processes underlying these aggregate statistics, but generalizations have been plagued by inconsistencies. These discrepancies indicate the hazards of generalizing from single cases. Yet the diversity of outcomes from different settings does not imply that it is impossible to generalize about U.S. migration. It simply reflects the fact that the pattern and course of U.S. migration are strongly shaped by factors acting at the community level. Generalizations can be made, but they must incorporate the con- tingent effects of community-level variables. These effects cannot be observed in single cases but only across a range of community types. Our review of findings from some thirty-two different communities suggests that four leading factors are crucial in determining patterns and processes of migration from Mexican communities to the United States. First, the age of the migration stream is crucial in determining the composition of the migrant flow. The length of time a community has been sending migrants to the United States determines the maturity of its networks, the immigration policies that applied to its migrants, and the total amount of migrant experience that has accumulated among its people. In general, the earlier that migration began, the more developed a community's networks will be, the more legalizations will have occurred, and the more experience community members will have accumulated in the United States. As a result, the longer a community has been involved in the migration process, the more likely that migrants will come from the poorest segments of the socioeconomic hierarchy, the greater the participation of women and children, the higher the proportion of documented migrants, and the greater the reliance on settled and recurrent migrant strategies. A second important community factor that shapes and conditions the process of migration is the niche in the U.S. occupational-industrial structure where migrants first became established. In general, agricultural 33 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review employment is associated with higher rates of legalization, greater participation by women and children, and a prevalence of temporary or recurrent migrant strategies. Although the occupational niche of the first migrants from a community is often a matter of chance, once migrants become established in a particular occupational-industrial position, they tend to channel other townspeople into the same structural location, thereby determining the character and composition of the subsequent stream. The position of a community within Mexico's political economy is also a crucial determinant of the composition and behavior of migrants to the United States. Obviously, rural agrarian communities will tend to send agricultural workers whereas urban communities send industrial or service workers, thereby conditioning the class composition of the migrant flow. But a community's position in Mexico's political economy also plays a strong role in determining migrant spending patterns. Migrants living in communities with access to urban markets, well-developed roads, accessible power supplies, good communications, and an even distribution of productive resources tend to spend larger shares of their U.S. earnings on productive ends than do migrants from isolated, poorly developed communities that lack this infrastructure. Finally, the distribution and quality of agricultural land plays a crucial role in shaping the processes of migration from rural communities. The degree of inequality in landholding obviously affects the class com- position of migration at the point of its initiation: recruitment from a community characterized by extreme inequality yields a stream of landless day laborers whereas recruitment from a town where land is more evenly distributed yields a flow of small landholders and ejidatarios. Inequality in landholding likewise affects the extent to which migrants are able to use their earnings to purchase land, and the quality of the land determines migrants' incentives for investing their earnings productively in farming. The local pattern of landholding also mediates the effect of structural shifts in the Mexican political economy, such as the Reparto Agrario, and strongly conditions the role they have played in generating migrants to the United States. The redistribution of land under the Reparto Agrario, rather than discouraging out-migration, actually encouraged international movement in most communities through several distinct mechanisms. In communities where land was owned by a large hacienda, the Reparto Agrario led to out-migration because it provided small parcels of land that were insufficient to support a family or gave land but no capital to enable farming. In communities where some households owned parcels but others owned nothing, the resulting political strife brought about the economic and political ruin of small landowners, who turned to U.S. migration as a means of supporting themselves. In other communities, 34 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. landowners and priests frightened townspeople into not applying for land under the Reparto Agrario, leaving them landless and open to re- cruitment by U.S. employers during the 1940s. The quality and distribution of land have also played important roles in conditioning the effects of the wave of agricultural modernization that swept Mexico during the 1960s and 1970s. Although the common wisdom is that adopting cash crops and capital-intensive production meth- ods led to greater out-migration via displacement of rural workers, comparative studies suggest that the process was more complex. Displacement did occur in communities with poor quality land and in places where land was unequally distributed, but in communities where land was of high quality and more evenly distributed among families, the advent of commercialized farming increased rural incomes, lowered risks to farm households, and thereby reduced the pressures for migration to the United States. The difficulty of generalizing about Mexico-U.S. migration has led many investigators to speak of "myths," "fallacies," "false assumptions," "exaggerations," or "games." The Colegio de Mexico published a book entitled Undocumented Migrants: Myths and Realities (1979). An article by David North discusses "five myths about illegal migration to the United States" (North n.d.). Cornelius (1979) speaks of the "new mythology of undocumented Mexican migration to the United States." Corwin calls estimates of undocumented migration a "numbers game" (1982). Garcia y Griego captions his book with the subtitle "Three False Assumptions about Emigration to the United States" (1988). Gustavo Verduzco com- ments on "the false assumptions about the Simpson-Rodino Law" (1987), and Margarita Nolasco refers to "the wishful calculations of returned migrants."3 Still others call for a "new focus" (Diez Canedo 1984) and a "new vision" (Garcia y Griego 1988). We have argued that this state of affairs follows from two factors: the political nature of the debate, which obscures the real consistency of facts about the number of migrants and the size of their remittances; and the fact that most case studies have failed to recognize the effect of structural variables operating at the community level, which strongly shape and condition the underlying process of migration. Our comparative analysis of communities suggests that a fruitful approach to developing more general statements about Mexico-U.S. migration is to focus on the ways in which community variables interact with individual and household processes to produce the manifold outcomes that we observe. 3. Margarita Nolasco, "Los alegres calculos de regresados," El Sol de Mexico, 4 July 1984, p. A-2. 35 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review REFERENCES ADELMAN, IRMA, J. EDWARD TAYLOR, AND STEPHEN VOGEL 1988 "Life in a Mexican Village: A SAM Perspective." Journal of Development Studies 25:5-24. ARIAS, PATRICIA 1988 "La pequefia empresa en el occidente rural." Estudios Sociologicos 6, no. 17:405-36. ARROYO, JESUS 1989 El abandono rural. Guadalajara: Editorial Universidad de Guadalajara. ARROYO, JESUS, ADRIAN DE LEON, AND BASILIA VALENZUELA 1990 Patterns of Migration and Regional Development in the State of Jalisco, Mexico. Commission for the Study of International Migration and Cooperative Economic Development, Working Paper no. 42. Washington, D.C.: Commission for the Study of International Migration and Cooperative Economic Development. BARKIN, DAVID, AND BILLIE R. DEWALT 1988 "Sorghum and the Mexican Food Crisis." LARR 23, no. 3:30-59. BEAN, FRANK D., BARRY EDMONSTON, AND JEFFREY S. PASSEL, EDS. 1990 Undocumented Migration to the United States: IRCA and the Experience of the 1980s. Washington, D.C.: Urban Institute Press. BEAN, FRANK D., THOMAS J. ESPENSHADE, MICHAEL J. WHITE, AND ROBERT F. DYMOWSKI 1990 "Post-IRCA Changes in the Volume and Composition of Undocumented Migration to the United States: An Assessment Based on Apprehensions Data." In BEAN, EDMONSTON, AND PASSEL 1990, 111-58. BEAN, FRANK D., ALLAN G. KING, AND JEFFREY S. PASSEL 1983 "The Number of Illegal Migrants of Mexican Origin in the United States: SexRatio-Based Estimates for 1980." Demography 20:99-109. BEAN, FRANK D., EDWARD E. TELLES, AND B. LINDSAY LOWELL 1987 "Undocumented Migration to the United States: Perception and Evidence." Population and Development Review 13:671-90. BEAN, FRANK D., GEORGES VERNEZ, AND CHARLES B. KEELY 1989 Opening and Closing the Doors: Evaluating Immigration Reform and Control. Santa Monica, Calif., and Washington, D.C.: Rand Corporation and the Urban Institute. BUSTAMANTE, JORGE 1979 "Emigracion indocumentada a los Estados Unidos." In CENTRO DE ESTUDIOS INTERNACIONALES 1979, 23-67. 1990 "Undocumented Migration from Mexico to the United States: Preliminary Find- ings from the Zapata Canyon Project." In BEAN, EDMONSTON, AND PASSEL 1990, 211-26. CALVO, THOMAS, AND GUSTAVO LOPEZ, EDS. 1988 Movimientos de poblacio'n en el occidente de Me'xico. Zamora: Colegio de Michoacan. CAMARA, FERNANDO, AND ROBERT VAN KEMPER, EDS. 1979 Migration across Frontiers: Mexico and the United States. Contributions of the Latin American Anthropology Group, vol. 3. Albany, N.Y.: Institute of Mesoamerican Studies, State University of New York. CARRERAS, MERCEDES 1974 Los mexicanos que devolvio la crisis 1929-1932. Mexico City: Secretaria de Relaciones Exteriores. CENIET (CENTRO NACIONAL DE INFORMACION Y ESTADISTICAS DEL TRABAJO) 1982 Los trabajadores mexicanos en Estados Unidos: resultados de la Encuesta Nacional de Emigracion de la Frontera Norte del Pais y a los Estados Unidos. Mexico City: Secretaria del Trabajo y Prevision Social. CENTRO DE ESTUDIOS INTERNACIONALES 1979 Indocumentados: mitos y realidades. Mexico City: Colegio de Mexico. CHAPMAN, LEONARD F., JR. 1976 "Illegal Aliens: Time to Call a Halt!" Reader's Digest, no. 109:188-92. CHAVEZ, LEO R., ESTEVAN T. FLORES, AND MARTA LOPEZ-GARZA 1990 "Here Today, Gone Tomorrow? Undocumented Settlers and Immigration Reform." Human Organization 49:193-205. 36 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. CORNELIUS, WAYNE A. 1976a "Outmigration from Rural Mexican Communities." Interdisciplinary Communications Program Occasional Monograph Series 5, no. 2:1-39. Washington, D.C.: Smithsonian Institution. 1976b Mexican Migration to the United States: The View from Rural Sending Communities. Migration and Development Monograph C/76-12. Cambridge, Mass.: MIT Center for International Studies. 1978 Mexican Migration to the United States: Causes, Consequences, and U.S. Responses. Migration and Development Monograph C/78-9. Cambridge, Mass.: Center for International Studies, Massachusetts Institute of Technology. 1979 "La migracion ilegal mexicano a los Estados Unidos: conclusiones de investigaciones recientes, implicaciones politicas y prioridades de investigacion." In CENTRO DE ESTUDIOS INTERNACIONALES 1979, 69-110. 1989 "Impacts of the 1986 U.S. Immigration Law on Emigration from Rural Mexican Sending Communities." Population and Development Review 15:689-706. 1990a Labor Migration to the United States: Development Outcomes and Alternatives in Mexican Sending Communities. Final Report to the U.S. Commission for the Study of International Migration and Cooperative Economic Development. La Jolla: Center for U. S.-Mexican Studies, University of California, San Diego. 1990b Labor Migration to the United States: Development Outcomes and Alternatives in Mexican Sending Communities. Commission for the Study of International Migration and Cooperative Economic Development Working Paper no. 38. Washington, D.C.: Commission for the Study of International Migration and Cooperative Economic Development. 1990c "From Sojourners to Settlers: The Changing Profile of Mexican Migration to the United States." Manuscript, Center for U.S. -Mexican Studies, University of California at San Diego, La Jolla. CORONA, RODOLFO 1987 "Estimacion del nuimero de indocumentados a nivel estatal y municipal." Mimeo, Centro de Estudios sobre Identidad Nacional en Zonas Fronterizas, Universidad Nacional de Mexico, Cuernavaca, Morelos. CORWIN, ARTHUR F. 1982 "The Numbers Game: Estimates of Illegal Aliens in the United States, 1970-1981." Law and Contemporary Problems 45:223-97. CRAIG, RICHARD B. 1971 The Bracero Program: Interest Groups and Foreign Policy. Austin: University of Texas Press. DIEZ CANEDO, JUAN 1984 La migracion indocumentada de Mexico a los Estados Unidos: un nuevo enfoque. Mexico City: Fondo de Cultura Econ6mica. DINERMAN, INA R. 1982 Migrants and Stay-at-Homes: A Comparative Study of Rural Migration from Michoacan, Me'xico. La Jolla: Center for U. S.-Mexican Studies, University of California, San Diego. DURAND, JORGE 1986 "Circuitos migratorios en el occidente de Mexico." Revue Europeanns des Migrations Internationals 2:45-67 (published in Poipiers, France). 1988 "Los mivrad6lares: cien afios de inversi6n en el medio rural." Argumentos: Estu- dios Criticos de la Sociedad 5:7-21. 1989a "Los hijos de Rodino." L'Ordinaire Mexique Amerique Centrale 122:57-60 (published by the Universite de Toulouse-Le Mirail). 1989b "Guanajuato, tierra de migrantes." Paper presented at the seminar "International Migration: Current State and Perspectives," Cocoyoc, Morelos, 4-6 October. 1990 "Guanajuato, cantera de migrants." Encuentro 4, no. 4:49-62. ESCOBAR, AGUSTIN, MERCEDES GONZALEZ, AND BRYAN ROBERTS 1987 "Migration, Labor Markets, and the International Economy: Jalisco, Mexico, and the United States." In Migrants, Workers, and the Social Order, edited by Jeremy Eades, 42-64. London: Tavistock. 37 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review ESCOBAR, AGUSTIN, AND MARIA DE LA 0 MARTINEZ 1990 Small-Scale Industry and International Migration in Guadalajara, Mexico. Commission for the Study of International Migration and Cooperative Economic Development, Working Paper no. 53. Washington, D.C.: Commission for the Study of International Migration and Cooperative Economic Development. FABILA, ALFONSO 1929 El problema de la emigracion de obreros y campesinos mexicanos. Mexico City: Talleres Graficos de la Nacion. FERNANDEZ, CELESTINO 1988 "Migracion hacia los Estados Unidos: caso Santa Ines, Michoacan." In LOPEZ, ED., 1988, 113-24. FONSECA, OMAR 1988 "De Jaripo a Stockton, California: un caso de migracion en Michoacan." In CALVO AND LOPEZ 1988, 359-72. FONSECA, OMAR, AND LILIA MORENO 1988 "Consideraciones historico-sociales de la migracion de trabajadores michoacanos a los Estados Unidos de America: el caso de Jaripo." In LOPEZ, ED., 1988, 65-84. GALARZA, ERNEST 1964 Merchants of Labor: The Mexican Bracero Story. Santa Barbara, Calif.: McNally and Loftin. GAMIO, MANUEL 1930a Mexican Immigration to the United States. Chicago, Ill.: University of Chicago Press. 1930b Numero, procedencia y distribucion geogrdfica de los inmigrantes mexicanos en los Estados Unidos. Mexico City: Talleres Graficos y Diario Oficial. 1931 The Mexican Immigrant: His Life Story. Chicago, Ill.: University of Chicago Press. GARCIA Y GRIEGO, MANUEL 1987 "International Migration Statistics in Mexico." International Migration Review 21: 1245-57. 1988 "Hacia una nueva vision del problema de los indocumentados en los Estados Unidos." In Mexico y Estados Unidos frente a la migracio'n de indocumentados, edited by Manuel Garcia y Griego and Monica Verea, 123-53. Mexico City: Miguel Angel Porru'a. GARCIA Y GRIEGO, MANUEL, AND FRANCISCO GINER DE LOS RIOS 1985 "ZEs vulnerable la economia a la aplicacion de politicas m denses?" In Me'xico-Estados Unidos, edited by Manuel Garcia y Griego and Gus- tavo Vega. Mexico City: Colegio de Mexico. GOLDRING, LUIN 1990 Development and Migration: A Comparative Analysis of Two Mexican Migrant Cir- cuits. Commission for the Study of International Migration and Cooperative Economic Development, Working Paper no. 37. Washington, D.C.: Commission for the Study of International Migration and Cooperative Economic Development. GONZALEZ, MERCEDES, AND AGUSTIN ESCOBAR 1990 The Impact of IRCA on the Migration Patterns of a Community in Los Altos, Jalisco, Mexico. Commission for the Study of International Migration and Cooperative Economic Development, Working Paper no. 41. Washington, D.C.: Commission for the Study of International Migration and Cooperative Economic Development. GRINDLE, MERILEE S. 1988 Searching for Rural Development: Labor Migration and Employment in Mexico. Ithaca, N.Y.: Cornell University Press. GUZMAN, RALPH C. 1979 "La repatriacion forzosa como solucion politica concluyente al problema de la emigraci6n ilegal: una perspectiva hist6rica." In CENTRO DE ESTUDIOS INTERNACIONALES 1979, 137-63. HANCOCK, RICHARD 1959 The Role of the Bracero in the Economic and Cultural Dynamics of Mexico: A Case Study of Chihuahua. Stanford, Calif.: Hispanic-American Society. 38 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. HEER, DAVID M., AND JEFFREY S. PASSEL 1985 "Comparison of Two Different Methods for Computing the Number of Undocu- mented Mexican Adults in the Los Angeles PMSA." Paper presented to the Population Association of America, 21-24 Apr., Boston, Massachusetts. HOFFMAN, ABRAHAM 1974 Unwanted Mexican Americans in the Great Depression: Repatriation Pressures, 19291939. Tucson: University of Arizona Press. IMF (INTERNATIONAL MONETARY FUND) 1987 International Financial Statistics: 1987 Yearbook. Washington, D.C.: International Monetary Fund. JASSO, GUILLERMINA, AND MARK R. ROSENZWEIG 1990 The New Chosen People: Immigrants in the United States. New York: Russell Sage. JONES, RICHARD C., ED. 1984 Patterns of Undocumented Migration: Mexico and the United States. Totowa, N.J.: Rowman and Allanheld. KEELY, CHARLES B., AND BAO NGA TRAN 1989 "Remittances from Labor Migration: Evaluations, Performance, and Implications." International Migration Review 23:500-525. KOSSOUDJI, SHERRIE A. 1990 "Playing Cat and Mouse at the Border: Does the INS Alien Apprehension Strat- egy Alter the Aggregate Supply of Illegal Labor?" Paper presented at the Annual Meetings of the Population Association of America, 3-5 May, Toronto, Ontario. KOSSOUDJI, SHERRIE A., AND SUSAN I. RANNEY 1984 "The Labor Market Experience of Female Migrants: The Case of Temporary Mexican Migration to the U.S." International Migration Review 18:1120-43. 1986 Legal Status as Union Membership: Legal and Illegal Wage Rates of Mexican Immigrants. Population Studies Center Research Report no. 86-103. Ann Arbor: Population Studies Center, University of Michigan. LESKO ASSOCIATES 1975 Final Report: Basic Data and Guidance Required to Implement a Major Illegal Alien Study during Fiscal Year 1976. Washington, D.C.: U.S. Immigration and Naturalization Service. LOPEZ, GUSTAVO 1986a La casa dividida: un estudio de caso sobre migraci6n a Estados Unidos en un pueblo michoacano. Zamora: Colegio de Michoacan. 1986b "Migraci6n temporal a Estados Unidos en una comunidad mexicana." In Se fue a volver: Seminario sobre Migraciones Temporales en America Latina. Mexico City: Pispal/Ciudad/CENEP. 1988 "La migraci6n a Estados Unidos en G6mez Farias, MichoacAn." In LOPEZ, ED., 1988, 125-34. LOPEZ, GUSTAVO, ED. 1988 Migraci6n en el occidente de Mexico. Zamora: Colegio de Michoacan. LOPEZ, GUSTAVO, AND SERGIO ZENDEJAS 1988 "Migraci6n internacional por regiones en Michoacan." In CALVO AND LOPEZ 1988, 81-102. LOWELL, B. LINDSAY, FRANK D. BEAN, AND RODOLFO 0. DE LA GARZA 1986 'The Dilemmas of Undocumented Immigration: An Analysis of the 1984 SimpsonMazzoli Vote." Social Science Quarterly 67:118-27. MARTIN, PHILIP L. 1990 "Harvest of Confusion: Immigration Reform and California Agriculture." International Migration Review 24:69-95. MASSEY, DOUGLAS S. 1981 "The New Immigration to the United States and the Prospects for Assimilation." Annual Review of Sociology 7:57-85. 1987 "Understanding Mexican Migration to the United States." American Journal of Sociology 92:1372-1403. 39 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review MASSEY, DOUGLAS S., RAFAEL ALARCON, JORGE DURAND, AND HUMBERTO GONZALES 1987 Return to Aztlan: The Social Process of International Migration from Western Mexico. Berkeley and Los Angeles: University of California Press. MASSEY, DOUGLAS S., AND LAURENCE BASEM n.d. "Determinants of Savings, Remittances, and Spending Patterns among Mexican Migrants to the United States." Sociological Inquiry, forthcoming. MASSEY, DOUGLAS S., KATHARINE M. DONATO, AND ZAI LIANG 1990 "Effects of the Immigration Reform and Control Act of 1986: Preliminary Data from Mexico." In BEAN, EDMONSTON, AND PASSEL 1990, 183-210. MASSEY, DOUGLAS S., AND FELIPE GARCIA ESPANA 1987 "The Social Process of International Migration." Science, no. 237:733-38. MINES, RICHARD 1981 Developing a Community Tradition of Migration: A Field Study in Rural Zacatecas, Mexico, and California Settlement Areas. Monographs in U.S.-Mexican Studies no. 3. La Jolla: Center for U. S.-Mexican Studies, University of California at San Diego. 1984 "Network Migration and Mexican Rural Development: A Case Study." In JONES 1984, 136-55. MINES, RICHARD, AND ALAIN DE JANVRY 1982 "Migration to the United States and Mexican Rural Development: A Case Study" American Journal of Agricultural Economics 64:444-54. MINES, RICHARD, AND DOUGLAS S. MASSEY 1985 "Patterns of Migration to the United States from Two Mexican Communities." LARR 20, no. 2:104-23. MORALES, PATRICIA 1981 Indocumentados Mexicanos. Mexico City: Grijalbo. NORTH, DAVID S. n.d. "Cinco mitos en torno a la inmigraci6n ilegal hacia Estados Unidos." Revista de la Universidad de Guadalajara 9, no. 2:11-20. NORTH, DAVID S., AND MARION F. HOUSTOUN 1976 The Characteristics and Role of Illegal Aliens in the U. S. Labor Market: An Exploratory Study. Washington, D.C.: Linton. OROZCO, JUAN LUIS 1990 "El negocio de los ilegales: Zganancias para quien?" Ph.D. diss., Estudios de America Latina, Paris. PASSEL, JEFFREY S. 1985 "Undocumented Immigrants: How Many?" Proceedings of the Social Statistics Section, American Statistical Association, 1985:65-72. 1986 "Undocumented Immigration." Annals, American Academy of Political and Social Science 487:181-200. PASSEL, JEFFREY S., AND KAREN A. WOODROW 1984 "Geographic Distribution of Undocumented Immigrants: Estimates of Undocumented Aliens Counted in the 1980 Census by State." International Migration Review 18:642-71. 1987 "Change in the Undocumented Alien Population in the United States, 1979-83." International Migration Review 21:1304-23. PLEHN, CARL C., IRA B. CROSS, AND MELVIN M. KNIGHT, EDS. 1929- University of California Publications in Economics, Berkeley: University of California 1932 Press. PORTES, ALEJANDRO, AND RUBEN C. RUMBAUT 1990 Immigrant America: A Portrait. Berkeley and Los Angeles: University of California Press. RANNEY, SUSAN I., AND SHERRIE A. KOSSOUDJI 1983 "Profiles of Temporary Mexican Labor Migrants to the United States." Population and Development Review 9:475-92. REICHERT, JOSHUA S. 1979 "The Migrant Syndrome: An Analysis of U.S. Migration and Its Impact on a Rural Mexican Town." Ph.D. diss., Princeton University. 40 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �MEXICAN MIGRATION TO THE U.S. 1981 "The Migrant Syndrome: Seasonal U.S. Wage Labor and Rural Development in Central Mexico." Human Organization 40:56-66. 1982 "A Town Divided: Economic Stratification and Social Relations in a Mexican Migrant Community." Social Problems 29:411-23. REICHERT, JOSHUA S., AND DOUGLAS S. MASSEY 1979 "Patterns of U.S. Migration from a Mexican Sending Community: A Comparison of Legal and Illegal Migrants." International Migration Review 13:599-623. 1980 "History and Trends in U.S.-Bound Migration from a Mexican Town." International Migration Review 14:475-91. REUBENS, EDWIN P. 1980 "Immigration Problems, Limited-Visa Programs, and Other Options." Appendix F in U. S. Immigration Policy and the National Interest: Report of the Select Commission on Immigration and Refugee Policy. Washington, D.C.: U.S. Government Printing Office. RIONDA, LUIS MIGUEL 1986 "Continuidad y accesi6n social en Copandaro." In Estudios Michoacanos II, edited by Carlos Herrej6n, 245-56. Zamora: Colegio de Michoacan. 1990 "Y jalaron pal norte." M.A. thesis, Colegio de Michoacan. ROBERTS, KENNETH D. 1982 'Agrarian Structure and Labor Mobility in Rural Mexico." Population and Development Review 8:299-322. 1984 'Agricultural Development and Labor Mobility: A Study of Four Mexican Subregions." In JONES 1984, 74-92. ROUSE, ROGER C. 1989 "Mexican Migration to the United States: Family Relations in the Development of a Transnational Migrant Circuit." Ph.D. diss., Stanford University. SAMORA, JULIAN 1971 Los Mojados: The Wetback Story. Notre Dame, Ind.: University of Not SANTIBANEZ, ENRIQUE 1930 Ensayo acerca de la inmigraci6n mexicana en los Estados Unidos. San Antonio, Tex.: Cleeg. SELBY, HENRY A., AND ARTHUR D. MURPHY 1984 The Mexican Urban Household and the Decision to Migrate to the United States. Occasional Papers in Social Change, no. 4. Philadelphia, Pa.: Institute for the Study of Human Issues. SHADOW, ROBERT D. 1979 "Differential Out-Migration: A Comparison of Internal and International Migration from Villa Guerrero, Jalisco (Mexico)." In CAMARA AND VAN KEMPER 1979, 67-84. STARK, ODED, AND J. EDWARD TAYLOR 1989 "Relative Deprivation and International Migration." Demography 26:1-14. STUART, JAMES, AND MICHAEL KEARNEY 1981 Causes and Effects of Agricultural Labor Migration from the M California. La Jolla: Center for U.S.-Mexican Studies, University of California, San Diego. TAYLOR, J. EDWARD 1986 "Differential Migration, Networks, Information, and Risk." In Research in Human Capital and Development: Migration, Human Capital, and Development, edited by Oded Stark, 4:147-71. Greenwich, Conn.: Jai. 1987 "Undocumented Mexico-U.S. Migration and the Returns to Households in Rural Mexico." American Journal of Agricultural Economics 69:626-37. 1988 "Life in a Mexican Village: A SAM Perspective." Journal of Development Studies 25:5-24. TAYLOR, PAUL S. 1929 "Mexican Labor in the United States: Migration Statistics." In PLEHN, CROSS, AND KNIGHT 1929, 237-55. 1930 "Mexican Labor in the United States: Dimmit County, Winter Garden District, South Texas." In PLEHN, CROSS, AND KNIGHT 1930, 393-464. 41 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms �Latin American Research Review 1931 "Mexican Labor in the United States: Bethlehem, Pennsylvania." In PLEHN, CROSS, AND KNIGHT 1931, 1-24. 1932 "Mexican Labor in the United States: Chicago and the Calumet Region." In PLEHN, CROSS, AND KNIGHT 1932, 25-284. 1933 'A Spanish-Mexican Peasant Community: Arandas in Jalisco, Mexico." IberoAmericano 4:1-80 (published by the University of California Press). TRIGUEROS, PAZ, AND JAVIER RODRIGUEZ 1988 "Migraci6n y vida familiar en Michoacan." In LOPEZ, ED., 1988, 201-21. U.S. INS (IMMIGRATION AND NATURALIZATION SERVICE) 1989 1988 Statistical Yearbook of the Immigration and Naturalization Service. Washington, D.C.: U.S. Government Printing Office. VERDUZCO, GUSTAVO 1987 "Los falsos supuestos de la Ley Simpson-Rodino." Foro Internacional 27:168-70. WARREN, ROBERT, AND JEFFREY S. PASSEL 1987 'A Count of the Uncountable: Estimates of Undocumented Aliens Counted in the 1980 United States Census." Demography 24:375-93. WIEST, RAYMOND 1973 "Wage-Labor Migration and the Household in a Mexican Town." Journal of Anthropological Research 29:108-209. 1979 "Implications of International Labor Migration for Mexican Rural Development." In CAMARA AND VAN KEMPER 1979, 85-97. 1984 "External Dependency and the Perpetuation of Temporary Migration to the United States." In JONES 1984, 110-35. YATES, P. LAMARTINE 1981 Mexico's Agricultural Dilemma. Tucson: University of Arizona Press. 42 This content downloaded from 24.4.24.213 on Fri, 04 Mar 2022 00:44:45 UTC All use subject to https://about.jstor.org/terms � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Mexican Migration to the United States: A Critical Review Creator An entity primarily responsible for making the resource Durand, Jorge Massey, Douglas, S Subject The topic of the resource Immigration & society Mexican Immigration Latin America Description An account of the resource An examination of “two particular salient issues — The number of Mexican migrants to the United States and the quantity of their monetary remittances to Mexico” Publisher An entity responsible for making the resource available Latin American Research Review Date A point or period of time associated with an event in the lifecycle of the resource 1992 Type The nature or genre of the resource Text Review Article Format The file format, physical medium, or dimensions of the resource PDF Identifier An unambiguous reference to the resource within a given context https://www.jstor.org/stable/2503748 Source A related resource from which the described resource is derived JSTOR Language A language of the resource English