 |
 |
 |
 |
NAFTA AND MEXICAN MIGRATION TO MICHIGAN by Robert Aponte
WORKING PAPER NO. 25 July 14, 1995
Submitted to the The argument here was originally presented at the conference, NAFTA: Free Trade Among Nations, at Michigan State University, November 1994. The author gratefully acknowledges the contribution to this effort of the Samora Institute's editorial staff along with the commentary of Rosemary Aponte, David Arsen, Steve Gold, Cory Krupp, Karen Roberts, and, especially, Peter Dorman, on earlier drafts. However, the usual disclaimers fully apply. Table Of Contents
Among the key selling points of the North American Free Trade Agreement (NAFTA) was the notion that the economic benefits of treaty's passage would slow Mexico-US migration. Whatever the ultimate extent of NAFTA's economic consequences, it is argued here that the treaty's implementation will stimulate more migration than it will stem. This is because neither the quality nor the quantity of the NAFTA-induced jobs on the Mexican side, the key ingredient in the migration-supression hypothesis, will be sufficient to deter many northern bound labor migrants. In addition, the economic integration process itself will entail job-losses in Mexico that will stimulate increased migration. On balance, therefore, outmigration will increase, rather than decrease. Despite this prediction, it is also argued here that Michigan, while sure to remain among the states most actively involved in financial dealings with Mexico, is not likely to be among the major receivers of the enhanced migration streams. Michigan will host little increased migration because most such migrants are likely to enter via existing trajectories which currently funnel large numbers of migrants. The best data available suggest that very little direct immigration from Mexico is currently steered toward Michigan. For example, the Mexican-origin population in the state grew at a very low rate during the 1980s, a decade of record Mexican immigration, and only a portion of this growth could have been due to migration. Indeed, in absolute terms, the state's Mexican-origin population grew faster in the 1970s than the 1980s, and it is likely that the portion of the growth due to migration came largely from other parts of the US, rather than from Mexico. Introduction | NAFTA and Migration | The NAFTA Jobs Debate: The Mexican Side | The Social Process Of Mexico-US Migration | NAFTA, Mexico-US Migration, and Michigan Communities | Conclusion | Notes | References The potential effects of NAFTA on migration, like the predicted effects of NAFTA on US and Mexican employment, are the subject of considerable disagreement (cf. Alarcon 1994, Andreas 1994, Bailey 1993, Cornelius and Martin 1993, Faux 1993, Hinojosa-Ojeda and Robinson 1992, Martin 1993, Weintraub 1994), with a general tendency for the treaty's proponents to expect decreased migration and vice versa. The conventional view is that the treaty will stimulate increased migration in the short to intermediate run, but clearly foster decreased migration in the intermediate to long (10-20 years) run (cf. Cornelius and Martin 1993, Hufbauer and Schott 1992, Martin 1993, Weintraub 1994). An alternative view, implied or expressed to varying degrees by various authors, is that NAFTA is likely to stimulate increased migration well into the foreseeable future (cf. Alarcon 1994, Andreas 1994, Hinojosa-Ojeda and Robinson 1992, Koechlin and Larudee 1992). The position taken here elaborates the latter perspective. In order to evaluate the viability of any of these views, one must first understand precisely how NAFTA is supposed to affect the current patterning of migration, particularly since the treaty has no provisions dealing with migration. First, it must be understood that the sustained migration waves are composed overwhelmingly of low-skilled labor migrants pursuing economic opportunities that are simply not available in Mexico, an impoverished country with rampant joblessness and correspondingly low wages (cf, Andreas 1994, Belous and Lemco 1993, Martin 1993, Monto 1994, Muller and Espenshade 1985). Second, the treaty is expected to substantially affect the mix of labor opportunities in both the US and Mexico, but most especially the latter for a variety of reasons. Key among these is size: as the smallest economy, any given volume of additional business will disproportionately affect it. In addition, Mexico's economy is expected to attract more foreign investment from the other NAFTA countries than vice versa, particularly for labor-intensive production, given its abundant pools of low-wage labor. Thus, the essential ingredient in the migration-suppression viewpoint, as elaborated below, is Mexican job growth. The mechanism by which this is supposed to work is fairly straightforward. The increased investment in Mexican production and the commensurate expansion in Mexican exports should create an abundance of jobs which will absorb potential migrants. While some jobs will also be lost, as less competitive Mexican producers are undermined by heightened competition, the balance is expected to be substantial net employment growth. The effects of NAFTA on US employment, by contrast, are considered largely irrelevant to any potential shifts in migration for the following reasons. First, NAFTA is clearly not expected to net any low-skilled work on the US side; thus, the treaty cannot induce more migration through any expansion of opportunities it may produce in the US. At the same time, however, it will also not decrease US based opportunities for those continuing north. Rather, compelling evidence supports the widely held assumption that US demand for low-skilled workers, willing to toil at menial and/or difficult tasks for exceptionally low-wages by US standards (ie., immigrant labor), will continue to increase in the foreseeable future with or without NAFTA (cf. Alba 1992, Hinojosa-Ojeda and Robinson 1992, Martin 1993, 1994, Muller 1992). Hence, if NAFTA is to stem the flow of migrants, it must produce a substantial expansion of opportunities on the Mexican side of the border!1 In focusing on jobs in Mexico, the debate over migration departs sharply from the larger debate in the US over the advisability of our entering into NAFTA. Whereas the larger debate focused on whether NAFTA would hurt employment prospects for US workers, largely assuming major job gains to Mexico, the migration debate focuses on the adequacy of job creation in Mexico, irrespective of job gains or losses in the US. One implication of this is that relatively little work by US based scholars has focused on the precise net effects, in terms of employment, that NAFTA should produce in Mexico. Despite this, it cannot be overemphasized that the expected migration-supression effect of the treaty ultimately rests almost entirely on the substantial employment growth NAFTA is supposed to produce south of the border. But, it is argued here that the earnings levels of the new jobs will also matter; if they are much lower than wages in accessible US based jobs, many laborers will perceive little incentive to forgo traveling north. While the pro-NAFTA literature often implies that Mexican wages will rise significantly under the treaty, such projections are ambiguous and unconvincing (cf., Lustig 1994, Hufbauer and Schott 1992). The view taken here is that NAFTA will fail to produce the jobs or wages suggested by proponents and that, for these and related reasons, the treaty will stimulate more migration than it will stem. First, the likelihood that net job growth will be at levels sufficiently high to deter migrants is remote or nil. Second, wage levels will NOT rise to anywhere near the levels necessary to be competitive with US standards for decades (if ever), thereby providing little attraction to most potential migrants. Third, the process of labor migration itself will prove relatively impervious to the beckoning of the opportunities promised by NAFTA. Introduction | NAFTA and Migration | The NAFTA Jobs Debate: The Mexican Side | The Social Process Of Mexico-US Migration | NAFTA, Mexico-US Migration, and Michigan Communities | Conclusion | Notes | References THE NAFTA JOBS DEBATE: THE MEXICAN SIDE The most fundamental and preliminary question on this issue is will NAFTA foster enough net job growth in Mexico to actually affect migration. Most observers agree that Mexican manufacturing employment will grow substantially under NAFTA, while agricultural work will substantially diminish, but the net outcome remains in dispute. A point on which substantial agreement exists, however, is that millions of rural Mexicans will be displaced from agriculture in the foreseeable future as a result of NAFTA. Indeed, the potential devastation to agricultural production in Mexico from NAFTA is so great that the schedules for easing competitive restrictions on key agricultural commodities are, by agreement, phased in gradually. The complete lowering of restrictions on the importation of US corn, for example, is phased in over a full 15 year period! The expected agricultural displacement is the main reason why proponents of NAFTA concede that, in the short term, NAFTA will actually stimulate increased migration. The idea is that after an initial upsurge from rural displacement, migration will substantially fall as potential migrants increasingly respond to the NAFTA-induced expansion in industrial opportunities (Cornelius and Martin 1993, Hufbauer and Schott 1993). It is clear, therefore, that any ultimate net gains to Mexican employment from NAFTA will necessarily result from a balance of both major gains (in manufacturing) and major losses (in agriculture). The varying estimates on NAFTA-induced employment shifts in Mexico range from net losses of up to 1.6 million jobs by the year 2000 in one pessimistic estimate (Koechlin and Larudee 1992) to a loss of 30 percent of current manufacturing employment (without regard to rural displacement) in another Alvarez and Mendoza 1992). Studies finding positive net gains for Mexican employment, the predominate position among US scholars, have yielded precious few estimates that specify the actual level of projected job growth. The notable exception to this, one of many studies based on economic simulations, suggests NAFTA will produce a net gain of some 690,000 jobs to Mexico sometime between 1995 and 1997, (Hufbauer and Schott 1992:52-57).2 The widespread view that Mexico will reap major benefits from NAFTA rests largely upon this and related simulation modeling studies. Indeed, most such work has concluded that, in keeping with the "comparative advantage" angle of standard trade theory, all three economies will benefit from liberalized trade, but especially Mexico (see reviews in Brown 1992, Hinojosa-Ojeda and Robinson 1992, Martin 1993, Weintraub 1992). As Martin's (1993) summary remarks put it "Mexico is the economic 'winner' in NAFTA, but all three countries will have bigger economies, more jobs, and higher wages," a generalized finding echoed by "Most government, academic, and industry studies..." (Martin 1993:38-9). Based on such generalizations, "experts," such as Nora Lustig of the Brookings Institution, confidently trumpet the good news that NAFTA will unquestionably produce "higher incomes in Mexico" with a "doubling" of the rate of growth in the nation's "overall economy" and "wages" (Lustig 1994:47). It is also likely that these ideas have gained wide acceptance because the greatly feared NAFTA-induced US employment declines are generally assumed to directly translate into Mexican jobs. Aside from the immense popularity of 1994 presidential candidate Ross Perot's quip that NAFTA would induce a "giant sucking sound" of US jobs migrating south of the border, there is the growing awareness of the maquiladora experience. Widespread publicity about the maquiladora program (e.g., Barlett and Steele 1991, 1992), in which hundreds of US firms created or moved numerous production facilities to Mexico, thereby displacing US workers, almost certainly facilitated implanting the expectation that NAFTA will extend the pattern.3 However, despite the widespread acceptance of this view, there is reason to doubt its accuracy and, especially, the findings of the simulation modelings. First, the models rest upon numerous dubious assumptions about the real world. The most serious of these is the assumption that full employment regimes characterize the economies in question, an obvious and serious flaw; others include the assumption of free competition (many producers, none with serious market share advantages) and balanced trade. All of the modeling techniques used in making NAFTA-impact projections rest upon these or related assumptions (see reviews in Cypher 1993, Marshall 1994, Stanford 1993). The problems with these kinds of modeling are concisely and expertly summarized in Stanford (1993). In brief, the idea is that such models necessarily begin with a set of assumptions of how an economy actually works. In turn, these assumptions are applied to a "starting point" set of real economic indicators, after which the model tells what should transpire in the various output variables (employment, trade, etc), given a change in input variables (e.g., lowering tariffs). The problem is that the prediction is wholly dependent upon the theoretical structure of the model itself, or its underlying assumptions. Referring to the so-called CGE (computable general equilibrium) models, the most widely used, Stanford (1993) notes:
Stanford goes on to note that the apparent consensus of pro-NAFTA modelers "reflects more a consensus of prior theoretical views than a consensus of quantitative evidence" (1993:100). Basing their views more on recent trends than on modeling techniques, some analysts have even argued that manufacturing employment will experience net declines or only miniscule growth in Mexico under NAFTA (Alvarez and Mendoza 1992, Rendon and Salas 1993, Tanski 1993). Their reasoning includes, first, the fact that much of Mexican industrial production, outside of the maquiladoras, is known to be highly inefficient and therefore itself vulnerable to competitive displacement. Second, there have been a number of massive layoffs already in the 1980s, particularly in steel and autowork (Alvarez and Mendoza 1992). While some 600,000 industrial jobs were generated during the 1980-92 period, a matching number were simultaneously lost. In fact, from 1985 to 1990, nearly 90 percent of all industrial job growth was due to maquiladora expansion (Pradilla Cobo 1993). As of 1992, there were actually fewer workers employed in industry than there had been in 1980 (Barkin 1992). Because this relative thinning of industrial workers occurred during a period when Mexico's economy was gradually being privatized and opened to foreign penetration, a process to be accelerated under NAFTA, these observers expect little industrial job additions from NAFTA, despite the hoopla to the contrary. These potential demand-side insufficiencies are matched by three mammoth supply-side surpluses which further undermine the NAFTA promise. First, the Mexican labor force hosts around one million new additions each year (Hinojosa-Ojeda and Robinson 1992, Martin 1993, Weintraub 1994). Yet, none has dared suggest that the treaty can produce jobs of that magnitude. In addition, there are the millions of Mexicans that are already un- or underemployed; it is reasonable to expect that many of these such individuals would also have to be absorbed if NAFTA were to truly have a major impact on labor migration. Finally, there are the hundreds of thousands of agricultural workers --some estimates run into the millions-- expected to be displaced by NAFTA and related policy shifts in Mexico (cf. Koechlin and Larudee 1993, Martin 1993). This combination of needs dwarfs the magnitude of opportunities promised by even the most optimistic of NAFTA scenarios. A still more compelling item undercutting the migration-supression hypothesis concerns wages. NAFTA adherents consistently argue or imply that the projected economic growth will ensure wage growth, that productivity differences are the key reason for the existing differentials, and that the newly induced growth will progressively eradicate each of these. However, the recent record belies all such claims. The critical claim that "productivity differences" fully account for the Mexican-US industrial wage gap is especially contradicted by a wide body of findings. Indeed, figures deriving from our own the Bureau of Labor Statistics show that Mexican industrial productivity rose by roughly 40 percent from 1980 to 1992, yet real wages dropped by a full 32 percent over the period (Marshall 1994, Shaiken 1993). Another popular argument by NAFTA proponents, reflecting real supply and demand factors, rather than "productivity" measures, is that the sheer increase in demand for industrial labor under NAFTA will force up wages. This argument is also belied by the maquiladora experience. Employment in maquiladora plants exploded from about 130,000 in 1980 to over 500,000 in 1992 (Shaiken 1993). Yet, real wages also dropped in this sector, Mexico's main growth sector, and average wages there are likely to be even lower than in other industrial jobs (cf. Pradilla Cobo 1993, George 1990, Martin 1993). A recent inspection of a maquiladora plant by an economist with a congressional team underscored the issue. The unannounced visit to a Sanyo plant by the group led to a candid interview with the plant's manager. The manager was asked how the plant's productivity compared with that of its counterparts in the US. He responded that, after four odd years in operation, the plant's productivity matched US standards by 100 percent. As explained by Jeff Faux, the economist on the trip (1993:6)
Similar accounts have been reported by Shaiken (1994) who studied four major foreign-owned plants in the 1991-92 period chosen for their superior quality. Taken together, the four plants employed over 7,000 employees. Despite rapid turnover (about 130 percent one year in one of the plants) and substantial worker dissatisfaction with wages and conditions, the plants were described as on par with high tech production facilities in both the US and Japan. Shaiken's summary remarks on this issue are instructive:
The substantial disjunct between the workers' earnings and their productivity were indirectly ascertained by managers. Typical commentary on the plant's problems offered by one of the interviewed managers made the point "You've got only one problem in that plant and that's wages. We took the cream of the work force,...and we gave them all this training, and we're just not paying them enough money" while another noted "we know this is not enough money for someone to survive" (Shaiken 1994:58). Worse, the managers freely admitted that the Mexican government's policies helped to suppress wages. "We even get help from the government making sure we don't settle too high because of the economic reforms and the fact that we're so visible" stated one official, after which he made the following reference to a concession recently granted by workers: "We suspect that one of the reasons for the ..[concession].. was the amount of pressure the government was putting on CTM [the union] to settle at a low level because of ..[our].. visibility," while still another supervisor pointed out "The labor ministry takes an active part in negotiations, especially in companies our size. And they steer the level of increases" (Shaiken 1994:58-9). In short, the wages paid to Mexican workers bears little resemblance to their productivity. Rather, the impact of catastrophic joblessness on worker bargaining power, the strictly enforced government policy of low wages, and the tight control exerted by government over the few unions allowed to operate, are the most widely held and believable reasons for the persistence of the rock-bottom Mexican pay scales (cf. Anderson et al 1995, Alvarez and Mendoza 1992, Hollings 1994, Marshall 1994, Shaiken 1994). Consistent with the scenario of widespread dissatisfaction over wages and conditions, turnover is rampant throughout much of the industrial sector, particularly among the maquiladoras, despite the paucity in alternative opportunities. Turnover rates as high as 200 percent to 300 percent in maquiladora plants have been reported by George (1990). Shaiken (1994) also reported high turnover rates in the four plants he studied. In the worst of them, turnover averaged at least 100 percent per year; poor wages and conditions were said to be the main causes. The problems of maquiladora jobs, in combination with their proximity to the US border, have actually been shown to stimulate border-crossings, a process engendering the term "stepping stone" migration (eg., Andreas 1994, Martin 1993, Faux 1993, Hollings 1994). As explained by Faux: (1993:6)
Martin (1993), a NAFTA proponent, provides a in-depth review of the decidedly mixed literature on this issue and cautions against exaggerated interpretations of the extent of "stepping stone." He nevertheless concludes that while its scale is probably small at present, its apparently growing and "could become a trampoline that brings more Mexican workers to the United States" (1993:126). There is little reason to believe that NAFTA will provide more attractive wages or conditions to Mexican workers, especially since labor standards were deliberately left out of the accord. Moreover, analytic studies done on the prospects for wage-convergence under NAFTA uniformly suggest that convergence, if it comes, could easily take many decades and will not necessarily consist, unilaterally, of upward movement on the Mexican side (upward convergence). Hufbauer and Schott (1992), the treaty's biggest boosters, expect that over three or four decades, Mexican per capita income could reach about half the level of the US; however, this scenario assumes their rosy predictions on NAFTA's effects on Mexican income actually come through. More critical work on the potential for Mexican wages to rise to levels approaching US standards has arrived at less optimistic conclusions (Valle Baeza 1993, Reynolds 1993). In summary, it has been argued that for NAFTA to produce a discernible impact on migration, the treaty must provide a substantial expansion in Mexican based opportunities. This point is not really controversial --NAFTA proponents claim the treaty will produce just such an expansion. But, the job creation projections by the treaty's proponents are unrealistic because they are based on dubious assumptions that, even if valid, fall far short of the mark. Second, the expected rates of pay will parallel prevailing levels, but these are currently so low that they possess little holding power despite the paucity of alternative opportunities. In the next section, support for the idea that NAFTA will fail to discernibly depress migration is provided by an entirely different set of items. Introduction | NAFTA and Migration | The NAFTA Jobs Debate: The Mexican Side | The Social Process Of Mexico-US Migration | NAFTA, Mexico-US Migration, and Michigan Communities | Conclusion | Notes | References THE SOCIAL PROCESS OF MEXICO-US MIGRATION There are a number of features associated with the process of Mexico-US immigration, particularly undocumented migration, which will strongly undercut the hypothesized deterrent effect of NAFTA on labor migration. The more general defining characteristics of Mexican migration to the US have been detailed in many treatises (cf. Browning and Rodriguez 1985, Martin 1993, Massey 1986, Massey et al 1987, Monto 1994, Muller and Espanshade 1985, Portes 1991, Portes and Borocz 1989, Waldinger 1982). Our limited interest here is in those characteristics of the movements which can be expected to mitigate the responsiveness of potential migrants to the emerging NAFTA opportunities. These are, first, that the movements are generated more by the historic establishment of social linkages to the host society, than by mere socioeconomic differences between the two places. Hence, the presence or absence of opportunities (such as those NAFTA may provide) do not, by themselves, determine the pace of migration. Second, the mass movements are socially organized and collectively implemented, rather than a mere aggregation of individual enterprises. Hence, inducing potential migrants to stay home is not easily accomplished, particularly if the economic inducements are no more attractive than those in the communities currently targeted as destinations. In general terms, migrations tend to be triggered when the more developed society intrudes into the sending society in a variety of ways, rather than just occurring in response to existing differentials in opportunities. As Portes and Borocz (1989) have noted, history is replete with examples "in which an absolute wage advantage in economically expanding areas has meant nothing to the population of more isolated regions; when their labor has been required, it has had to be coerced out of them" (1989:608). Rather, the migrations require the establishment of ties --social, political, economic-- which facilitate the initial movements and set the stage for subsequent migrations. The initial exchanges usually set into motion both the push and pull factors leading to migration. For example, the traditional means of subsistence in the sending society are often replaced by the introduction of modern forms of production from the host society. These processes generally displace the peasantry, stimulating the "push" toward alternative available opportunities. The more common initiator to the major migration waves to the intrusive society, however, is the "pull" of labor recruitment. Accordingly, labor recruitment by major agricultural concerns, railroad builders, steel mill operators, and so forth, all played a key role in initiating the early major flows from Mexico. A key facilitator of this recruitment during and after World War II was the US government's Bracero program, which permitted and regulated the use of Mexican agricultural "guestworkers" until the program was disbanded in the 1960s. Once the use of the workers became established, many factors came into being which facilitated the permanence or regularity of their presence. A major example of this concerns agricultural production. There, employers became dependent on immigrant labor because native workers had become increasingly intolerant of the undesirable conditions (particularly seasonality) and meager wages of agricultural work. Employers therefore sought and developed ways (pressuring against strict immigration controls) to insure the repeated use of such workers. Thus, a measure of permanence was created. In other instances, at least some members of the recruited workforce remained in their areas of employment after dismissal. Once a "critical mass" of the group accumulated, they became the US-based bedrock of associational linkages, or social networks, spanning the two societies. In due course, these emerging networks would attract, channel, and accommodate additional migrants. The networks serve as "bridges" for sojourners from the sending to the host communities by providing information, job-contacts, and assistance, along with assorted other forms of inducements (stories about life and fortune in "El Norte") to potential migrants. It is across such forged linkages that the massive movements unfold, and the role played by the networks can be likened to that of the underground railroad of the pre-Civil War era. Through network sponsorship, many incoming migrants find work within a few days of arrival; oftentimes jobs await them. The process becomes so regularized that it has been called a "culture of emigration" (Martin 1993). Indeed, sending communities often dispatch half or more the working aged men on at least temporary forays north and large numbers from one sending community will relocate in the very same destinations, creating, in effect, "satellite" settlements or "transnational" communities. This helps to explain why the settlement of the nation's Mexican origin population remains largely in its traditional patterning: most of the population can be found in five states of the Southwest (the so-called "big 5"), CA, TX, NM, AZ, CO, and the Chicago area. These were the areas from which the major recruitment efforts originated in the US and, therefore, most of the social-network trajectories over which migrants travel lead to these general areas. Alternatively, this helps to explain why some communities in Mexico, the so-called traditional sending areas, export so many people, while other similarly impoverished areas, export so relatively few. The former are among those areas originally tapped by the major recruiters and are, thus, deeply imbedded in the web of migratory networks. The latter were "outside the loop" on that score and their potential migrants must expend more time, effort and resources in order to launch similar forays. Of course, new linkages do form, but they are very slow to generate the volume of the mature and currently activated ones. These conditions affect the relative attractiveness of NAFTA-induced and other Mexican-based employment opportunities in a number of critical ways. Most importantly, they reduce the "costs" and uncertainties of heading north. Thus, the "returns" to workers from opportunities that hope to deter migration need to be far better than those already available. Moreover, a related and critical effect of these migration processes is that sending communities become dependent upon remittances from their northern brethren. An estimated 3.2 billion was remitted in 1990 alone (Lozano Ascencio 1993). Hence, not only do the networks reduce the "costs" of migration, but, they induce a strong dose of dependence on Yankee-style wages. Accordingly, Alarcon (1994) reported that in one traditional sending area he studied, migration patterns failed to exhibit any discernible change despite the area's hosting of substantial growth in manufacturing production. In summary, the processes by which the migrations ensue will undercut the attractiveness of NAFTA-style employment. This provides yet another reason for concluding that NAFTA cannot and will not succeed in appreciably reducing migration. Introduction | NAFTA and Migration | The NAFTA Jobs Debate: The Mexican Side | The Social Process Of Mexico-US Migration | NAFTA, Mexico-US Migration, and Michigan Communities | Conclusion | Notes | References NAFTA, MEXICO-US MIGRATION, AND MICHIGAN COMMUNITIES The state of Michigan currently ranks third among the fifty states in volume of business conducted with Mexico. There is little reason to expect that ranking to shift, although with NAFTA, the volume of business can be expected to increase. However, while it has been argued here that NAFTA will stimulate more migration than it will stem, Michigan is unlikely to be among the major receivers of the stepped-up volume in those movements. This is because the best available evidence suggests that Michigan is not currently receiving many immigrants from Mexico, and given the near certainty that future migrants will mainly venture over established trajectories, the state is unlikely to become a major receiver anytime in the foreseeable future. There are many strands of evidence supporting this generalization. The most immediate concerns growth. A recent analysis of the social and demographic characteristics of Latinos (Hispanics) in the Midwest showed that the Mexican-origin population of Michigan grew relatively slowly during the 1980s, a decade of record-breaking Mexican migration to the US as a whole. The pattern of slow growth, itself, is suggestive of little immigration. Moreover, between 1980 and 1990, Michigan's Mexican origin population grew only from about 112,000 to just over 138,000, or by slightly more than 26,000. By contrast, the Mexican origin population of the state grew by nearly 47,000 between 1970 and 1980. Thus, any Mexican migration to the state is likely to have slowed, rather than increased, during the record Mexican immigration of the 1980s. Indeed, there is good reason to believe that only a small amount of the full Mexican origin growth (73,000) over both decades was accounted for by immigrants. First, the relatively meager growth is likely to stem from the balance of two sources of growth, natural increase (excess of births over deaths) and net migration (more entries than exits), rather than just representing arrivals from other areas. Thus, the total number of inmigrants cannot possibly match, much less exceed, the 73,000 figure unless Michigan's Mexicans experienced substantial out-migration or major aberrations in typical mortality/fertility patterns. No evidence exists to support either of these possibilities. A second reason is that the portion of the growth that actually represents new arrivals may consist more or solely of "internal" migrants (from other parts of the US) than of "immigrants" (from abroad/Mexico). This is supported by the reported characteristics of the state's Mexican origin population as of 1990. Whereas the reported growth over the two decades preceding 1990 represents fully 34 percent of Michigan's 1990 population, only 10 percent of the state's 1990 Mexican origin residents were actually born in Mexico (US Bureau of the Census 1993). Finally, extensive historical work on the evolution of the Mexican presence in the Midwest indicates that by the post World War II era, both agricultural and industrial employers using Mexican origin workers primarily utilized Chicano or Mexican American workers, rather than Mexicans (Valdes 1991, Vargas 1993). Even migrant workers, whose numbers have leveled off since the 1970s, tend to come from Texas rather than Mexico. In view of the fact that nearly all of the predictable increase in Mexico-US migration is likely to ensue over the currently active trajectories, and these appear not to lead to Michigan, the state is unlikely to gain many new immigrants. Of course, new linkages do form, and a number of areas that previously held few Mexicans are now home to newly formed such communities. One example is New York City. There, a sizable Mexican community has recently begun to flower (Sassan and Smith 1992, Smith 1992). While the census enumerated some 62,000 Mexicans in residence as of 1990, one researcher estimates the true number is closer to 96,000 (Smith 1992). Moreover, within this group, sizable numbers hail from the very same communities of origin in Mexico, and the connections resemble the formation of "transnational communities," as earlier discussed. Nevertheless, the maturing of these communities, particularly their growth, ensued quite slowly until very recently (Smith 1992). Therefore, while emerging linkages between Michigan and Mexico could produce migration to the state, it will not happen rapidly. Introduction | NAFTA and Migration | The NAFTA Jobs Debate: The Mexican Side | The Social Process Of Mexico-US Migration | NAFTA, Mexico-US Migration, and Michigan Communities | Conclusion | Notes | References It has been argued here that NAFTA will stimulate more migration than it will stem. Recent events noted below, which have wreaked havoc on the Mexican economy, ensure the basic thrust of the argument--migration will increase, probably dramatically, regardless of whether Michigan receives many more immigrants. Moreover, the causes of the calamitous downturn in the Mexican economy are far more consistent with the critique advanced here than with the prescriptions of the treaty's proponents. Rather than ushering in a period of political reform and social stability in Mexico, as promised by proponents, NAFTA's effective implementation triggered precisely the opposite reaction. On the very day the treaty took effect, January 1, 1994, the world was stunned by news of the first major anti-government uprising in Mexico in decades, the Chiapas rebellion. The "Zapatista" National Liberation Army rebels cited many grievances, but chose to tie the uprising to the start of NAFTA because, to them, the treaty represented an extension of the kinds of "reforms" which were driving Mexico's workers and peasantry deeper and deeper into poverty and despair (Nairn 1994). The Chiapas incident was followed by a number of prominent political assassinations (cf. Serrill 1995, Watson 1995). While these events surely unnerved many potential investors, the single most important event undercutting the NAFTA promise has been the economic crisis unleashed by the severe devaluation of the Mexican Peso. The exchange value of the Peso fell some 50 percent between December 1994 and March 1995, a devastating reduction in purchasing power (DePalma 1995b). The resulting crisis has been catastrophic in Mexico and has sent major shocks to the US and other parts of Latin America. In Mexico, for example, inflation, interest rates, business and bank defaults, and layoffs have all shot up, while tens of thousands (possibly hundreds of thousands) of US workers also stand to lose their jobs (Levinson 1995, Myerson 1995). The Clinton administration responded by masterminding a $50 billion dollar "bailout," of which the US contributed $20 billion, but repairing the depressed state of the economy, by all accounts, will be years in the making and painful in the extreme (cf. DePalma 1995a, Golden 1995, Levinson 1995). While few might have predicted the extent of the peso's fall, the overvaluation was hardly a secret. As noted by Newsweek's Marc Levinson (1995:32), "the problems that would lead to the peso's collapse were anything but secret." The overvaluation was, during Salinas' presidency, a strategy to stimulate foreign investment. Levinson notes that both the World Bank and the US Treasury favored devaluation by as much as 30 percent, although this was not made public. In addition, the major bond rating agencies, Moody's and Standard & Poor's, each labeled investing in Mexico "a high risk credit." He goes on to quote a California University professor's candid assessment that "Anyone who conducted business in Mexico or traveled in Mexico ... knew the peso was overvalued" (1995:33). Similarly realistic accounts of weaknesses in the Mexican economy could be found in the work of NAFTA opponents, particularly those focused on Mexican issues (cf. Cypher 1993, Faux 1993, Hollings 1994). In much of the key literature supporting NAFTA, however, almost no references to these problems are made. For example, in the two major works on NAFTA by the treaty's most celebrated boosters, Hufbauer and Schott (1992, 1993), not a single reference to the potential perils of the peso's overvaluation can be found. Indeed, the authors instead advance the fantastic claim that NAFTA would lead to a real increase in the peso exchange rate of nearly 30 percent! Likewise, not a word is devoted to the grave social costs of Mexico's economic liberalization strategy, of which NAFTA is part and parcel, let alone the potential for social unrest that such a strategy undoubtedly entailed. Despite the oversights of NAFTA proponents, one thing is certain: the economic crisis at hand will increase migration. On the basis of interviews of Mexican workers and US immigration experts, the Wall Street Journal published a story detailing the consensus view that migration will increase, perhaps as much as 25 percent from traditional sending areas (Rose and Solis 1995). Moreover, even before the crisis unfolded, the failure of NAFTA to produce a rate of job-growth sufficient to dampen migration was evident. As Anderson et al (1995) have noted, during the first six months of 1994, the entire (formal) Mexican economy expanded by only around 84,000 jobs --far, far fewer than the treaty's promoters implied or promised. In addition, the displacements from agricultural production tied to NAFTA will add even more pressure to migrate. The conclusion here, therefore, is that NAFTA can only lead to more, rather than less migration. However, due to the nature of the migration processes, relatively few migrants should find their way to areas in the US not currently receiving Mexican immigrants. For this reason, Michigan is unlikely to host much increased immigration even if the state enjoys increased trade or capital exchanges with Mexico. Given the current crisis in Mexico, however, even the potential upsurge in business may itself be in jeopardy for some time to come. Introduction | NAFTA and Migration | The NAFTA Jobs Debate: The Mexican Side | The Social Process Of Mexico-US Migration | NAFTA, Mexico-US Migration, and Michigan Communities | Conclusion | Notes | References 1. Demand for immigrant labor will continue indefinitely in both agricultural (Martin 1993, 1994, Mines et al. 1992) and urban settings (cf. Muller 1991, Muller and Espanshade 1985, Piore 1979, Sassan 1992, Sassan and Smith 1992). In agricultural work, the seasonality, low wages, and increasingly distant locales (from native population concentrations) have led to increasingly lower proportions of native workers, but more immigrants (Martin 1993, Mines et al. 1992). In addition, the "Bracero Program," a US government guestworker program promoting use of Mexican workers by Southwestern growers, established links that facilitated immigrants' use even after the program's demise, afterwhich the mere availability of the cheaper labor influenced the structuring of the industry away from mechanization. As Martin notes, "Agricultural engineers note that machines are available to harvest practically every fruit and vegetable grown in the United States, but that machines replace hand-pickers only when it is economically rational to make the switch--that is when the cost of machine harvesting is cheaper than the cost of hand harvesting" (1993:80), a process not in the immediate offing. Finally, increased acreage devoted to growing, including vast areas under preparation for future harvesting, all foster increased demand for agriculturally based immigrant labor (Martin 1994). Demand for urban workers willing to take low quality jobs will also increase because advanced economies cannot easily eradicate or make attractive low level jobs (Piore 1979, Sassan 1992). Jobs that are highly unstable, that entail little social status (eg, dog walkers, dishwashers) and, especially, that entail very low wages and few upward trajectories (eg, household help), increasingly attract fewer native workers, but suffice to draw immigrants. Factors such as the rise of single-parent and dual earner households, for example, give rise to increased demand for some such workers (eg, household help, child care), while in other cases, the mere presence of the cheaper workers is likely to influence demand, as with agriculture. 2. Hufbauer and Schott (1992) cite an obscure study which projected that NAFTA would produce 1.5 million jobs in Mexico (KPMG Peat Marwick 1991). However, it is not widely available and Hufbauer and Schott's description of the study do not make clear when the target would be reached or if the estimate took account of the expected rural displacement of Mexican producers. 3. Maquiladoras are mostly foreign owned (mostly US) or operated (Mexican owned-US operated/joint ventures) plants in Mexico, mostly near the US border, authorized by a special treaty. They mainly process or assemble imported products for reexport to the US, capitalizing on the workers' low wages and favorable US tariff treatment (Hufbauer and Schott 1992). By the 1990s, most US owned maquiladoras replaced or absorbed the expansion of similar productive facilities in the US, thereby discernibly shifting some US employment to south of the border (Barlett and Steele 1992). Introduction | NAFTA and Migration | The NAFTA Jobs Debate: The Mexican Side | The Social Process Of Mexico-US Migration | NAFTA, Mexico-US Migration, and Michigan Communities | Conclusion | Notes | References
|
|
|
|
For more information, contact: |
