This paper analyzes migration from Mexico to the United States over the past decades and seeks to understand why it began to fall starting in 2008. After providing some background on the historical trends of labor migration to the United States, the article focuses on the place of Mexicans and other Latin Americans in the United States labor market as the context that frames Mexican migration. This work then analyzes the impact of the 2008-2009 recession on Latin American laborers, especially Mexicans, and how migration flows from Mexico have evolved in recent years. The conclusion derived from the analysis of the data and trends observed is that the decline in Mexican migration starting in 2008 can primarily be explained by the behavior of labor demand in the United States, largely determined by the recession.

INTRODUCTION

The Mexicans who chose to stay in United States territory when the Treaty of Guadalupe Hidalgo was signed in 1848 became the first Mexican immigrants to settle in the lands of Mexico’s northern neighbor. Migration flows between the two countries, which persist into the present day, have gone through a series of phases, with marked fluctuations in magnitude and intensity, a response to both structural and contextual changes on both sides of the border.¹ Starting in the final decades of the twentieth century, the tremendous growth in the number of Mexicans immigrating to the "North"—as many migrants commonly refer to the United States—turned this flow into the largest migration corridor (in terms of bilateral movement) in the contemporary age (IOM, 2013).

The greatest relative rise in the number of Mexicans living in the United States took place between 1970 and 1980, when growth reached 189%, and the figure went from 760,000 to nearly 2.2 million (Passel et al., 2012). Meanwhile, the largest absolute jump happened between 1990 and 2000, when the number of Mexicans went from 4.5 to 9.4 million. There was also a rather significant increase of 3.1 million people between 2000 and 2007, when the total rose from 9.4 to more than 12.5 million.

Recent data from BBVA Research indicate that 51.1% of immigrants living in the United States in 2013 arrived in 1996 or later, while 24.1% came to the country in the decade prior (1986 to 1995), and the remaining 24.1% immigrated before 1986 (Li Ng and Ramírez, 2014). Passel and Cohn (2011) estimate that on average, 500,000 unauthorized Mexicans came to the United States annually between 2001 and 2006, although flows declined markedly between 2007 and 2009. Based on these figures, the total number of Mexicans living in the United States began to decrease starting in 2009 or 2010 and now wavers around 12 million; it has yet to return to its previous peak. The explanation for such a remarkable and abrupt change in the trends that prevailed over the past three decades has sparked debate in the academic and political worlds, and even among the media, in both countries.

Particularly striking are statements made regarding the zero net migration rate and the factors behind this phenomenon (see Cave, 2011; Massey, 2012; Passel and Cohn, 2012). These declarations have elicited diverse responses from some Mexican scholars, who have claimed that the facts need to be analyzed in greater detail to prevent drawing erroneous conclusions (Durand, 2011; García Zamora, 2012; Alarcón, 2012; Aragonés, 2011 and 2012). The purpose of this paper is not to rehash the terms of the debate surrounding "zero net migration," but rath er to make some observations in this regard. The main objective of this article is therefore to explore factors that better explain the role that Mexican migrants have played in the United States labor market, and the shifting labor demand for these migrants, which, in the perspective of this paper, has been the principal determinant of migration flow dynamics over the past decades.

BRIEF CONSIDERATIONS REGARDING THE "ZERO NET MIGRATION RATE"

First and foremost, it should be noted that in the phrase "zero net migration rate," the word "net" is just as important as the word "zero." In the first decade of the twenty-first century, the rate of deportation for unauthorized or "undocumented" immigrants rose by over 100%. Nearly 400,000 people were deported in 2009, of whom more than 70% were Mexicans (Passel and Cohn, 2011). If we multiply the percentage of Mexicans deported in 2012 (69.5%) and 2013 (65.5%) by the total number of people deported from 2008 to 2013, we can estimate that somewhere between 1.5 and 1.6 million Mexicans were deported over this six-year period (TRAC, 2014; ICE, n.d.). Based on an average of 250,000 new immigrants every year, and the drop in employment that led an additional number of Mexican to return to their country of origin voluntarily, we reach the "zero net migration rate."

It is also relevant to mention that both Massey (2012) and Passel and Cohn (2012) have focused their analyses mainly on the sharp drop in the number of unauthorized immigrants. They also point to the simultaneous slight increase in the number of Mexicans entering the United States with visas for temporary workers.

On the other hand, despite the decreasing birth rate and the more or less favorable behavior of macroeconomic indicators in Mexico, which should have offered better labor opportunities, the persistence of poverty and a precarious labor market continued to produce conditions that encouraged emigration. In the point of view of this paper, the reason behind the radical reversal, beginning in 2008, resides in the economic, political and social conditions in the United States, which reduced the demand for Mexican immigrant labor, and were the principal drivers behind the declining migration flow. To support this statement, this paper will take a brief look at historical labor migration trends to the United States and the impact of the 2008-2009 crisis on Latin American laborers, and more specifically, Mexican immigrants.

BACKGROUND: HISTORICAL TRENDS OF LABOR MIGRATION TO THE UNITED STATES

Throughout history, attracting immigrant labor has been fundamental to the economic development of the United States. To a certain extent, migration flows have adjusted to the ebb and flow of economic activity in the country. The industrial boom at the end of the twenty-first century marked the beginning of various decades of rapid growth in immigration, both in relative and absolute terms. By 1910, the figure of 13.5 million foreign-born people living in the United States hit a record high in percentage terms: 14.7% of the population (Gibson and Lennon, 1999).

In 1921, the first quantitative restrictions on the arrival of new immigrants were enacted. Even so, the total number of immigrants continued to rise, reaching 14.2 million people in 1930 (11.6% of the total population), until the Great Depression in the 1930s reversed the trend. From that moment forward, due to the combined impact of reduced overall migration flows, death, the deportation of many Mexicans and the voluntary return of some other groups to their countries of origin, the number of immigrants living in the United States fell to 9.6 million in 1970, or around 4.7% of the total population, which had meanwhile grown from 122.8 to 203.2 million people (Gibson and Lennon, 1999).

The 1970s once again ushered in a new trend, as the number of immigrants began to grow rapidly, surpassing 38 million in 2007, when the figure constituted 12.6% of the total population and 15.7% of the Economically Active Population (EAP). Besides these shifts in the number of immigrants, which in some measure have reflected overall economic growth trends in the United States, the primary origin countries of immigrants have also evolved. From the early days of colonization to its subsequent development as an independent nation, the United States was principally populated by European immigrants and their descendants. Native groups were pushed further west and, in many cases, to the brink of extinction. In the first half of the twentieth century, Europeans, mainly from Italy, Germany and the United Kingdom, were the predominant immigrant group, up until the 1970s, when new trends began to appear in migration flows, not only in the United States, but also on the global level.

Among other factors, the reconstruction period after the Second World War and the subsequent consolidation and expansion of the European Union led to growing prosperity for many European countries, which slowed the outflow of emigrants and turned many of these countries into destinations for migrants from other parts of the world. The unfavorable evolution of terms of exchange for the export of primary products from many Latin American nations, in addition to widespread economic difficulties and population growth, in turn, made many of these countries net population exporters. Moreover, political regimes and armed conflict in parts of Central America, Asia and Eastern Europe led to emigration from these countries at the end of the twentieth century.

As such, at the dawn of the twenty-first century, the top origin region of immigrants to the United States was Latin America, followed by Asia, and to a lesser degree, Europe. In 2007, 31% of immigrants came from Mexico, followed by various other countries of origin: Philippines, India and China, with 4% each, and then El Salvador, Vietnam, South Korea and Cuba, with 3% each. At that time, Canada accounted for only 2% of immigrants. According to data from the Migration Policy Institute (MPI), the top ten countries of origin did not include a single European nation, and altogether, European immigrants only constituted about 13% of foreign-born people living in the United States (MPI, 2009; U.S. Census Bureau, 2008). At the same time, around 27% of immigrants were Asian and around 54% were Latin American.

As the United States population began to age rapidly at the beginning of the twenty-first century, immigrants in general, and especially Latin American immigrants, became an increasingly important component of the labor force. In fact, the growth rate of the EAP in the United States has been declining since the 1960s, when it rose 29.2% as baby boomers born in the post-war period aged up and women joined the workforce. In the 1990s, the EAP grew only 11.5%; however, without the addition of new immigrants, this increase would only have been 5%. Between 2001 and 2006, there was a similar trend; that is, half or more of the rise in the EAP was due to immigration. There is rather widespread consensus among economists that without new immigrants to feed the EAP of the United States, not only would job growth have been restricted, but also economic growth in the United States during the years described (Sum et al., 2002; Council of Economic Advisers, 2007).

In the specific case of Mexicans, following the 2001 recession, and in spite of enhanced border security after September 11 that same year, the number of migrants reaching the United States continued to grow until mid-2006. Moreover, during this time period, there was a much higher incidence of undocumented, as compared to authorized, immigrants (Passel and Cohn, 2009). However, the number of undocumented Mexicans arriving to the United States has fallen sharply in recent years (2007-2013), not because living conditions have improved in Mexico, but rather because job prospects in the United States have diminished. The 2008-2009 recession has, at least for the time being, slowed the arrival of migrant workers, especially undocumented laborers. These fluctuations are a sign of the growing complementarity, and in some sense, de facto integration, of the Mexican and United States labor markets. Migration flows from Mexico adapt, in large part, to the conditions of labor demand in its northern neighbor.

In addition, the labor market in the United States has undergone radical changes over the past decades as a result of the ways in which the country has dealt with globalization and growing international competition. Broadly speaking, both skilled and unskilled jobs have become less stable and many have become frankly precarious. To reduce the cost of labor, many companies have restricted wages, while technology innovations have made it possible to eliminate thousands of positions. But even as traditional manufacturing jobs are disappearing, others have been added in countless new service sectors. In general, the patterns of labor supply and demand have been drastically altered. Although many of the new jobs created in the United States are undesirable to those who have been displaced from their manufacturing positions, which paid fairly well, there is a constant and growing supply of Mexican immigrants willing to accept nearly any type of job that will give them an income in dollars.

The spectacular economic growth achieved by the United States in the 1980s and 1990s generated a boom in the labor demand on both ends of the spectrum; that is, for highly skilled and unskilled jobs. The United States has managed to attract a wide range of professionals from all around the world. In recent years, Asia stands out as the principal region of origin for highly skilled workers. Unskilled labor, on the other hand, tends to come from Mexico and includes an indeterminate, but considerable, number of undocumented immigrants (Levine, 2012; Passel and Cohen, 2009, 2011). In fact, the issue of undocumented immigration has been at the heart of some of the most critical immigration policy debates in the United States in recent decades. Despite rhetoric against the trend, estimates of the number of undocumented workers indicate that the practice of employing them when and where it is convenient is fairly widespread.

Some immigrant groups have clearly gravitated towards certain labor niches. While the majority of Asians work in technical and professional realms, Mexicans and other Latin Americans are concentrated in construction and less-skilled sectors, such as manufacturing and, above all, services. There are also distinct patterns in terms of the places of origin of immigrants, the human capital they possess and the income levels they can achieve in the United States. The greater income levels of European and Asian immigrants match their higher levels of education, even with respect to the American-born population, and therefore reflect a favorable position in the labor market. By contrast, the high percentage of undocumented immigrants and generally lower levels of schooling place the majority of undocumented immigrants in extremely vulnerable circumstances with regard to working conditions and wages.

GENERAL CONTEXT: MEXICANS AND OTHER LATINOS IN THE UNITED STATES LABOR MARKET

Because the principal motivation to come to the United States is to obtain a job that pays in dollars, it is unsurprising that Mexicans² and their descendants are the group with the highest EAP participation rate in the United States, at 66% in 2013 (USDOL, Tables 5 and 6, 2014). The rate of 77.3% for Mexican men was far higher than that of any other group (Whites, 70.5%; African Americans, 63.5%; Asians, 73.0%). Although the rate for Mexican women, at 54.3%, was somewhat lower than that of other groups (Whites, 56.9%; African Americans, 59.2%; Asians, 57.1%), it was still higher than the EAP participation rate of women in Mexico, which was a mere 42% in the first half of 2014 (INEGI, 2014). These trends have held true over the past three decades or even longer.

At the same time, the unemployment rate for Latinos of Mexican origin and Latinos in general has been higher than that of Whites³ and lower than that of African Americans. In 2013, the rates recorded were: Whites, 6.5%; African Americans, 13.1%; Latinos, 9.1%; Mexicans, 9.0%; Puerto Ricans, 13.6%; and Cubans, 7.7% (USDOL, Tables 5 and 6, 2014). Unemployment rates reflect the ups and downs of overall economic activity, falling and rising, respectively, in response. In general, the unemployment rate among Mexicans has been lower than that of Puerto Ricans and higher than that of Cubans. Cubans tend to experience an unemployment rate slightly below that of the total white population (which, as has been mentioned, includes the majority of Hispanics).

Nor should it come as a surprise that recent immigrants hold the least desirable jobs with the lowest wages in the United States, although these jobs do pay much more than what they could earn in their countries of origin. Such has been the experience of the vast majority of Mexican immigrants, as they frequently have little schooling and poor knowledge of English. These educational disadvantages persist into the second and third generations and affect the work opportunities available to Latinos of Mexican origin born in Mexico.

United States Department of Labor data (see Table 1) reveal that in 2013, the Mexican-origin, that is, immigrants and their American-born descendants, was distributed fairly uniformly among four of the five principal employment categories: 20.3% in sales and office occupations; 18.0% in production, transportation and material moving; 17.7% in natural resources, construction and maintenance; and 17.3% in management, professional positions and other related occupations. Their share of the latter sector was markedly lower than that of any other ethnic or racial group. Meanwhile, the share of Mexican-origin workers in the services sector (26.7%) was the highest of any group, slightly above that of African Americans and also Puerto Ricans. Only 2.6% of Mexican workers were employed in activities related to agriculture, fishing and forestry, which since 2004 has no longer appeared as an independent category, because it employees less than 2% of the total EAP, but is still a much higher percentage than that of any other group. More than 40% of agriculture and livestock workers at the national level are Latino (USDOL, 2014).

Looking at the subcategories of the principal categories (shown in Table 1), 9.8% of Mexicans were employed in professional and related occupations. A similar percentage, 9.5%, work in manufacturing, and 11.3% are employed in construction and extraction jobs; these two sectors offer some well-remunerated jobs for highly skilled and experienced workers, but the majority are low-wage and unskilled positions. Another 10.8% hold jobs as office workers and administrative support. This category includes many sectors in which women predominate and wages tend to be low. The same is true of the sales area, which accounts for 9.5% of Mexican workers. Meanwhile, 9.9% and 9.1%, respectively, work in food preparation and service or building and grounds cleaning or maintenance. Wages are very low in these positions. Within each of the broader categories, Mexicans and other Latinos tend to be concentrated in a few sectors: certain specific branches of light manufacturing—rather than heavy—, building and grounds cleaning and maintenance, food preparation and service, cashiers in self-service stores and retail sales, and specialized positions in bricklaying, etc., to name a few.

Between 1990 and 2007, the share of Latino workers in the EAP rose from 7.5% to 14.3%. Data by industry (see Table 2) show that at the same time, some sectors came to depend more and more on Latino labor. In the time period described, the percentage of Latino workers in the following branches at least doubled: retail baking grew from 9.0% to 22.3%; industrial baking went from 13.0% to 31.7%; cement production and related products changed from 8.2% to 19.9%; farming and forestry jobs from 15.4% to 37.1%; barber shops went from 10.0% to 21.5%; drycleaning and laundry services rose from 14.6% to 31.2%; and animal slaughtering and processing from 17.0% to 35.2% (USDOL, 1991, 2008-2014).

However, the most spectacular increases took place in construction and carpet and rug mills. The percentage in construction grew from 8.5% in 1990 to 25.3% in 2007, when the sector employed approximately three million Latino workers. Because the crisis had a strong impact on the sector, by 2010, the number of Latino workers had fallen to 2.2 million, accounting for 24.4% of all employees in the industry. In other words, although the crisis had a substantial effect on the number of people employed in the construction industry, the percentage of Latinos did not fall significantly.

In carpet and rug mills, the share of Latino labor rose from 10.1% in 1990 to 29.4% in 2007. It fell sharply to 19.2% in 2008 and then shot up again to reach 49% of all 59,000 sector employees in 2010. When the total number of employees fell in 2011 and again in 2012 to 50,000, the percentage of Latinos dropped to 30%. The share of Latinos then grew again in 2013, when the total number of employees rose to 55,000. In any event, although the sector employs a low number of people, in Dalton, Georgia, known as the "carpet capital," the core of the industry in the United States, Latinos make up more than a third of the local population.

The occupational and industrial concentration of Latinos is closely intertwined with their geographic distribution. At the beginning of the twenty-first century, 75% of the Latino population was located in only seven states. However, a collection of states in the Southeast—where the Latino population is still fairly small—experienced tremendous growth rates of more than 200% to nearly 400% between 1990 and 2000 in the number of Latinos residing there, precisely due to the job opportunities they offered. Mexicans and other Latinos are frequently actively recruited to work in jobs in the meatpacking, chicken processing or carpet and rug industry, positions that are undesirable to local residents. To consolidate this type of market niche, it would seem that a flow of Latino immigrants and jobs that nobody else wants to do or wages that others would not accept are sufficient. The same is true of agricultural workers in states such as California, Texas and Oregon. The demand for labor to work in these undesirable and poorly remunerated jobs grew discernibly at the end of the twentieth century simultaneous to the arrival of new waves of immigrants from Mexico and other Latin American countries who were more than willing to take them on.

The most disaggregated statistics on occupation only record the percentage of "Hispanics or Latinos" with respect to the total number of people employed in the category. They do not distinguish among the various groups that make up the Latino population or between immigrants and Latinos born in the United States, such that it is difficult to analyze the various occupational profiles in greater detail. What is evident, though, based on the information available, is the percentage, and therefore the number, of Hispanic workers in each sector of the list of occupational categories detailed and published by the United States Department of Labor.

Given the preponderance of Mexicans in the set—nearly two-thirds of Hispanic workers are of Mexican origin—and the fact that discrepancies with the other two principal groups, Cubans and Puerto Ricans frequently run counter to each other, the data for the entire Hispanic population is a good approximation of the labor insertion of Mexicans. Moreover, slightly over half (51%) of Latinos who have held jobs in the United States in recent years are immigrants (Motel, 2012).

If the data is analyzed over various years, it emerges that the occupations employing the highest numbers of Latino workers are primarily unskilled, low-wage jobs that do not require higher education. The same is true of occupations with relatively high percentages of Latino workers (that is, rather higher than the total percentage of Latinos among employed people) (Levine, 2001, 2008, 2010). Moreover, the vast majority of jobs with the highest percentages of Latino workers in 2007, and also in 2013, also had high percentages of undocumented workers, right before the outbreak of the financial crisis at the end of 2007 (Passel, 2006).

IMPACT OF THE CRISIS ON LATINO WORKERS

During the most severe years of the crisis (2008-2009), Latinos lost a proportionate part—14%, equivalent to 863,800 jobs—of the 6.2 million jobs destroyed in the United States in those years. The most striking case was the construction industry, where Latinos lost 720,000 jobs. Contrary to the above, their participation in some sectors actually saw a slight increase, surely due to the lower cost of employing Latinos, but the group suffered losses in many occupations where the concentration of Latinos had been high. Table 3 displays the net changes in the number of Latinos employed by industrial category from 2007 to 2010 and between 2007 and 2013. In 2010, some sectors began to recover slightly and Latinos regained 300,000 jobs, although total employment continued to fall with additional losses of 813,000 jobs, such that the Latino share of the labor force grew to 14.3% and continued to rise to 15.6% in 2013 (USDOL, 2008), 2009, 2010, 2012, 2013, 2014).

An analysis of employment figures with a breakdown of occupational categories (see Table 4) reveals that in the period 2007-2010, Latinos suffered net losses in jobs in the majority of occupations with high numbers of Latino workers. Annual figures reveal that in general, the most drastic losses took place between 2008 and 2009, although the patterns of behavior of each occupation varied and are surely tied to discrepancies in regional and local behavior, with variations for different population groups, and between Latinos and Mexican immigrants. Undoubtedly, the specific behavior of many occupations and industries, and their impacts on specific groups within the EAP, deserve a much more detailed analysis than what would be possible in the space available in this article.

As indicated previously, the decline in the number of jobs for Latinos began to turn around in 2010, even though total employment continued to fall. Starting in 2011, there were successive increases in total employment, but it did not reach pre-recession levels until mid-2014. By 2012, the number of Latinos with jobs had already surpassed the previous peak reached in 2007. Rakesh Kochhar et al. (2010) asserted that in the first year of economic recovery, starting in July 2009, the unemployment rate for immigrants started to fall slightly (down 0.6%) even as the rate for American-born workers continued to rise (up 0.5%). In any event, by mid-2010, the number of immigrants employed remained below pre-recession levels. This slight increase in immigrant employment saw a simultaneous decrease in income of 4.5%, as compared to only a 1% decrease in the income of American-born workers. Moreover, it turns out that Latino immigrants suffered the greatest losses in terms of income; their weekly wages fell 1.3% from 2008 to 2009 and an additional 5.8% between 2009 and 2010 (Kochhar, 2010: 7).

MEXICAN IMMIGRANTS

As argued above, the overall labor situation of Latinos is a good mirror for the conditions facing Mexican-origin workers, which includes both immigrants as well as American-born workers. However, there are significant differences in the socioeconomic indicators of the two groups that constitute the Mexican-American population. Data published by the Pew Research Center in 2011 helps draw comparisons between Latinos as a whole, Latinos of Mexican origin born in the United States and Mexican immigrants (González-Barrera and López, 2013). As might be expected, the majority of the immigrant group is disadvantaged when it comes to speaking English, level of education and in many cases, irregular immigration status, which restricts opportunities to access more skilled and better paying jobs.

The percentage of men is slightly higher among immigrants and their rate of participation in the EAP is also higher (see Table 5). The percentage of people employed is greater, and therefore the unemployment rate is lower among immigrants, although in both cases it is above the general population rate. In 2011, the overall unemployment rate was 8.9%, while for Mexican immigrants, it was 10.3%, and it was even higher among the population of Mexican origin, 14.1%. It is clear that employment is the principal reason that immigrants want to live in the United States and is also a key factor in keeping them in the country. Many of those who lose their jobs are forced to return to Mexico, as happened during the most severe years of the recession, in 2008-2009.

Immigrants lag other groups in terms of education, which negatively impacts their occupational profile and income level. They are much more concentrated in construction, manufacturing, agriculture and unskilled service sectors, such as cleaning, maintenance, etc. However, people of Mexican origin are much more likely to be employed as professionals and sales managers and in administrative positions than Mexican immigrants. The average income of Mexican immigrants is lower than that of the group of Mexican origin and Latinos as a whole. The differences are even more striking for people holding full-time, year-round jobs. The difference is fairly marked if we look at households, despite the fact that the households of immigrants tend to contain more people with jobs.

Following the mass deportation of Mexicans in the 1930s, due to the Great Depression, non-temporary migration to the United States remained rather stagnant for decades. But as the United States labor market began to transform in the 1970s, migration flows picked up once again, such that starting in the 1980s, the arrival of new immigrants was the principal growth factor for the Mexican population in the United States. This trend held strong for two decades. However, beginning in the twenty-first century, despite the fact that the migration flow grew strongly in absolute terms, the growth of the Mexican population born in the United States once again surpassed the increase due to the arrival of new immigrants. Heightened challenges to cross the border following September 11, 2001 put the brakes on circular migration and created an incentive for immigrants to stay longer or more permanently in their destinations. This fact encouraged family members who had remained behind in Mexico to try to cross the border, and led to a significant spike in the birth rate among Mexicans now residing in the United States. Currently, approximately 65% of the Mexican-American population was born in the United States, and only 35% was born in Mexico (Pew, 2012, 2013). Even so, in 2011, it was estimated that the number of Mexican immigrants in the EAP was still slightly higher than the number of Mexican descendants born in the United States (7.635 million versus 7.557 million).

It is undoubtedly difficult to accurately count people who go out of their way not to be counted, which is the case for unauthorized or “undocumented” immigrants. Recent estimates signal that around 55% of the approximately 11.1 million unauthorized immigrants living in the United States are Mexican (González-Barrera and López, 2013). As such, a little over half (51%) of Mexican immigrants are unauthorized, while 32% are legal permanent residents and 16% have obtained United States citizenship. Even so, it is challenging to gather precise data on the number of people born in Mexico and currently living in the United States and to know with certainty how migration flows have behaved in recent years. The U.S. Census Bureau and other reputable sources, such as BBVA Research and Pew Research—which calculate estimates using data from the U.S. Census Bureau—have come up with figures that vary by the hundreds of thousands and have identified changes in trends starting at different dates (see table 6).

In any event, there is some consensus among experts (Massey, 2010; Passel et al., 2012) that the massive influx of Mexican migrants that the United States experienced in the early years of the twenty-first century came to a halt sometime between 2006 and 2010 and that, as a result, the total number of Mexican immigrants residing in the United States has relatively stagnated since then. The most cited reasons for this shift in the trend from decades prior are the heightened costs and challenges associated with crossing the border since 2001, in addition to the increase in deportations beginning in 2008, and, perhaps more than anything else, declining job prospects following the crisis that began in 2008-2009. Moreover, these two events—the terrorist attacks and the economic crisis—have both exacerbated anti-immigrant sentiment in some sectors of the American population and fostered a hostile environment for immigrants from Mexico and other Latin American countries.

Data provided by the U.S. Census Bureau (2014) reveal that the number of Mexican immigrants in the American EAP fell from 7.523 million in 2007 to 7.483 million in 2008. It saw a slight recovery in 2009 (7.515) and by 2010 (7.582), it had already surpassed 2007-levels. It fell again in 2011, to 7.403 million, and then spiked to 7.681 million in 2012 (see Table 6). Employment numbers among Mexican immigrants also rose and fell during this time period, going from 7.108 million in 2007 to 6.510 million in 2009, before rising slightly to 6.628 million in 2010 and falling once again to 6.530 million in 2011. The figure rose again to 6.892 million in 2012, but has yet to reach 2007 levels.

The total changes appeared differently at the level of occupation structure. The most notable change was the absolute and relative decrease in construction jobs, the realm most affected by the recession, and the sector from which, according to these data, Mexican immigrants lost around 670,000 jobs. In production, the decrease amounted to 80,000. Losses were less than 10,000 jobs in sales and installation, maintenance and repairs. At the same time, employment in services rose by 329,000. In the rest of the occupational categories: management, business and finance; office and administrative; and farming, forestry and fishing, the employment increase ranged from 20,000 to 60,000 jobs. Employment data by industry reflect the same trends. In summary, in overall terms, job opportunities for Mexican immigrants in the United States fell between 2007 and 2012. As of yet, there is still no clear sign of sustained recovery in the demand for unskilled immigrant labor (see Table 7).

CONCLUSIONS

This paper has demonstrated that from the 1970s to 2007, Mexican immigration to the United States grew vigorously. Starting in 2008, the rate began to stagnate, or even decline, and the number of Mexican immigrants living in the United States fluctuated, figures that vary depending on the source consulted. Keeping in mind that migration flows between Mexico and the United States and how they fluctuate are due to the interaction of a variety of economic, political and social factors and the way in which they manifest themselves, both in the short and long term, on both sides of the border, trying to isolate any single causal explanation for this phenomenon would not make sense.

However, this article considers that the most recent trends observed in Mexican migration can primarily be explained by the behavior of U.S. labor demand. The changes in attitudes towards immigration and stricter immigration policy following the September 11 attacks were not enough to make a dent in the arrival of Mexican immigrants. In fact, given the growing demand for cheap labor during a time period of strong economic growth, and despite heightened challenges and higher costs to enter the United States without a visa, immigration numbers hit record highs at the end of the twentieth century and in the first half-decade of the twenty-first. It was only starting in the 2008-2009 recession that employment dropped sharply and immigration slowed.

Officially, the recession, which has been called the most severe since the Great Depression, ended in June 2009. Macroeconomic growth has stayed weak since then, however. At some times, there have even been fresh fears of a relapse. In fact, growth declined by 1.5% in the first quarter of 2011, followed by a positive rate of 2.9% in the second quarter and once again a decrease of 2.1% in the first quarter of 2014, followed by a second-quarter jump of 4.3% (U.S. Bureau of Economic Analysis, 2014). It was not until five years later, in June 2014, that the number of jobs returned to its 2007 high. By then, the unemployment rate (6.1%) was still greater than that of 2007 (4.6%) and the share of EAP (62.8%) was lower than the figure of pre-recession years, around 66%.

In 2012, the number of Mexican immigrants in the EAP of the United States was a little bit higher than what it was in 2007. Even so, the number employed in 2012 was significantly lower and the unemployment rate among immigrants was still double what it was before the recession. If Mexican immigrants living in the United States are the principal source of information for their relatives and countrymen in Mexico about the job supply for new immigrants, it is reasonable to assume that they would also let them know about the lack of jobs and growing deportations. Any current measurement of the number of immigrants, or any statement on their intentions and motivations to immigrate, must take into account that between 2008 and 2013, approximately 1.5 to 1.6 million Mexicans were deported from the United States.

Regardless of the behavior of macroeconomic indicators, we must also pay mind to the precarious state of employment in Mexico and wage discrepancies between the two countries as other factors that motivate immigration. Official sources recognize that the rate of labor informality was 58.78% in July 2014. In fact, 27.17% of people with jobs worked in the informal sector and another 31.61% more did not have any formal labor link to their employers (INEGI, 2014b). At the beginning of 2014, more than 80% of wage earners in Mexico received an income below half of the United States minim wage (calculations made by the author based on data from INEGI, 2014a). According to official data for 2012, 53.3 million Mexicans, or 45.5% of the population, were living in “poverty” (Coneval, 2013). This publication also reported a total of 60.6 million people, or 51.6% of the population, earning an income “below the welfare line.”

In light of these conditions, it is hard fathom exactly why the supply of cheap labor from Mexico fell significantly beginning in 2008, when the number of Mexican immigrants living in the United States began to stagnate. However, the substantial reversal came about starting in 2008 precisely because the labor conditions in the United States changed. The economic crisis also further exacerbated anti-immigrant sentiment, already percolating under the surface in certain sectors of American society, and fostered a politically and socially hostile environment for many immigrants, especially undocumented Mexicans. It remains to be seen how the Mexican labor supply and the labor demand in the United States will interact once the American economy has fully recovered.

BIBLIOGRAPHY

Alarcón, Rafael (2012), “El debate sobre la migración cero”, in Letras Migratorias, Newsletter, Observatorio de Migración Internacional, December 2012 (consulted February 9, 2015), available at: http://www.omi. gob.mx/work/models/OMI/Resource/821/1/images/Newsletter_Alar-con_Rafael_10_12_2012.pdf

Aragonés, Ana María (2011), “¿De qué país están hablando?” La Jornada, July 16, http://www.jornada.unam.mx/2011/07/16/opinion/019a1pol

_______ (2012), “¿Migración cero?” La Jornada, October 9, 2012, http://www.jornada.unam.mx/2012/10/09/opinion/021a1pol

Cave, Damian (2011), “Better Lives for Mexicans Cut Allure of Going North”, New York Times, July 6, 2011, http://www.nytimes.com/ interactive/2011/07/06/world/americas/immigration.html

Consejo Nacional de Evaluación de la Política de Desarrollo Social-Coneval (2013), “Pobreza en México. Resultados de pobreza en México 2012 a nivel nacional y por entidades federativas” (consulted January 7, 2014), available at: http://www.coneval.gob.mx/Medicion/Paginas/Medici%C3% B3n/Pobreza%202012/Pobreza-2012-aspx

Council of Economic Advisers (2007), “Chapter 9, Immigration”, in The Economic Report of the President 2007 (consulted November 29, 2007), available at: http://www.gpo.gov/fdsys/browse/collection.action? collectionCode=ERP

Durand, Jorge (2011), “Migración a la baja”, La Jornada, January 17, 2011, http://www.jornada.unam.mx/2011/07/17/opinion/019a2pol

Durand, Jorge, and Douglas S. Massey (2003), Clandestinos: migración México- Estados Unidos en los albores del siglo XXI, Mexico, Universidad Autónoma de Zacatecas and Miguel Ángel Porrúa.

García Zamora, Rodolfo (2012), “Cero migración: Declive de la migración internacional y reto del empleo nacional”, Migraciones Internacionales, vol. 6, no. 4, July-December, 2013, pp. 273-283.

Gibson, Campbell J. and Emily Lennon (1999), “Historical Census Statistics on the Foreign-born Population of the United States: 1850-1990” (consulted September 29, 2012), available at: http://www.census.gov/ population/www/documentation/twps0029/twps0029.html

González-Barrera, Ana and Mark Hugo López (2013), A Demographic Portrait of Mexican-Origin Hispanics in the United States, Washington, D.C.: Pew Hispanic Center, available at: http:// www.pewhispanic.org/files/2013/05/2013-04_Demographic-Portrait-of- Mexicans-in-the-US.pdf, Consulted January 7, 2014.

Instituto Nacional de Estadística y Geografía (INEGI) (2014a), “Población ocupada por niveles de ingresos”, Banco de Información Económica (consulted August 7, 2014), available at: http:// www.inegi.org.mx/ sistemas/bie/

______ (2014b), “Indicadores oportunos de ocupación y empleo. Cifras preliminares durante julio de 2014”, Boletín de Prensa No. 380/14, August 22, 2014, Aguascalientes, Ags. (consulted August 30, 2014), available at: http://www.inegi.org.mx/inegi/contenidos/espanol/prensa/ comunicados/ocupbol.pdf

International Organization for Migration (2013), “World Migration Report 2013”, Geneva: International Organization for Migration, consulted October, 2014, available at: http://publications.iom.int/bookstore/free/ WMR2013_EN.pdf

Kochhar, Rakesh, in collaboration with C. Soledad Espinosa and Rebecca Hinze- Pifer (2010), After the Great Recession: Foreign Born Gain Jobs; Native Born Lose Jobs, Washington, D.C., Pew Hispanic Center (consulted March 16, 2011), available at: http://www.pewhispanic.org/2010/10/29/ after-the-great-recession-brforeign-born-gain-jobs-native-born-lose-jobs/

Levine, Elaine (2001), Los nuevos pobres de Estados Unidos: los hispanos, Mexico, Miguel Ángel Porrúa, Instituto de Investigaciones Económicas y Centro de Investigaciones sobre América del Norte, UNAM, pp. 203.

______ (2008), “Transnacionalismo e incorporación laboral de migrantes mexicanos en Estados Unidos y las perspectivas de ascenso socioeconómico para sus hijos”, in Elaine Levine (ed.), La migración y los latinos en Estados Unidos: visiones y conexiones, Mexico, Centro de Investigaciones sobre América del Norte, UNAM, pp. 253-276.

______ (2010), “Los hispanos/latinos en Estados Unidos”, in José Luis Sánchez Gavi and Adriana Sletza Ortega Ramírez (coords.), Migración México-Estados Unidos: Textos introductorios, Puebla, Benemérita Universidad Autónoma de Puebla and Lunarena Editorial, pp. 149-187.

______ (2012), “La polarización del mercado laboral inmigrante en los Estados Unidos: Inmigrantes mexicanos y asiáticos”, in Rocío Rosales Ortega, Ludger Brenner and Cristóbal Mendoza (coord.), Geografía económica y social: actores, instituciones y procesos globales, Mexico, Siglo XXI Editors and UAM-Iztapalapa, pp. 304-329.

Li Ng, Juan José and Telesforo Ramírez García (2014), Situación migración México Primer Semestre 2014, Mexico, BBVA Research (consulted July 8, 2014), available at: https://www.bbvaresearch.com/wpcontent/ uploads/2014/07/1407_SitMigracion _1S14.pdf

Massey, Douglas (2010), “Interview with Douglas Massey: Measuring the Effects of US Policy on Latin American Migration Patterns”, The American Academy of Political and Social Science, posted June 6, 2010 (consulted July 3, 2010), available at: http://aapss.org/news/2010/07/06/ douglas-massey-measuring-the-effects-of-us-policy-on-latin-american-migration-patterns

______ (2012), “America is Losing as Many Illegal Immigrants as it’s Gaining”, blog posted April 12, 2012 (consulted February 9, 2015), available at: http://blogs.reuters.com/great-debate/2012/04/12/america-is-losing-as-many-illegal-immigrants-as-its-gaining/

Migration Policy Institute (2009), MPI Data Hub, Migration Facts, Stats, and Maps (consulted July 6, 2013), available at: http://www.migrationinformation.org/datahub/

Motel, Seth (2012), Statistical Portrait of Hispanics in the United States, 2010, Washington, D.C., Pew Hispanic Center (consulted February 21, 2013), available at: http://www.pewhispanic.org/2012/02/21/statistical-portrait-of-hispanics-in-the-united-states-2010/

Passel, Jeffrey (2006), “Size and Characteristics of the Unauthorized Migrant Population in the U.S. Estimates Based on the March 2005 Current Population Survey” (consulted April 8, 2007), available at: http:// www.pewhispanic.org/files/reports/61.pdf

Passel, Jeffrey S. and D’Vera Cohn (2009), “Mexican Immigrants: How Many Come? How Many Leave?” Washington, D.C., Pew Hispanic Center (consulted September 18, 2011), available at: http://pewhispanic.org/files/reports/112.pdf

______ and D’Vera Cohn (2011), “Unauthorized Immigrant Population: National and State Trends, 2010”, Washinton, D.C., Pew Hispanic Center (consulted February 3, 2011), available at: http://pewhispanic. org/files/reports/133.pdf

______, D’Vera Cohn and Ana González-Barrera (2012), “Net Migration From Mexico Falls to Zero –And Perhaps Less”, Washington, D.C., Pew Hispanic Center (consulted April 23, 2012), available at: http://www. pewhispanic.org/files/2012/04/Mexican-migrants-report_final.pdf

Sum, Andrew, Neeta Fogg and Paul Harrington (2002), “Immigrant Workers and the Great American Job Machine: The Contributions of New Foreign Immigration to National and Regional Labor Force Growth in the 1990s”, report for the National Business Roundtable, Washington, D.C., USA.

Transactional Records Access Clearinghouse (TRAC) at Syracuse University (2014), “ICE Deportations: Gender, Age and Country of Citizenship” (consulted September 6, 2014), available at: http://trac.syr.edu/ immigration/reports/350/

US Bureau of Economic Analysis (2014), News Release, “Gross Domestic Product: Second Quarter 2014 (Second Estimate), August 28, 2014” (consulted September 3, 2014), available at: www.bea.gov/ newsrelases/national/gdp/2014/pdf/gdp2q14_2nd.pdf

US Census Bureau (2008), The Hispanic Population in the United States: 2008 Detailed Tables, Current Population Survey Annual Social and Economic Supplement 2008, Tables 1 and 2 (consulted February 7, 2010) available at: http://www.census.gov/population/hispanic/data/2008.html

______ (2014), Foreign-Born Tables Package, Annual data Tables, www.cen-sus.gov/population/foreign/data/cps.html

US Department of Labor (USDOL), Bureau of Labor Statistics (1991), Employment and Earnings, vol. 38, no. 1, Washington, D.C., US Government Printing Office.

______ (2007), Employment and Earnings, vol. 54, no. 1, Washington, D.C., US Government Printing Office.

______ (2008), Employment & Earnings, vol. 55, no. 1 (consulted July 9, 2008), available at: http://www.bls.gov/opub/ee/empearn 200801.pdf ______ (2009), Employment & Earnings, vol. 56, no. 1 (consulted March 10, 2009), available at: http://www.bls.gov/opub/ee/empearn200901.pdf

______ (2010), Employment & Earnings, vol. 57, no. 1 (consulted March 20, 2011), available at: http://www.bls.gov/opub/ee/empearn201001.pdf

______(2011), “Annul Average Household Data”, Employment & Earnings Online (consulted April 3, 2011), available at: http://www.bls. gov/opub/ee/2011/cps/annual.htm

______ (2012), “Annul Average Household Data”, Employment & Earnings Online (consulted April 12, 2012), available at: http://www.bls. gov/opub/ee/2012/cps/annual.htm

______ (2013), “Annul Average Household Data”, Employment & Earnings Online (consulted March 15, 2013), available at: http://www. bls.gov/opub/ee/2013/cps/annual.htm

______ (2014), “Annul Average Household Data”, Employment & Earnings Online (consulted April 3, 2014), available at: http://www.bls. gov/opub/ee/2014/cps/annual.htm

US Immigration and Customs Enforcement (ICE) (n.d.), “ICE Total Removals, Through August 25th, 2012” (consulted September 9, 2014), available at: https://www.ice.gov/doclib/about/offices/ero/pdf/ ero-removals1.pdf