NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Research Council (US) Panel on Hispanics in the United States; Tienda M, Mitchell F, editors. Hispanics and the Future of America. Washington (DC): National Academies Press (US); 2006.

Cover of Hispanics and the Future of America

Hispanics and the Future of America.

Show details

3The Demographic Foundations of the Latino Population

, , and .

The news that Hispanics have become the nation's largest minority was no demographic surprise. Its fruition had been predicted at least 30 years ago. This news event, though, was important because the appearance of Latinos on the American scene could no longer be denied—neither in the nation's vital economic or educational policies nor in politics. Nowhere. Once considered a sleeping giant, the Latino population has not only grown tremendously but also now constitutes a significant presence throughout most of the United States. Once confined to a small number of states, the Latino population has migrated to new regions, including much of the South, moved into new sectors of the economy, and become an important voting bloc in many states. Its impact is heightened by the fact that it is considerably younger than an aging non-Latino America, making its potential impact on America's future all the greater.

This chapter reports on the factors that account for this growth. Overall, it describes how relatively high rates of immigration and fertility have shaped the growth and the creation of an especially youthful age structure among the Latino population. In particular, it examines how changing immigration policies, social networks, and other factors have led to immigration from Latin America and then how a changing labor market as well as immigration policies have affected migration patterns in the United States and prompted the regional dispersion of Latinos. These demographic foundations are fundamental for understanding nearly every aspect of Latino well-being covered in this book, including their spatial distribution and family structure, their position in the educational system and the labor market, and their access to health care and the political system. A notable example of the importance of this population was its role in the recent presidential election: the Hispanic vote may have influenced the outcome (Cobble and Velaquez, 2004). Given the demographic destiny of the Latino population, that influence is likely to grow with its dispersion into new states and as immigrants become citizens and their children reach voting age.

At the same time, the Latino population has become increasingly diverse by national origin. Mexicans continue to constitute the large majority of Latinos in the United States, driving the demographic behavior of Latinos in general as well as mainstream American attitudes toward the Hispanic group. However, many other groups have also become part of the new immigration from Latin America, as the previous chapter has shown. While Mexicans, Puerto Ricans, and Cubans constituted almost all Latinos in the United States just 30 years ago, Dominicans, Central Americans, and South Americans have doubled or tripled their numbers in the past two decades. This chapter also shows that Latino national groups vary greatly in their age structure and extent of regional dispersion. Specifically, Cubans have an old age structure and have become increasingly concentrated, two patterns that are unlike the rest of the Latino population, whereas Mexicans and Central Americans are especially young and have migrated throughout the United States.


Demographic growth or decline is a result of births and deaths, also known as natural growth and net migration, which is the balance of immigration and emigration. The growth of the Latino population is mostly the result of two of these components, births or fertility and immigration. While many assume that the growth is due almost entirely to immigration, relatively high rates of Latino fertility now constitute roughly half of all population growth. That fertility is comprised largely of births to immigrants, but a sizeable component can also be attributed to the U.S.-born. Table 3-1 breaks down Latino population growth in the 1970s, 1980s, and 1990s into that resulting from either immigration or fertility to both immigrants and the U.S.-born.

TABLE 3-1. Percentage Components of Latino Population Growth by Decade, 1970–2000 .


Percentage Components of Latino Population Growth by Decade, 1970–2000 .

The last row of the table shows that, in the 1990s, nearly half (48 percent) of Latino population growth was due to immigration, 28 percent can be attributed to fertility among immigrants, and the remaining 24 percent resulted from the fertility of U.S.-born Latinos. The same table also shows that, in relative terms, the share of Latino growth due directly to immigration in 1990–2000 declined compared with the decade before, when immigration accounted for fully 56 percent of growth. Even though immigration increased in the 1990s in absolute numbers, its share of total Latino population growth declined, although fertility to immigrants continued at about the same rate. In both the 1980s and 1990s, at least three-quarters of Latino population growth has been due to immigrants, either by their own migration or through their childbearing.

Fertility went from accounting for 44 percent of all growth in the 1980s to representing 52 percent of growth in the 1990s. The growth from fertility to U.S.-born Latino parents represents the greatest growth share, increasing from 17 percent in the 1980s to 24 percent in the 1990s. This signals a reemergence of the so-called third generation, which begins to echo the presence of a large third generation, as in the period prior to 1980. Births to such parents had represented fully 39 percent of Latino growth in the 1970s but dropped steeply to less than half (17 percent) in the 1980s.


The growth and increasing national diversity of immigration from Latin America was shown in Chapter 2, but in this chapter we seek to describe its sources. The volume and socioeconomic characteristics of the immigrant Latino population are largely related to economic and political factors and the social networks that have since perpetuated immigration. In the cases of Mexico and Puerto Rico, the current migration flows were initiated with intense labor recruitment, but once the process was set in motion, economic factors and social networks continued to fuel further immigration. In the Cuban case, the causes were, and continue to be, fundamentally political. For the Dominican Republic, political causes, including dictatorship and military intervention, played an important role at first but social and economic causes perpetuated the process. A similar process occurred in the cases of Guatemala, El Salvador, and Nicaragua, where immigration was political as well as economic. Finally, in the case of South America, socioeconomic factors have been predominant in immigration to the United States, although political violence in Colombia and Peru has been an important push factor for migration from those countries.

For more than a century, Mexican immigration to the United States was distinguished by two components: a migration that settled and another between Mexico and the United States that was circular, largely facilitated by the porosity of the border. Similar phenomena occurred in the case of the Puerto Ricans, who could easily go back and forth between Puerto Rico and the mainland, facilitated by low airfares and the fact that they are U.S. citizens. According to Massey, Durand, and Malone's (2002) theory of cumulative causation, each act of migration, especially through the large volume of return migration, alters the social context in which subsequent migration decisions are made, thus increasing the chances of future migration. Immigration from Mexico and Puerto Rico has advanced considerably through such social networks, although Puerto Rican immigration has slowed as the networks have already incorporated a large part of the eligible population in the immigration process. The other national groups are arguably at earlier stages.

Demographic, political, and economic factors in Latin America also help to account for the large increases in immigration from that region in the 1980s and 1990s. In the 1970s, many countries began birth control programs, as a demographic transition with continuing high fertility rates and decreasing mortality rates led to sharp population growth. The effects of Latin America's baby boom began to be felt in the labor force, as new entrants to the labor force had increasing difficulty finding work and thus opted for migration abroad. At the same time, the 1970s and 1980s were turbulent political times, especially in Central America, where civil wars and other types of armed conflict generated intense emigration (Hamilton and Stoltz Chinchilla, 2001; Menjivar, 2000).

Finally, it is important to recognize the continuing attraction of the United States, as Latin American countries struggle economically. Practically all Latin American countries have suffered recurring economic crises, currency devaluation, and runaway inflation that have left millions of people in poverty, many of them from sectors of the middle class who seek to maintain a decent quality of life through migration. The effect of neoliberal economic policies has been mixed across countries, with Chile being a successful case while Peru, Argentina, and Ecuador have had negative experiences thus far (Huber and Solt, 2004; Walton, 2004).

Effects of U.S. Immigration Law on Immigration from Latin America

If economic, political, and social factors provoked and perpetuated immigration from our southern neighbors, the history of immigration law may be most responsible for the diversity of Latino immigration in terms of legal status, class selectivity, and destination. With the exception of Puerto Rico, migratory flows from Latin American countries to the United States have been shaped, to varying degrees, by the opportunities, limitations, and exceptions that different laws and regulations concerning migration have offered since the turn of the 20th century. Among the national-level immigration laws that have been ratified in the United States, three deserve special attention because of their direct effects upon migration from Latin America:

  1. The 1965 reform known as the Immigration and Nationality Act, and its numerous subsequent corrections;
  2. The 1986 Immigration Reform and Control Act (IRCA); and
  3. The 1996 Illegal Immigration Reform and Immigrant Responsibility Act (IIRAIRA).

In addition, there has been special legislation related to the specific cases of Cubans and Central Americans.

The Immigration Act of 1965, also known as the Hart-Celler Act, established an annual limitation of 170,000 visas from all Eastern Hemisphere countries with no more than 20,000 per country. The civil rights revolution in the 1960s sought to mitigate discrimination based on race and ethnicity, and thus the 1965 law sought to revoke the previous immigration policy, which favored European immigration and severely impeded Asian immigration. By replacing the old quotas with numerical country ceilings that were uniform across all Eastern Hemisphere countries, the change effectively eliminated the restrictions on immigration from Asia (Jasso and Rosenzweig, 1990). It also stressed family reunification as grounds for admission, along with a very inclusive definition of family relationships. Thus the 1965 act's legacy is to have opened the door to large-scale immigration from the Eastern Hemisphere. This law also ended the Bracero Program and created a general visa policy for the Western Hemisphere with no numerical limits.

In 1968, a limit of 120,000 visas for the Western Hemisphere was set, although this amendment does not determine fixed, per-nation quotas. For 10 years, Mexicans secured roughly half of all the Western Hemisphere visas under the quota. However, with a 1976 amendment, a quota of 20,000 visas per country was established and continues to the present (Jasso and Rosenzweig, 1990; Reimers, 1992, p. 87).

Consequently, this law greatly reduces the number of legally authorized immigrants who can come from Mexico and effectively increases the number of immigrants that can come from the rest of Latin America. Given historically high levels of legal Mexican immigration for the past century or more, a long common border with the United States and persistently great socioeconomic inequality between the two countries, the tightening of immigration restrictions for Mexico has generated greater levels of undocumented immigration.

This system of hemispheric limits soon was in crisis because of the almost 360,000 Cuban refugees who were allowed to enter the country between 1965 and 1978, a fact that came to modify the established ceiling for the Western Hemisphere. A case involving the exceptional number of visas given to Cubans was taken to court. The judge in that case, known as Silva versus Levi, ruled that the rights of 145,000 people, primarily Mexicans, had been violated. As a result, visas that came to be known as Silva Cards (Cartas Silva) were issued to the victors in the lawsuit (Reimers, 1992) and the immigration law was eventually reformed in 1980 to separate refugees from the per-country quota system.

In the next stage, the 1986 IRCA greatly affected the legal status of some 2.7 million undocumented migrants. This law outlined two amnesty programs, which included a general program named the Legally Authorized Worker (LAW) program for undocumented immigrants who had resided five years or more in the United States, and the Special Agricultural Workers (SAW) program for those who had worked in agriculture during the past six months. Fully 70 percent of those taking advantage of the LAW program and 81 percent of the SAW program were Mexicans. To some degree, Central Americans had a more difficult time qualifying for these programs, since many of these new immigrants did not meet the residency requirements (Donato, Durand, and Massey, 1992; Durand, 1998; Massey et al., 2002).

The major impact of this law was to improve the legal status of Latino immigrants in general (and Mexicans in particular), by regularizing the situation of most existing undocumented migrants. Legalization allowed Latinos to become more geographically dispersed throughout U.S. territory, because it permitted unrestricted movement. Undocumented workers tend not to move very far from their workplaces, because when traveling they become conspicuous and risk greater exposure to being detained by the Immigration and Naturalization Service (INS). However, once they were able to secure legal papers through IRCA, more than 3 million individuals were able to travel with greater ease to look for work or new opportunities in other areas. The economic crisis in California that resulted from the end of the Cold War and the decline of the local aerospace industry led to further immigration away from that state. At the same time, many immigrants who had taken advantage of the IRCA programs found better wages and opportunities in other states. The 1990s economic recovery in the rest of the United States increased the demand for workers (Donato et al., 1992).

The IIRAIRA of 1996 constituted a serious blow to the community of Latin American origin because it restricted a wide range of support programs and services to which the migrant population had previously had access, regardless of legal status. Undocumented workers consequently suffered, substantial hurdles were placed to prevent the entrance of refugees, and resident immigrants with work permits who were not citizens were penalized. In many ways, this federal law contained many of the same elements as California's Proposition 187, which voters in that state passed in 1994, although it avoided California's ban on K–12 education and emergency services for the undocumented, which had been ruled unconstitutional by the U.S. Supreme Court (Hood and Morris, 2000; Weintraub, 1997).

Nevertheless, as so often happens, laws bring unexpected consequences. In this case, the new restrictions on the immigrant community actually fostered the empowerment of that very group, as the number of applications for naturalization soon rose substantially. The year 1996 was a landmark in this regard, as the number of Mexican migrants who applied for naturalization tripled compared with the year before—an interesting development given this particular group's traditional reluctance to change its nationality, although it was expected from the amnesty program under IRCA. Similar patterns emerged among immigrants from Cuba, the Dominican Republic, El Salvador, Guatemala, Colombia, Ecuador, and Peru. Figure 3-1 shows that the number of naturalizations for immigrants from Latin America increased from 65,000 in 1991 to 529,000 in 1996.

FIGURE 3-1. Number of Hispanics naturalizing each year from 1991 to 2002.


Number of Hispanics naturalizing each year from 1991 to 2002. SOURCE: U.S. Immigration and Naturalization Service (2002).

Three factors were particularly responsible for the change in the number of naturalizations: (1) the 1986 IRCA legislation allowed residents to naturalize, (2) the 1996 IIRAIRA law placed restrictions on social benefits for illegal immigrants, and (3) the constitutional reforms in several Latin American nations, such as Mexico, Argentina, Costa Rica, and Peru, allowed migrants to acquire a new nationality without losing their original one. Of course, naturalization data are affected by bureaucratic rhythms, particularly the time required to process applications, but the large increase between the first and second half of the 1990s is indisputable. As in much of the data on Hispanics, the Mexican case drives these results. There were 22,000 naturalizations from that country in 1991, 254,000 in 1996, and an average of 150,000 in subsequent years.

In synthesis, there were no restrictions on immigration for Latin Americans prior to 1965. Hemispheric restrictions began in 1968 and by country in the hemisphere in 1976. Regardless of the quota system, this legislation did not impede the growth of the Latino population through immigration, although it clearly fueled increased diversification of national origins. Also, the 1986 IRCA legislation had transformed earlier undocumented Hispanics into permanent immigrants and encouraged greater geographical mobility within the country (Massey et al., 2002). The 1996 law stirred the Latino community to action, as many individuals decided to apply for naturalization in order to avoid losing a number of rights they had enjoyed up to that time. These changes consequently have implications for the nature of Latino political participation in the United States, as Chapter 11 discusses.

These different migration laws not only affected the Latino immigrant population directly, but also brought them into direct contact with governmental institutions and the U.S. legal system. Through personal procedures and collective social struggles—especially of a legal nature—Latinos began to interact with a variety of U.S. institutions. In short, they became more politically integrated as they learned how to operate in North American society. This dynamic of institutional interaction can be better appreciated through the analysis of some specific cases (Georges, 1990; Hamilton and Stoltz Chinchilla, 2001; Pessar, 1995).

U.S. immigration laws have treated specific Latin American countries distinctly, which may help explain why particular countries have sent more immigrants than others. For example, the only country to have ever experienced a guest-worker program on a large scale is Mexico. Without question, migration from Mexico has been subject to more exceptions to U.S. immigration law than that of any other country, beginning in 1917, when Mexican immigrants were exempted from the clause that required migrants to pass a literacy test (Cardoso, 1980; Jasso and Rosensweig, 1990). From 1942 to 1964, the Bracero Program allowed about 5 million Mexican agricultural workers to enter the United States to work legally on a seasonal basis (Calavita, 1992). In 1965, these former guest-workers (braceros) were offered the opportunity to legalize their migratory status if their employers or relatives were willing to support their applications, as there were no fixed limits or country quotas for Western Hemisphere nations.

It was not until 1978 that Mexico had to adapt to the quota system that the United States established at the world level (20,000 visas). This controlled migratory flow primarily involved people who tended to settle in particular cities, especially Los Angeles (Ortiz, 1996). Mexicans used the family reunification measures to bring their relatives into the country. According to Jasso and Rozenweig (1990), in 1985, 22.7 percent of spouses sponsored by native-born U.S. citizens were from Mexico, whereas the next highest group, Fillipinos, stood at 4.9 percent. Later still, in 1986, Mexican immigrants were also the main beneficiaries of the IRCA reforms, through which perhaps 2.3 million people succeeded in regularizing their migratory status. Finally, in the 1990s, almost 80,000 H2A and H2B visas were issued to Mexicans to enter the country legally as temporary agricultural and service workers, despite the fact that those particular visas had traditionally been granted to migrant workers from the Caribbean (Portes and Rumbaut, 1990; Smith-Nonini, 2002) (see Box 3-1).

Box Icon

BOX 3-1

H2A and H2B Visas. The H2A temporary agricultural visa is a nonimmigrant visa that allows foreign nationals to enter the United States to perform agricultural labor or services of a temporary or seasonal nature. Limits are not fixed and vary annually. (more...)

Finally, the U.S. government initiated the Diversity Visa Program, also known as the Green Card Lottery, in 1990. It offers 55,000 visas each year to citizens of “under-represented countries” in order to facilitate immigration from countries other than those that send large numbers of migrants. This is the easiest and most cost-effective way to secure an immigration visa, because persons chosen under this system have the right to migrate with their families to the United States. In some cases, they are given prepaid airfare. In Latin America, citizens of Colombia, the Dominican Republic, El Salvador, and Mexico are considered ineligible.

Cuba became a special case beginning in 1959, both with respect to the entry of refugees and to the systems of quotas, lotteries, deportations, and regularization. What is most notable in the Cuban case is the series of advances and retreats that have characterized the migratory policies of the U.S. and Cuban governments, which open and close their doors as a function of changing political conjunctures. The Cuban government allowed several thousand emigrants to leave the island freely from the harbor at Camarioca in 1965, and in 1980 some 124,000 more were permitted to leave the country from the port of Mariel and were eventually received as refugees by the United States (Portes and Stepick, 1993). Beginning in the Johnson administration, air travel from Cuba was allowed, making it possible for 360,000 more Cubans to enter the country (Reimers, 1992, p. 124). Between 1961 and 1988, almost half a million Cubans (486,426) became permanent residents (Jasso and Rosenzweig, 1990). In the Cuban case, there has even been legislation (the Law of Adjustment) related to the so-called raft people, who are granted refuge only if they succeed in reaching American soil, a right that is denied to those who are captured or rescued while still at sea.

Dominicans also received preferential treatment after the fall of Trujillo, who governed as president from 1930 to 1961 and exercised an almost exclusively personal authority over the issuing of passports (Grasmuck and Pessar, 1991). During the 1965 U.S. invasion, President Johnson liberalized the authorization of visas as a kind of escape valve designed to diffuse social conflicts, so from 1965 to 1966 the number of resident visas increased by 74 percent. Those who benefited most were young men who had supported the opposition movement (Georges, 1990). Later, the economic reforms instituted by the Balaguer regime spurred an intense flow of migrants to the United States. Many of these Dominican migrants entered the country with legitimate tourist visas and then simply stayed on indefinitely, while others opted for an indirect emigration route that took them first to Puerto Rico, where they crossed the dangerous Strait of La Mona (Duany, 1990). During their stay in Puerto Rico, many Dominicans worked in agriculture or the informal economy, although others considered their time on the island as simply a stopover on their journey to the United States (Duany, Hernandez Angueira, and Rey, 1995).

Central American countries, such as Nicaragua and, to a lesser extent, El Salvador and Guatemala, have also benefited from special regularization programs. In the 1980s, President Reagan's support for the contras (antisandinistas) was manifested, among many other ways, by a very liberal policy for issuing tourist visas to Nicaraguans, many of whom later became illegal immigrants but were finally granted refugee status. After several years of legal struggles, Nicaraguans were granted refugee status in 1997, thereby earning the right to reside in the United States on a permanent basis, thanks to legislation called the Nicaraguan Adjustment and Central American Relief Act (NACARA). This particular act was also utilized by some Salvadoran, Guatemalan, and Cuban migrants (Menjivar, 2000).

Nonetheless, there was a clear reluctance to recognize most Salvadorans and Guatemalans as refugees even if they were fleeing civil wars, so they were classified as economic immigrants. Many of them were deported in the 1980s, while a few received asylum and others launched lawsuits with the support of a variety of nongovernmental and religious organizations (Hamilton and Stoltz Chinchilla, 2001). Finally, in 1990, a form of temporary protection was granted (temporary protected status, or TPS1), which gave those migrants permission to work, although it did not grant them permanent residence status, as had been the case with the Nicaraguans. This measure was renewed several times until 1997, when the courts established a legal status designed to protect all migrants who found themselves in similar situations. In this way, many Salvadorans and Guatemalans succeeded in regularizing their migratory status (Menjivar, 2000). If IRCA was for Mexicans the means of legal recourse that allowed many undocumented migrants in their communities to legalize their situation, for Central Americans the legislative processes provided by NACARA and TPS were fundamental to their establishing permanent residence on American soil.

Finally, migrants from South American countries depend mostly on the established quota system that allows them to enter the United States as tourists, although once there, they often take advantage of the family reunification provisions. Few South Americans have petitioned for refugee status, and there have been no cases of special legislation or programs related to these nations. Nonetheless, the number of migrants that decide to remain in the United States after their visas expire is increasing, adding to the ranks of undocumented migrants.

Undocumented Immigration

In effect, the limits on immigration from Latin America, which began in 1978, along with the growing demand for low-wage labor have led to the growth of an undocumented population. The Latino population in the United States today includes a high proportion of undocumented migrants, a situation that made regularization programs so significant in the past. In practice, there are now two main modalities of undocumented migration. The first is the preferred method of Mexican and Central American migrants, who enter the United States surreptitiously by crossing its southern border. The second is that used by individuals who first obtain a tourist visa to enter the country and then stay beyond the document's expiration date. Estimates for 2002 indicate that there are 9.3 million illegal migrants in the United States (Passel and Fix, 2004).

Careful demographic studies on the size of the undocumented population are relatively recent. At first, the INS estimated that there were between 8 and 12 millon undocumented in the 1970s but with no empirical evidence. This led to more careful examination by demographers. Systematic demographic studies began to be carried out based on the 1980 census, especially with respect to calculating the size of the Mexican-origin population. Demographers began to reach a consensus that, in 1980, the Mexican undocumented population was between 1.7 and 2.3 million and that other nationalities constituted roughly another 1.5 million (see Bean, Telles, and Lindsay Lowell, 1987, for a review of sources). This estimate appears to have been accurate, since IRCA legalized 2.3 millon Mexicans in 1986 and, on the basis of estimates by Massey et al. (2002), 600,000 undocumented immigrants remained. Thus, on the eve of IRCA in 1986, there were 2.9 million undocumented Mexicans. Soon after, the undocumented population began to grow again and now included a large Central American contingent. By 1997 with NACARA and with TPS in 2000, close to 3 million undocumented Central Americans in the United States had become legal residents (Hamilton and Stoltz-Chinchilla, 2001; Menjivar, 2000), so from 1986 to 2000, the undocumented population varied widely. By 2002, according to estimates by Passel and Fix (2004), the total undocumented population had grown to 9.3 million, the majority of whom are Hispanic. Mexicans constitute 57 percent of all undocumented, and other Latin Americans, primarily Central Americans, constitute 23 percent (Passel and Fix, 2004). Thus, four-fifths of all undocumented migration to the United States originates in Latin America.

Several factors explain the presence of many undocumented Hispanic immigrants. Historically, efforts were made to create a model of temporary migration from Mexico different from the flows that originated in Europe. On the other hand, the absence of specific legislation related to Mexico once the Bracero Program was eliminated clearly fostered undocumented immigration, which was tolerated by the U.S. government and used as a mechanism for regulating secondary labor markets where wages were low and working conditions poor.

Mexicans came to depend on a complex network of social relations to reduce the costs and risks associated with crossing the border and live as illegals. In the absence of official contracts, as during the Bracero Program period, these Mexicans sought the support of relatives and friends for finding work and a place to live. At the same time, employers, who could no longer seek workers from an official bracero system, began to utilize migrant networks to hire and train workers. As a result, the social networks matured and became increasingly complex. They served not only the migrants themselves but also employers' demand for low-cost and reliable labor (Durand, 1994; Durand and Massey, 2003; Massey, Alarcón, Durand, and González, 1987; Passel and Fix, 2004). Growing Mexican immigration, fostered through the networks of formerly unregulated migrants, simply overwhelmed the relatively small visa quotas suddenly imposed on Mexico in 1978.


In addition to immigration from Latin America, the extent of Latino population growth in the United States depends on fertility or the childbearing rates of Hispanic women, which tend to be higher than for non-Hispanic whites and blacks. Also, Hispanic women vary widely by nationality in the number of children they bear. Table 3-2 shows total fertility rates and the mean number of children ever born to women ages 35 to 44 in 2000. The total fertility rate is an age-standardized measure of recent fertility among women ages 15 to 44 and has become the standard childbearing measure. All Latino groups shown have higher fertility rates than non-Hispanic white women in the United States. Mexican women clearly had the highest fertility, registering a rate of 3.3 in 2000 and at the other extreme, Cuban women had only 1.9 children. This compares with 1.9 for non-Hispanic whites and 2.2 for blacks.

TABLE 3-2. Total Fertility Rates by Ethnicity and Children Ever Born Rates to Women Ages 35 to 44 by Ethnicity and Generation Since Immigration, Women 35 to 44, 2000 .


Total Fertility Rates by Ethnicity and Children Ever Born Rates to Women Ages 35 to 44 by Ethnicity and Generation Since Immigration, Women 35 to 44, 2000 .

We also computed the mean number of children ever born to women ages 35 to 44. This provides a measure of nearly complete fertility, although it misses the fertility experiences of younger women. However, it permits us to disaggregate fertility for various immigrant generations, since data are not available to break down the total fertility rate in this way. By limiting the age cohort to 35 to 44 rather than including all women of childbearing age, we avoid most, though not all, of the distortion caused by different age structures, especially problematic for the second generation, which tends to be very young.

Immigrant Mexican women in this age group have had 2.7 children compared with 2.1 for the children of immigrants and 2.3 for grandchildren and later generations. Thus, there is limited support for convergence toward the fertility of black and white native-born groups, although the second generation has even lower fertility than the third, which continues to have considerably higher fertility than black and white natives.2 Interestingly, the fertility rates of white, black, and Asian immigrants are lower than those of the well-established and much more common third-generation black and white population. Non-Mexican Latino populations are generally too small to capture a reliable fertility index, except for the case of Puerto Ricans. The so-called Puerto Rican second generation, those who were born on the U.S. mainland of island-born parents, have only slightly lower fertility than those born in Puerto Rico.

Fertility rates for immigrants to the United States are often higher than those for their compatriots who stayed behind. This is especially true in the case of Mexicans, for whom an immigrant total fertility rate of 3.3 in 2000 was well above the 2.6 recorded for Mexico in the same year. The reasons for these differences are unclear, but perhaps Mexican immigration is selective of families and persons from rural areas where fertility is higher. A Cuban immigrant fertility rate of 1.9 is above that of 1.6 for Cuba; this may reflect the social status of those who immigrated as well as the profound changes that occurred on the island since the time when most Cubans immigrated. Puerto Rican immigrant fertility of 2.2 is also above that for Puerto Rico, which recorded a total fertility rate of 1.9. Thus, although we might expect lower fertility among immigrants who migrate to the United States because they are coming to a more urban and modern destination and because the migration process itself may interrupt the likelihood of having children, their fertility tends to be higher.3

Age and Sex Distributions

We have shown that the Latino population is one of the fastest growing segments of the U.S. population, and that growth is fueled by a combination of fertility and immigration. However, the extent of fertility and the nature of immigration, specifically the age and sex of immigrants, have effects not only on the growth rate of the population but also on the population's age and sex structure. That is, it determines, along with mortality, how a population is distributed by age and the sex balance at each age. This age–sex distribution, represented in this section by population pyramids, reflects the importance of Latinos in different stages of the life cycle and thus the extent of their participation in such areas as education, family formation, the labor market, and their use of health care services.

The particular combination of relatively high fertility and the migration of persons at young working ages leads to a Latino population that is particularly young, as Figure 3-2 shows, especially when compared with the age structure of the non-Hispanic white and black populations (Figures 3-3 and 3-4). The population pyramids represent the age and sex distribution of these groups, and they are further broken down by generation since immigration. The Latino population pyramid's shape reveals a small middle-aged and senior population; the population under age 45 comprises the vast majority of the population. Almost 40 percent of the Latino population is under age 20, and 65 percent is under age 35. Foreign-born Latinos comprise 43 percent of the population, compared with 31 percent who are second generation and 26 percent in the third generation. However, the age distributions are quite diverse across generations. The foreign-born population, represented by the lightest shade, is diamond shaped around a middle band, comprised of 30- to 34-year-olds. This form characterizes a labor migrant population that brings few children with them. In terms of sex, the immigrant population is only slightly more male than female, making it more gender-balanced than earlier Latino immigrant waves.

FIGURE 3-2. Age–sex composition of the Latino population of the United States, 1998–2002.


Age–sex composition of the Latino population of the United States, 1998–2002. SOURCE: Data from Current Population Surveys, 1998–2002.

FIGURE 3-3. Age–sex distribution of the white population of the United States, 1998–2000.


Age–sex distribution of the white population of the United States, 1998–2000. SOURCE: Data from Current Population Surveys, 1998–2000.

FIGURE 3-4. Age–sex distribution of the black population of the United States, 1998–2000.


Age–sex distribution of the black population of the United States, 1998–2000. SOURCE: Data from Current Population Surveys, 1998–2000.

By the second and third generations, there is little difference between the percentage distribution of males and females by age, as expected from a population that is no longer immigrant. For the third generation, the age–sex distribution is only slightly younger than the distribution for the entire U.S.-born population. In terms of age and sex, there is growing convergence with the rest of the U.S. population.

The youthfulness of the Latino population is especially apparent in comparison to the age pyramids for blacks and whites who are not Hispanic. Their pyramids show the large bulge of a baby boom generation born for nearly 20 years after World War II. The bulge in the 40 to 54 age groups in the year 2000 for whites and blacks implies that a large group of U.S.-born will soon reach retirement age. Although the baby boom had had a significant echo effect, reflecting that generation's births, their fertility greatly declined compared with that of their parents. Latinos have thus largely supplanted non-Hispanic whites and blacks as new labor force entrants and increasingly constitute large numbers of the school-age population.

Figures 3-5, 3-6, and 3-7 show analogous population pyramids for three of the largest Latino national groups: Mexicans, Cubans, and Dominicans. These three groups illustrate the internal variation in the structure of the Latino population. The age–sex distribution of the Mexican population in the United States, shown in Figure 3-5, looks very similar to that of the Latino population. This is not surprising considering that Mexicans make up 63 percent of the Latino population. The Mexican population is also a young population: 42 percent are under age 20 and almost 70 percent are under age 35. First-generation immigrants constitute the largest percentage of the population over age 20, while second-generation immigrants dominate the population under age 20. Again, the male and female populations are relatively similar in terms of both age distribution and nativity, although there are slightly more first-generation males (41 percent) than females (36 percent). For the Mexican-origin population, the second and especially the third generation are significant in all age groups, reflecting an especially long-standing immigration.

FIGURE 3-5. Age–sex composition of the Mexican-origin population by nativity, 1998–2002.


Age–sex composition of the Mexican-origin population by nativity, 1998–2002. SOURCE: Data from Current Population Surveys, 1998–2002.

FIGURE 3-6. Age–sex composition of the Cuban-origin population by nativity, 1998–2002.


Age–sex composition of the Cuban-origin population by nativity, 1998–2002. SOURCE: Data from Current Population Surveys, 1998–2002.

FIGURE 3-7. Age–sex composition of the Dominican-origin population by nativity, 1998–2002.


Age–sex composition of the Dominican-origin population by nativity, 1998–2002. SOURCE: Data from Current Population Surveys, 1998–2002.

Figure 3-6 demonstrates that the Cuban population in the United States departs markedly from the age–sex distributions of Mexicans and the general Latino population. While the shapes of the pyramids for all Latinos and for Mexicans specifically resemble triangles, reflecting young and gender-balanced populations, the Cuban graph shows a much older population, in which only 22 percent of Cubans are under age 20 and 40 percent are under age 35. This is consistent with their fertility levels, which are more similar to those of the mostly U.S. native black and white populations, which have similar distributions. These numbers provide a stark contrast to those for Mexicans, for whom the majority of the population is less than 35 years old.

Also, this graph shows that the vast majority of Cuban origin persons over the age of 40 are immigrants, apparently the result of refugee immigration flows concentrated in the 1960s. Only 6 percent of all Cubans in the United States are third generation, and these tend to be in the youngest age categories. This is in comparison to 27 percent who are second generation and 67 percent who are first generation. The sex distribution of the population is relatively equal, although there are slightly more first-generation males than females under age 20. Finally, the alternating periods of migratory flux and reflux, plus the age-related selectivity imposed by the Cuban government, which makes it more difficult for young people to migrate under the pretext of their obligation to perform military service, have left their mark on the peculiar distribution of ages and generations that make up the Cuban-origin community (Grenier and Pérez, 2003).

The other Latino national groups are nearly all part of the wave of new immigrants that came after 1970. Thus, the majority of adults of the new Latino subgroups (Dominicans, Central Americans, South Americans) are immigrants. The pyramids for Central Americans and South Americans are shown in Appendix Figures A3-1 and A3-2 at the end of this chapter. We show only a population pyramid for Dominicans in Figure 3-7.4 The age distribution of Dominicans is particularly young, even more than other Latino groups, and, like the new immigration from Central and South America, the large majority of those over age 30 are immigrants. However, Dominicans are different from the other major Latino groups because of the preponderance of females, especially in the first generation: 64 percent of Dominican women are first-generation immigrants, compared with 54 percent of men. Although not shown, the age distribution for Central Americans is especially concentrated in the 25- to 45-year-old categories, which may reflect a sudden surge of immigration in the late 1970s and early 1980s, when civil wars broke out in Guatemala, El Salvador, and Nicaragua. Finally, the South American population, also not shown, is relatively old, apparently reflecting a largely middle-class immigration with low fertility.

FIGURE A3-1. Age–sex composition of persons of Central American origin by nativity, 1998–2002.


Age–sex composition of persons of Central American origin by nativity, 1998–2002. SOURCE: Data from Current Population Surveys, 1998–2002.

FIGURE A3-2. Age–sex composition of persons of South American origin by nativity, 1998–2000.


Age–sex composition of persons of South American origin by nativity, 1998–2000. SOURCE Data from Current Population Surveys, 1998–2002.

With the exception of the Cubans, the Latino second generation is largely young and the offspring of the recently arrived immigrants. The white and black populations, in contrast, contain very few immigrants or children of immigrants and most of the non-Hispanic black and white immigrants are in the middle age ranges. Immigrants among the Latino population dominate the 20- to 39-year-old age bars, mostly with first-generation persons, while their second-generation U.S.-born children fill the ranks of the youngest age brackets. Thus, with continuing immigration in the next 20 years, the second generation will continue to dominate the under age 20 categories, while immigrants will compete for the 20- to 40-year-old brackets with the grown children of immigrants. Prior to 20 years ago, there were far fewer immigrants available to become mothers. Now, the rapidly growing population of immigrant mothers has contributed to a large and young second generation. Seen from another perspective, between ages 0 and 19, the Latino population is dominated by the children of immigrants. This baby boom caused by the growth in immigration portends a significant change in the ethnic composition of the U.S.-born adult population. A large Hispanic component born and socialized in the United States is entering the labor force and in certain parts of the country will dominate the blue-collar labor force in at least the next 40 years. The Latino proportion of the U.S. electorate is also certain to grow, which could have a large impact on U.S. leadership at the beginning of the 21st century. Overall, this infusion of young persons into the U.S. population will allow it to keep a relatively young age profile, in contrast to the aging populations of most industrialized countries.

The age structures of the Latino populations also have implications for education, the labor force, health care, and the future of the social security system, as subsequent chapters show. With time, the age structures demonstrate the progressive Hispanification of increasingly older age groups from children to adults to seniors, as Latinos themselves age. Also, massive industrial restructuring since the 1970s has removed and downgraded many formerly high-paying working-class jobs, and the new economy has taken advantage of the new infusion of low-cost labor provided by immigration, particularly from Latin America. The dispersion of Latinos into areas of new industrial growth may also reflect and fuel the changing spatial structure of industry. At the same time that young Latino immigrants are overrepresented and growing as a proportion of the young working population, their immigrant status means they are less likely to be qualified to vote. Thus, immigrants are contributing substantially to the well-being of the native population in several ways, but they are largely unable to participate in the electoral process, because of foreign birth or young age. For the children of these immigrants, who are automatically citizens, their relatively young age will prevent their electoral participation in the near future. The impact that Latinos will have on these institutions also depends on the extent to which they populate particular regions of the United States. Traditionally, they have been regionally concentrated, so their effects are often locally specific, but that is changing.

Geographical Dispersion

Traditionally, the Hispanic population in the United States has been concentrated near the southern border with Mexico, the Northeast, and Florida. Such concentration is typical of immigrant groups who build communities in a few areas before branching out. Also, in the case of Mexicans, the U.S. Southwest, their territorial area of concentration, was formerly Mexican territory that was annexed in the 19th century. Table 3-3 shows the leading three states for several Latino national groups. California and Texas served as both ports of entry and areas of absorption for immigrants arriving from Mexico (and later Salvadorans and Guatemalans, the two largest Central American groups—Hamilton and Stoltz Chinchilla, 2001; Massey et al., 1987, Menjivar, 2000). On the East Coast, New York, New Jersey, and Florida are the preferred destinations of Caribbean and South American immigrants (Pessar, 1995; Portes and Stepick, 1993). As the traditional areas of settlement and because of the force of social networks that continues to draw immigrants to these states, 7 out of every 10 Hispanics reside in these five states. Other important immigrant-receiving states are Illinois, where Mexicans began to arrive in the 1920s and Puerto Ricans in the 1940s, and Arizona and New Mexico, which were, like Texas and California, formerly part of Mexican territory.

TABLE 3-3. Top Three Places of Residence at the 2000 Census of the Foreign-Born Latino Population by Place of Birth .


Top Three Places of Residence at the 2000 Census of the Foreign-Born Latino Population by Place of Birth .

Despite the concentration of the Latino population in only a few states, we should also emphasize two more recent processes that are expanding with great intensity: (1) the diversification of national origin in urban areas once considered to be the capitals of migrants from particular countries and (2) a greater geographical dispersion in new destinations (Durand and Massey, 2003). New York City, for example, was well known as Puerto Rican territory, California and Texas has been home to numerous Mexican immigrants, and Florida was the preferred destination of Cubans. Today, however, these concentrations of migrants are being diluted, and there is an increasing diversity in the interaction among distinct national groups of Latino migrants.

The states of New York and, to a greater extent, Florida are paradigmatic cases regarding Latino concentration, where a process of diversification of the Hispanic population is taking place and there is potentially more interaction among Hispanics of different national groups. Although New York State, and especially the City of New York itself, have long been the main place of residence for Puerto Ricans, they are now the principal destination for Dominicans, Ecuadorians, Panamanians, and Paraguayans. In addition, New York is the second most common destination for Colombians, Guatemalans, and Venezuelans, and it is home to many Salvadorans, Hondurans, Nicaraguans, Argentineans, Chileans, Bolivians, Peruvians, and Cubans (Durand and Massey, 2003; Durand, Massey, and Charvet, 2000). Indeed, the fourth largest group is now Mexicans, who were of an almost negligible number in that city two decades ago (Smith, 1993).

A similar situation has developed in Florida. Although Cubans predominate there, this state is also the primary place of settlement for Honduran, Nicaraguan, and Colombian migrants and the second place for Peruvians, Dominicans, Guatemalans, Bolivians, and Ecuadorians. Meanwhile, many small cities and towns in the state of New Jersey have absorbed innumerable Hispanics from diverse places of origin in the Caribbean and South America, especially Cuba, the Dominican Republic, Peru, Ecuador, and Colombia. By contrast, in California and the border states, Mexicans continue to predominate, although a growing presence of Salvadorans and Guatemalans is increasingly evident, as are South Americans and even Caribbean migrants (Durand and Massey, 2003; Durand et al., 2000). This greater interaction among Hispanics of different nationalities could promote greater social and political ties among Latino national groups as well as stronger panethnic identities.

The geographical distribution of Hispanics in the 21st century no longer seems to obey the traditional patterns of concentration, in which networks of social relationships, ethnic enclaves, and niches in labor markets function as mechanisms of attraction and permanence for this population. The past experiences of earlier waves of European immigrants of diverse nationalities show that while they were initially attracted to a particular place, they tended to disperse and form new population centers in other areas, where they once again experienced processes of concentration and residential segregation. Table 3-4 shows the Latino populations of the top 20 states in 1990 and 2000. The top 9 states remain in roughly the same order over the course of the decade, but the next 11 states reveal Latino growth in places where they were few before. The table shows growth in new states that had only small Latino populations in 1990. Between 1990 and 2000, the Hispanic population more than tripled in Nevada, grew four times in Georgia, and nearly quintupled in North Carolina. It doubled in several others.

TABLE 3-4. Latino Population (1000s) in 1990 and 2000 and Percentage Growth for Top 20 States in 2000 .


Latino Population (1000s) in 1990 and 2000 and Percentage Growth for Top 20 States in 2000 .

The new geography of the Latino population in the United States of the past two decades cannot be generalized for all Latinos. While Mexicans, Central Americans, Puerto Ricans, and Dominicans have dispersed in the past two decades, other national groups have not. Cubans, for example, have increased their residential concentration in Florida, from 60 percent in 1980 to 74 percent in 2000. As an example of residential stability, nearly 50 percent of Ecuadorians resided in New York in both 1980 and 2000.

The primacy of California and New York as Latino states is weakening, even though they continue to be the main ports of entry for immigrants from Latin America. In 1980, California was the residence of 58 percent of the Mexican population, but it fell by 15 percentage points to 43 percent in 2000. Similarly, 72 percent of Salvadorans lived in California in 1980 but only 44 percent in 2000. Guatemalans went from 58 percent in California in 1980 to 44 percent in 2000. California continues to be the principal place of residence for these three national groups, but the trend away from concentration in that state is substantial.

Similarly, New York was the state of residence for 47 percent of Puerto Ricans in 1980 but only 31 percent in 2000. At the same time, the total population of Puerto Ricans in New York declined. Just under 8 of 10 (77 percent) Dominicans lived in New York, but by 2000 only about 6 of 10 (59 percent) did. Finally, Cuban New Yorkers constituted 10 percent of all Cuban Americans in 1980 but only 4 percent in 2000. Thus, Latinos are moving away from the two traditional ports of entry. This includes both Latinos who first go to these states and then move to other states as well as a large number who are moving directly from their countries of origin to less traditional states.

The movement of these populations into these new regions has much to do with the profound structural changes that have taken place in U.S. labor markets, the availability of cheaper housing in the new regions, and processes of social and geographic mobility that affected the black and white populations (Durand and Massey, 2003; Hernández-León and Zúñiga, 2000; Stull, Broadway, and Griffith, 1995). For example, Latinos—and especially Mexicans—began to emigrate from Los Angeles to the state of Nevada in the 1980s in search of better wages. In the space of just 20 years, they have practically come to control the service industries linked to hotels, casinos, and restaurants, jobs that were previously held by native black and white workers and before that by Polish and Italian immigrants (Durand, 1994; Martinez Curiel, 2003). In addition to higher wages, Nevada offers migrants other advantages, such as ample opportunities to work double shifts and to bring their families, due to the availability of work for both men and women. Moreover, the arrival of Latinos in Nevada coincided with the recovery of the construction industry, which pursued a new phase of hotel and casino development.

Perhaps the most important industrial development for immigrants in the South is in the poultry industry, especially in chicken, duck, and turkey processing (Kandel and Parrado, 2004). There, processes of industrial conversion have been characterized by the relocation of plants to rural areas outside the cities and by efforts to reduce production costs, especially those of labor. As part of this process, older workers have been indemnified and labor unions dissolved, thus opening the way for the hiring of cheap, easily controlled workers (Griffith, 1995). New opportunities for employment have also emerged on the East Coast, especially in industries related to fish products. Fish, seafood, and crab processing plants have also gone through a process of conversion in which unionized black workers have been replaced by Latino migrants (Smith-Nonini, 2000).

Opportunities for employment also opened up in cities, especially in construction, the service sector, and manufacturing industries. The opening up of labor markets in the construction industry in Atlanta during the period leading up to the 1996 Olympic games is a particularly well-known case, because this sector had been a well-remunerated and specialized niche practically reserved for whites (Rosenfeld and Tienda, 1999). Over time, new Latino migrants have entered the construction industry and have come to occupy higher and better paid positions in it. In fact, many Latinos have become construction contractors themselves. Another example, also from the state of Georgia, is the revitalized carpet industry—in the city of Dalton—that hires only Latino migrants of Mexican and Central American origin, as Hernández-León and Zúñiga (2000) have documented.

Mexicans have gained a reputation as good workers in the agricultural labor markets in the South and on the East Coast, which have traditionally been dominated by black, poor white, and Caribbean workers. Today, almost the entire tobacco harvest depends on temporary workers who arrive from Mexico with H2A visas (see Box 3-1). Similarly, Mexican workers have begun to predominate in the agricultural belt between Florida and New Jersey by moving in a south-to-north direction in accordance with changing harvesting seasons (Griffith, 2000).

Finally, in the Midwest—especially in Minnesota, Iowa, and Nebraska—the growth of the Latino population has come largely as a response to the restructuring of the meat and poultry processing industries. As in the South, meat processing plants have been relocated outside the cities, many black and white workers have been laid off, and labor unions have been broken (Stull et al., 1995). These recently hired Latino workers have joined groups of earlier migrants who have lived and worked in this region since the 1920s and whose main points of reference are the cities of Chicago, Gary, Detroit, and St. Paul (Durand and Arias, 2000; Valdés, 1982, 1991, 2000).

Using the index of dissimilarity at the state level, Table 3-5 summarizes the extent of segregation or, alternatively, dispersion for several Hispanic subgroups across the 50 states. Dissimilarity indexes range from 0 to 100, in which 0 indicates complete dispersion into the general U.S. population and 100 indicates complete segregation. The table shows changes in state distribution from 1980 to 2000 and levels for foreign- and native-born in 2000. The first row shows the extent of dissimilarity for Hispanics with non-Hispanics since 1980 and reveals stability in the extent of segregation by state, when Hispanics are viewed as a single group. However, for subgroups, the patterns over time vary substantially. Mexican geographical dispersion remained fairly constant from 1980 to 1990 but then markedly dropped from 65 to 57, which reflects their recent emergence in nontraditional states. The last two columns show that dispersion is greater among the foreign-born. Thus, the cross-sectional evidence for 2000 demonstrates that Mexicans are bucking the trend of many other immigrant groups, who concentrate in the traditional ports of entry of earlier immigrants and then they or their descendants disperse, as they become incorporated into American society. In contrast, the dispersal of Mexicans is now being created in the immigrant generation, which is settling in new destinations and, to a great extent, eschewing traditional ones. These trends may have major implications for the way that Latinos are perceived and incorporated by American society in the future, as their geographic isolation diminishes.

TABLE 3-5. Dissimilarity Indexes at the State Level Between Hispanics and Hispanic Subgroups with the Remaining U.S. Population for 1980, 1990, and 2000 and by Foreign or U.S. Birth in 2000 .


Dissimilarity Indexes at the State Level Between Hispanics and Hispanic Subgroups with the Remaining U.S. Population for 1980, 1990, and 2000 and by Foreign or U.S. Birth in 2000 .

Puerto Rican dispersion has increased slowly but steadily since 1980. Dominicans, for which no data are available in 1980, also became slightly more dispersed between 1990 and 2000. However, Cubans, the most concentrated subgroup in a single state, have continued to steadily increase their geographic isolation in each of the last two censuses. For all the Caribbean groups, immigrants are clearly more concentrated in fewer states than the U.S.-born, and this is especially true for Cubans. Thus, Puerto Ricans, Dominicans, and Cubans follow the traditional path of regional concentration followed by dispersion for later generations, and this is especially true for Cubans. Foreign-born Cubans, despite having been in the United States for a relatively long time compared with other immigrant groups, registered a full 73.3 dissimilarity rate, indicating extremely high concentration of immigrants, perhaps because of an exceptionally successful ethnic enclave in Southern Florida (Portes and Stepick, 1993). Still, U.S.-born Cubans had a rate of only 51.8, suggesting substantial movement of their children out of that state.

For the other groups, only 1990 and 2000 data are available. Rates of Central American segregation across states, like those for Mexicans, dropped sharply from 1990 to 2000. Central Americans registered the greatest dispersion among all of the subgroups. Dominicans and South Americans have registered barely noticeable declines in segregation, while Central Americans are increasingly dispersed, registering the lowest segregation by state among all Hispanic subgroups. South Americans as a whole follow the traditional path of dispersion by generation, while generational differences for Central Americans are small, suggesting a pattern that is between the traditional immigrant-native settlement path and the new settlement path of the Mexicans. Thus, the increasing distribution of Hispanics, as a group, is driven mostly by the new geographical trends in settlement among Mexican and Central American immigrants.


The Hispanic population is likely to continue its growth trend for at least the near future, especially considering that Latin America immigration has not ebbed. The labor market and political incentives appear to continue firmly in place. Based on the most likely future immigration scenario, in which immigration continues at about the current rate, Jeffrey Passel (2003) of the Urban Institute predicts that the Hispanic population will grow from 35 million in 2000 to 101 million by 2050, nearly a three-fold increase. Consequently, the Hispanic component will constitute fully 24.6 percent of the U.S. population, nearly double its 12.5 percent in 2000, meaning that Latinos are likely to account for about 45 percent of U.S. population growth in the next 50 years. Even if immigration were to suddenly end, the youthful composition and higher fertility rates of Latinos ensure continued growth.

Even if there were no immigration after 2000, the Latino population would still increase to 63 million in 2050 or to about 20 percent of the U.S. population. The extent to which immigration will continue at the same rate depends partly on the actions of the U.S. government to effectively conduct border control, as it has suggested in recent years. Also, immigration from Latin America may begin to slow, as new entrants to the labor force of sending countries begin to ebb as a result of rapidly decreasing fertility over the past 20 years. Finally, immigration may also decrease in an optimistic scenario in which Latin American economies recuperate and their labor force needs begin to increase.

Currently, 40 percent of the Latino population is foreign-born. However, under the middle immigration scenario, the share of immigrants among the population will decline to 24 percent. The children of immigrants or the second generation currently represents 28 percent of Latinos, but by 2050 their population will quadruple and will represent 39 percent. The so-called third generation (and beyond) will grow from 32 to 37 percent.

Regarding the national composition of the Latino population, the order by population size is likely to change in the near future as Dominicans and other national populations that have become part of the new immigration stream overtake the Cuban and Puerto Rican populations, which grew less than 20 percent each decade since 1980. Dominican immigrants, for example, grew at about 100 percent in each of the past two decades. Cuban immigration has had only intermittent immigration for the past 30 years. Similarly, Puerto Rican immigration cannot grow much, since a large portion of the island's population is already engaged in the immigration process to the mainland, making the pool of potential immigrants relatively small. Like Dominicans, the Mexican population also doubled in the 1980s and then again in the 1990s. The Central American countries of El Salvador, Guatemala, and Honduras grew even more.

Latin American immigrant populations, which seem to be small today, may grow in size, depending on the expansion of networks and especially the relative opportunities between their countries and the United States. This is particularly true for South Americans, who constitute the majority of the Latin American population and are only beginning to participate in the immigration process to the United States. Brazil, in particular, with a population of 170 million but less than 1 million immigrants in the United States, may potentially become a leading sender of immigrants to the United States, especially if conditions in that country greatly deteriorate. Brazilians are Latin Americans, and whether we consider them part of the Latino/Hispanic population will depend on how their identity in the United States is shaped, how others will consider them, especially the U.S. Census Bureau, and especially whether we use a Latino or Hispanic label. For now, the small size of that population makes their inclusion of little consequence.

It is hard to tell whether the growing spread of Latino populations is likely to continue, although we are inclined to think that they are. Apparently, these immigrants are gaining footholds in such nontraditional places as the South, but it is probably too early to tell how solid they will become. The experience of Cubans and Vietnamese, who were relocated throughout the United States but the majority of whom later resettled in one metropolitan area, may be telling. However, those immigrants were refugee populations, and decisions about their settlement were made by others. Today's Latino populations seem to be choosing these destinations for the economic opportunities they provide. With the integration process experienced by most immigrants and their descendants, the more common trend is residential dispersion, especially by the second and third generation.

An issue that is likely to increase in importance, especially with the emergence of a third generation, is identification as Latino or Hispanic. The data presented in official statistics depend on persons identifying as Latinos, Hispanic, or in one of the national subgroups. This does not seem to be a major issue for immigrants from Latin American countries. However, many of these second- and third-generation descendants of Latino immigrants may no longer identify as Latino or Hispanic, especially if their parents or grandparents include non-Latinos. The ability to identify as white rather than Latino may be especially appealing to light-skinned people and children of mixed marriages, given the higher status of (non-Hispanic) whites in American society. Eschbach and Gomez (1998), for example, found that as many as 20 percent of high school students identified as Hispanic in a survey during their sophomore year and as white in a survey two years later. These trends will surely vary by national group and the place of settlement. The experience in the South, with its historically important black–white dichotomy, will be especially interesting.

For the labor force, Passel expects the Latino labor force to grow from representing 11.5 percent of the labor force in 2000 to 25 percent in 2050. This is the middle immigration assumption, but the Latino share would increase under all of the immigration scenarios. Under the no-immigration assumption, the Latino labor force would still increase by 82 percent, while the non-Latino labor force would shrink. Regarding the educational composition of the labor force, Latinos with a college education will grow from 10 to 25 percent, which is greater than that for the overall labor force. At the end, the number of those without a high school diploma will shrink from 39 to 17 percent of the Latino workforce.


This chapter has emphasized differences among national groups, but we have also sought to stress commonalities where they occur. In addition, there are some signs that Latinos have begun to adapt and incorporate themselves in the United States as a pan-ethnic group. Despite class, race, and national differences, Latinos have found ways to coalesce to some extent based on a shared language and cultural traits and especially because of a common identity forged by categorization by U.S. natives as a single racial or ethnic group. The extent to which a Latino or Hispanic identity forms largely depends on similar interests and structural foundations, such as a common geography. The historically separate regional concentration of particular national groups represents a prime reason for separateness, but the growing dispersion of various national groups into overlapping places presents growing opportunities for pan-Latino identity. For now, though, Latinos tend to favor identities based on national origin.

One issue we have dealt with in this chapter but that does not arise in many of the subsequent chapters, largely because of the lack of data, is the issue of illegal status. Many Latinos continue to confront serious problems in their everyday lives because of their illegal and lower class status, and this may have particularly important implications for their future. Most Latinos are in the working class, have low levels of education, and receive low salaries, characteristics that impede their ability to assimilate into the country they have chosen. However, the undocumented among them especially suffer from these and also from the lack of access to formal institutions. The demographic foundations today have important implications for the future of the Latino population in various social dimensions, as the rest of the book shows.

The demographic trends outlined in this chapter have large and direct implications for all of the issues treated in subsequent chapters. Such demographic characteristics as population growth, concentration in young ages, significant rates of naturalization, and dispersion into new locales are important for understanding education, family life, labor force participation, voting, the criminal justice system, and health care. The changing geography of Latinos should also focus the attention of the public on their situation in places where they never existed before. The dispersion of the Latino population into new states may also suggest changing prospects for assimilation into American society. However, the next chapter shows that this dispersion has been accompanied by a resegregation in the cities of the new areas.


  1. Bean FD, Telles EE, Lindsay Lowell B. Undocumented migration to the United States: Perceptions and evidence. Population and Development Review. 1987;13(4):671–689.
  2. Bean FD, Lee J, Batalova J, Leach M. Immigration and fading color lines in America. In: Carnevale E, Kent MM, Tarmann A, editors. The American people. New York: Russell Sage Foundation; 2004.
  3. Calavita K. Inside the state: The Bracero Program, immigration, and the INS. New York: Routledge; 1992.
  4. Cardoso L. Mexican emigration to the United States, 1897–1931. Tucson: University of Arizona Press; 1980.
  5. Cobble S, Velaquez J. Looking back, looking forward. The Nation. 2004 December:11.
  6. Donato KM, Durand J, Massey DS. Changing conditions in the U.S. labor market: Effects of the Immigration Reform and Control Act of 1996. Population Research and Policy Review. 1992;11(2):93–115.
  7. Duany J. Los dominicanos en Puerto Rico: Migración en la semi-periferia. San Juan, Mexico: Ediciones el Huracán; 1990.
  8. Duany J, Hernández Angueira L, Rey CA. El barrio Gandul: Economía subterránea y migración indocumentada en Puerto Rico. San Juan, Mexico: Ediciones Nueva Sociedad; 1995.
  9. Durand J. Más allá de la línea: Patrones migratorios en el Occidente de México. México D.F. CONACULTA; 1994.
  10. Durand J. Política, modelo y patrón migratorios: El trabajo y los trabajadores mexicanos en Estados Unidos. San Luis PotosíEl Colegio de San Luis, Serie Cuadernos del Centro; 1998.
  11. Durand J, Arias P. La experiencia migrante: Historia e iconografía de la migración México Estados Unidos. Guadalajara, Mexico: Universidad de Guadalajara; 2000.
  12. Durand J, Massey DS. Clandestinos: Migración México Estados Unidos en los albores del siglo XXI. México D.F: Editorial Miguel Angel Porrua; 2003.
  13. Durand J, Massey DS, Charvet F. The changing geography of Mexican immigration to the United States: 1910–1996. Social Science Quarterly. 2000;81(1):1–15.
  14. Eschbach K, Gomez C. Choosing Hispanic identity: Ethnic identity switching among respondents to High School and Beyond. Social Science Quarterly. 1998;79(1):74–90.
  15. Georges E. The making of a transnational community. New York: Columbia University Press; 1990.
  16. Grasmuck S, Pessar PR. Dominican international migration. Los Angeles: California University Press; 1991. Between two islands.
  17. Grenier G, Pérez L. The legacy of exile: Cubans in the United States. Boston, MA: Allyn and Bacon; 2003.
  18. Griffith D. Hay trabajo: Poultry processing, rural industrialization and the latinization of low-wage labor. In: Stull DD, Broadway MJ, Griffith D, editors. Any way you cut it: Meat processing and small town America. Lawrence: University Press of Kansas; 1995. pp. 129–151.
  19. Griffith D. Work and immigration: Winter vegetable production in South Florida. In: Tardanico R, Rosenberg MB, editors. Poverty or development. New York: Routledge; 2000. pp. 139–178.
  20. Hamilton N, Stoltz Chinchilla N. Seeking community in global city: Guatemalans and Salvadorans in Los Angeles. Philadelphia, PA: Temple University Press; 2001.
  21. Hernández-León R, Zúñiga V. Making carpet by the mile: The emergency of a Mexican immigrant community in an industrial region of the U.S. historic South. Social Science Quarterly. 2000;81(1):49–65.
  22. Hood MV III, Morris IL. Brother, can you spare a dime? Racial/ethnic context and the Anglo vote on Proposition 187. Social Science Quarterly. 2000;81(1):194–206.
  23. Huber E, Solt F. Successes and failures of neoliberalism. Latin American Research Review. 2004;39(3):150–164.
  24. Jasso G, Rosenzweig MR. The new chosen people: Immigrants in the United States. New York: Russell Sage Foundation; 1990.
  25. Kandel W, Parrado E. U.S. industrial transformation and new Latino migration. Prepared for publication by the Migration Policy Institute, Washington, DC, based upon a presentation at the 2003 Annual Meeting of the Population Association of America; Minneapolis, MN. May 1-3; 2004. [accessed January 2006]. Available:http://www​.migrationinformation​.org/Feature/display​.cfm?ID=217.
  26. Martin JA, Hamilton BE, Ventura SJ, Menacker F, Park MM. Births: Final data for 2000. Hyattsville, MD: National Center for Health Statistics; 2002.
  27. Martínez Curiel E. Hasta que la muerte nos separe. Guadalajara, Mexico: Universidad de Guadalajara; 2003.
  28. Massey D, Alarcón R, Durand J, González H. Return to Aztlán. Berkeley: University of California Press; 1987.
  29. Massey D, Durand J, Malone N. Beyond smoke and mirrors. New York: Russell Sage Foundation; 2002.
  30. Menjívar C. Fragmented ties: Salvadoran immigrant networks in America. Berkeley: California University Press; 2000.
  31. Ortíz V. The Mexican origin population: Permanent working class or emerging middle class? In: Waldinger R, Bozorgmehr M, editors. Ethnic Los Angeles. New York: Russell Sage Foundation; 1996. pp. 247–278.
  32. Passel JS. Hispanic projections: The future and the past. Presentation to the Panel on Hispanics in the United States; Washington, DC. 2003.
  33. Passel JS, Fix M. Undocumented immigrants: Facts and figures. Urban Institute Immigration Studies Program. 2004 unpublished.
  34. Pessar PR. A visa for a dream. London: Allyn and Bacon; 1995.
  35. Portes A, Rumbaut R. Immigrant America: A portrait. Los Angeles: University of California Press; 1990.
  36. Portes A, Stepick A. City on the edge: The transformation of Miami. Los Angeles: University of California Press; 1993.
  37. Reimers DM. Still the golden door. New York: Columbia University Press; 1992.
  38. Rosenfeld MJ, Tienda M. Mexican immigration, occupational niches, and labor market competition: Evidence from Los Angeles, Chicago, and Atlanta, 1970–1990. In: Bean FD, Bell-Rose S, editors. Immigration and opportunity. New York: Russell Sage Foundation; 1999.
  39. Smith R. Los ausentes siempre presentes: The imagining, making, and politics of a transnational community between New York City and Ticuani, Puebla. Institute for Latin American and Iberian Studies, Columbia University; 1993. Working Papers on Latin America.
  40. Smith-Nonini S. Nadie sabe, nadie supo: El programa federal H2A y la explotación de mano de obra mediada por el estado. Relaciones. 2002;23(90):55–86.
  41. Stull DD, Broadway MJ, Griffith D. Any way you cut it: Meat processing and small town America. Lawrence: University Press of Kansas; 1995.
  42. United Nations. Database, Department of Economic and Social Affairs, Population Division. New York: United Nations; 2003. World population prospects 1950–2050: The 2002 revision.
  43. U.S. Census Bureau. Index of census 2000 data sets. 2000. [accessed May 20, 2004]. Available:http://www2​​/census_2000/datasets​/Sim_Hispanic_Totals.
  44. U.S. Immigration and Naturalization Service. Statistical yearbook of the Immigration and Naturalization Service. Pittsburgh, PA: Superintendent of Documents; 2002.
  45. Valdés DN. El pueblo mexicano en Detroit y Michigan: A social history. Detroit, MI: Wayne State University; 1982.
  46. Valdés DN. Al norte: Agricultural in the Great Lakes region, 1917–1970. Austin: University of Texas Press; 1991.
  47. Valdés DN. Barrios norteños: St. Paul and Midwestern Mexican communities in the twentieth century. Austin: University of Texas Press; 2000.
  48. Walton M. Neoliberalism in Latin America: Good, bad, or incomplete? Latin American Research Review. 2004;39(3):165–184.
  49. Weintraub S. U.S. policy and Mexican immigration. In: Bean FD, de la Garza R, Roberts B, Weintraub S, editors. At the crossroads: Mexican migration and U.S. policy. New York: Rowman and Littlefield; 1997. pp. 283–298.



The TPS provides nationals of certain regions or nationalities residing in the United States a temporary stay of removal and work authorization due to armed conflict or other extraordinary conditions in their home country.


It is important to note, however, that the so-called generations that we use are based on cross-sectional data and thus do not capture generational change. That is, the second generation in 2000, for example, consists not of children of the first generation in 2000 but rather children of a first generation cohort in earlier years. Also, the so-called Mexican third generation is actually comprised of many generations, including direct descendants of residents in the Southwest states prior to the U.S. occupation in 1848.


Fertility rates for Mexico, Cuba, and Puerto Rico in 2000 can be found in United Nations (2003).


The age–sex pyramids for Dominicans, other Central Americans, and South Americans do not contain data on third-generation immigrants in the United States. The Current Population Survey contained only nativity data for the aggregated group of Central-South Americans. According to the cross-tabulation of nativity and country of origin, only 823 Central-South Americans, or 4 percent, are identified as third-generation Americans.

Copyright © 2006, National Academy of Sciences.
Bookshelf ID: NBK19901


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (2.4M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...