The nation’s 53 million Hispanics comprise 17% of the total U.S. population but just 10% of all voters this year, according to the national exit poll. To borrow a boxing metaphor, they still “punch below their weight.”

However, their share of the electorate will rise quickly for several reasons. The most important is that Hispanics are by far the nation’s youngest ethnic group. Their median age is 27 years—and just 18 years among native-born Hispanics—compared with 42 years for that of white non-Hispanics. In the coming decades, their share of the age-eligible electorate will rise markedly through generational replacement alone.

According to Pew Hispanic Center projections, Hispanics will account for 40% of the growth in the eligible electorate in the U.S. between now and 2030, at which time 40 million Hispanics will be eligible to vote, up from 23.7 million now.²

Moreover, if Hispanics’ relatively low voter participation rates and naturalization rates were to increase to the levels of other groups, the number of votes that Hispanics actually cast in future elections could double within two decades.

If the national exit poll’s estimate proves correct that 10% of all voters this year were Hispanic, it would mean that as many as 12.5 million Hispanics cast ballots. But perhaps a more illuminating way to analyze the distinctive characteristics of the Hispanic electorate—current and future—is to parse the more than 40 million Hispanics in the United States who did not vote or were not eligible to vote in 2012. That universe can be broken down as follows:

• 11.2 million are adults who were eligible to vote but chose not to. The estimated 44% to 53% turnout rate of eligible Hispanic voters in 2012 is in the same range as the 50% who turned out in 2008. But it still likely lags well below the turnout rate of whites and blacks this year.³
• 5.4 million are adult legal permanent residents (LPRs) who could not vote because they have not yet become naturalized U.S. citizens. The naturalization rate among legal immigrants from Latin America and the Caribbean trails that of other legal immigrants by a sizable margin—49% versus 72%, according to a Pew Hispanic analysis of the 2011 March Current Population Survey (CPS). The new Pew Hispanic survey finds that a major reason Hispanic immigrants naturalize is to gain civil and legal rights, including the right to vote. The flexing of electoral muscle by Hispanic voters this year conceivably could encourage more legal immigrants to become naturalized citizens.
• 7.1 million are adult unauthorized immigrants and would become eligible to vote only if Congress were to pass a law creating a pathway to citizenship for them. Judging by the immediate post-election comments of leading Democratic and Republican lawmakers, the long-dormant prospects for passage of such legislation appear to have been revived by Latinos’ strong showing at the polls.
• 17.6 million are under the age of 18 and thus too young to vote—for now. That vast majority (93%) of Latino youths are U.S-born citizens and thus will automatically become eligible to vote once they turn 18. Today, some 800,000 Latinos turn 18 each year; by 2030, this number could grow to 1 million per year, adding a potential electorate of more than 16 million new Latino voters to the rolls by 2030.

Thus, generational replacement alone will push the age- and citizen-eligible Latino electorate to about 40 million within two decades. If the turnout rate of this electorate over time converges with that of whites and blacks in recent elections (66% and 65%, respectively, in 2008), that would mean twice as many Latino voters could be casting ballots in 2032 as did in 2012.

This turnout could rise even more if naturalization rates among the 5.4 million adult Hispanic legal permanent residents were to increase over time—and/or if Congress were to pass a comprehensive immigration reform bill that creates a pathway to citizenship for the more than 7 million unauthorized Hispanic immigrants already living in the U.S.

The Pew Hispanic Center survey finds that more than nine-in-ten (93%) Hispanic immigrants who have not yet naturalized say they would if they could. Of those who haven’t, many cite administrative costs and barriers, a lack of English proficiency and a lack of initiative. For example, according to the survey, only 30% of Hispanic immigrants who are LPRs say they speak English “pretty well” or “very well.”

In addition to all these factors, there is the as-yet-unknowable size and impact of future immigration. About 24 million Hispanic immigrants have come to U.S. in the past four decades—in absolute numbers, the largest concentrated wave of arrivals among any ethnic or racial group in U.S. history. Some 45% arrived in the U.S. legally, and 55% arrived illegally.⁴

Assuming Hispanic immigration continues into the future —even at the significantly reduced levels of recent years—the Hispanic electorate will expand beyond the numbers dictated by the growth among Hispanics already living in the U.S. And because immigrants tend to have more children than the native born, the demographic ripple effect of future immigration on the makeup of the electorate will be felt for generations.

In 2008, the Pew Research Center projected that the Hispanic share of the total U.S. population would be 29% by 2050 (Passel and Cohn, 2008). Since that projection was made, the annual level of Hispanic immigration has declined sharply (Passel, Cohn and Gonzalez-Barrera, 2012). Because of this decline, the share of Hispanics in 2050 now appears unlikely to reach 29%. However, the 2008 projection also included a “low immigration scenario” that showed the Hispanic share of the U.S. population would be 26% by mid-century (Passel and Cohn, 2008)—still much higher than today’s 17%.

Who Naturalizes and Who Doesn’t

A record 15.5 million legal immigrants were naturalized U.S. citizens in 2011, according to a Pew Hispanic Center analysis of Census Bureau data. In addition, the share of the nation’s legal immigrants who have become U.S. citizens has reached its highest level in three decades—56%. However, naturalization rates among legal immigrants from Latin America and the Caribbean (49%), especially Mexican legal immigrants (36%), remain below those of other immigrants (72%).

In the new Pew Hispanic Center survey, when asked in an open-ended question why they decided to naturalize, almost one-in-five (18%) naturalized Hispanic immigrants said that acquiring civil and legal rights—including the right to vote—was the main reason. This response was closely followed by an interest in having access to the benefits and opportunities derived from U.S. citizenship (16%) and family-related reasons (15%). Other reasons included viewing the U.S. as home (12%) and wanting to become American (6%).

The Pew Hispanic survey also explored the reasons Hispanic immigrants who are legal permanent residents haven’t yet tried to become citizens. According to the survey, when asked in an open-ended question why they had not naturalized thus far, 45% identified either personal barriers (26%), such as a lack of English proficiency, or administrative barriers (18%), such as the financial cost of naturalization.

About this Report

This report explores the growing size of the Hispanic electorate and the reasons Hispanic immigrants give for naturalizing to become a U.S. citizen—and for not naturalizing.

The report uses several data sources. Latino vote shares are based on the National Election Pool national exit poll as reported on November 6, 2012, by CNN’s Election 2012 website. Data on Latino immigrants’ views of naturalization are based on the Pew Hispanic Center’s 2012 National Survey of Latinos (NSL). The NSL survey was conducted from September 7 through October 4, 2012, in all 50 states and the District of Columbia among a randomly selected, nationally representative sample of 1,765 Latino adults, 899 of whom were foreign born. The survey was conducted in both English and Spanish on cellular as well as landline telephones. The margin of error for the full sample is plus or minus 3.2 percentage points. The margin of error for the foreign-born sample is plus or minus 4.4 percentage points. Interviews were conducted for the Pew Hispanic Center by Social Science Research Solutions (SSRS).

For data on the legal status of immigrants, Pew Hispanic Center estimates use data mainly from the Current Population Survey (CPS), a monthly survey of about 55,000 households conducted jointly by the U.S. Bureau of Labor Statistics and the Census Bureau. It is best known as the source for monthly unemployment statistics. Each March, the CPS sample size and questionnaire are expanded to produce additional data on the foreign-born population and other topics. The Pew Hispanic Center estimates make adjustments to the government data to compensate for undercounting of some groups, and therefore its population totals differ somewhat from the ones the government uses. Estimates of the number of immigrants by legal status for any given year are based on a March reference date. For more details, see Passel and Cohn (2010).

This report was written by Director Paul Taylor, Research Associate Ana Gonzalez-Barrera, Senior Demographer Jeffrey S. Passel and Associate Director Mark Hugo Lopez. Ana Gonzalez-Barrera took the lead in developing the survey questionnaire’s naturalization section. Passel and D’Vera Cohn provided comments on earlier drafts of the report. The authors also thank Scott Keeter, Leah Christian, Cohn, Richard Fry, Cary Funk, Rakesh Kochhar, Rich Morin, Seth Motel, Kim Parker, Passel, Eileen Patten and Antonio Rodriguez for guidance on the development of the survey instrument. Motel provided excellent research assistance. Fry, Morin and Patten number-checked the report text and topline. Marcia Kramer was the copy editor.

A Note on Terminology

The terms “Latino” and “Hispanic” are used interchangeably in this report.

“Foreign born” refers to persons born outside of the United States, Puerto Rico or other U.S. territories to parents neither of whom was a U.S. citizen.

The following terms are used to describe immigrants and their status in the U.S. In some cases, they differ from official government definitions because of limitations in the available survey data.

Legal permanent resident, legal permanent resident alien, legal immigrant, authorized migrant: A citizen of another country who has been granted a visa that allows work and permanent residence in the U.S. For the analyses in this report, legal permanent residents include persons admitted as refugees or granted asylum.

Naturalized citizen: Legal permanent resident who has fulfilled the length of stay and other requirements to become a U.S. citizen and who has taken the oath of citizenship.

Unauthorized migrant: Citizen of another country who lives in the U.S. without a currently valid visa.

Eligible immigrant: In this report, a legal permanent resident who meets the length of stay qualifications to file a petition to become a citizen but has not yet naturalized.

Legal temporary migrant: A citizen of another country who has been granted a temporary visa that may or may not allow work and temporary residence in the U.S.

Hispanics now make up 22% of all children under the age of 18 in the United States–up from 9% in 1980–and as their numbers have grown, their demographic profile has changed.

A majority (52%) of the nation’s 16 million Hispanic children are now “second generation,” meaning they are the U.S.-born sons or daughters of at least one foreign-born parent, typically someone who came to this country in the immigration wave from Mexico, Central America and South America that began around 1980. Some 11% of Latino children are “first generation”–meaning they themselves are foreign-born. And 37% are “third generation or higher”–meaning they are the U.S.-born children of U.S.-born parents.

In 1980, only three-in-ten Latino children were second generation, while nearly six-in-ten were in the third generation or higher. These shifts are noteworthy because many social, economic and demographic characteristics of Latino children vary sharply by their generational status. A Pew Hispanic Center analysis of U.S. Census Bureau data finds that:

• 43% of first-generation Latino children, 21% of those in the second generation and 5% in the third generation or higher are not fluent in English.
• 47% of first-generation Latino children have parents who have less than a high school education, compared with 40% of second-generation children and 16% of Latino children in the third generation or higher.
• 34% of first-generation Latino children live in poverty, compared with 26% of those in the second generation and 24% in the third generation or higher.
• 69% of first-generation Latino children live in married-couple families, compared with 73% of second-generation children and just 52% in the third generation or higher.

Most of these data reflect the classic pattern of socioeconomic gains for immigrant families from one generation to the next. However, there are also some contrary trends. As indicated above, the chances of being raised by a single parent are much greater among Latino children in the third generation or higher than among first- or second-generation Latino children.¹

Another characteristic that separates Latino children along generational lines is their legal status. Building on earlier research² the Pew Hispanic Center estimates that 7% of all Hispanic children are unauthorized immigrants. But this share varies sharply by generational status. Two-thirds of the 1.7 million foreign-born Hispanic children are unauthorized, while none of the 6 million Hispanic children in the third generation or higher are unauthorized (as the U.S.-born children of U.S.-born parents, by definition they are U.S. citizens at birth). As for those in the middle–the second generation–about four-in-ten have at least one unauthorized immigrant parent and are therefore living in a family whose immigration status is legally mixed.

Projects by the U.S. Census Bureau indicate that by 2025, nearly three-in-ten children in this country will be of Latino ancestry. Pew Hispanic Center population projects indicate that the generational composition of Hispanic children will change yet again between now and then. Today’s large cohort of second-generation Hispanic children will eventually form families and produce third-generation offspring. The Center’s projections show that the share of Hispanic children who are second generation will likely peak not much above the current level of 52% and then begin falling roughly a decade from now. The share of Hispanic children who are third generation or higher will likely hit bottom at 35% around 2015 and then begin rising. Given past trends, this change in the generational composition of Hispanic children could well have an impact on their employment, education and social outcomes in future years.

About this Report

This report present findings from several existing and new Pew Hispanic Center analyses of U.S. Census Bureau data as well as Pew Hispanic Center population projections. The analysis of the legal status of Hispanic children underlying Table 1 utilized the augmented March 2008 Current Population Survey. See Passel and Cohn (2009) for further details. The methods and assumptions underlying the generational projections of Hispanic children from 2010 to 2025 are described in detail in Passel and Cohn (2008). The historical and current profile of Hispanic children derives from new analyses of Decennial Census and American Community Survey data. See the data appendix for further details.

A Note on Terminology

Unless otherwise noted, this report uses the following definitions of the first, second, and third or higher generations:

First: Foreign-born; an individual who is not a U.S. citizen at birth or, in other words, who is born outside the U.S., Puerto Rico or other U.S. territories and whose parents are not U.S. citizens.

Second: An individual who is a U.S. citizen at birth (including people born in the United States, Puerto Rico or other U.S. territories, as well as those born elsewhere to parents who are U.S. citizens) with at least one first-generation parent.

Third or higher: An individual who is a U.S. citizen at birth with both parents U.S. citizens at birth.
The terms “Latino” and “Hispanic” are used interchangeably in this report, as are the terms “foreign born” and “immigrant.”

Recommended Citation

Richard Fry and Jeffrey S. Passel. “Latino Children: A Majority Are U.S.-Born Offspring of Immigrants.” Washington, D.C.: Pew Hispanic Center (May 2009).

If current trends continue, the demographic profile of the United States will change dramatically by the middle of this century, according to new population projections developed by the Pew Research Center.¹

The nation’s population will rise to 438 million in 2050, from 296 million in 2005, and fully 82% of the growth during this period will be due to immigrants arriving from 2005 to 2050 and their descendants. (Figure 1)

Of the 117 million people added to the population during this period due to the effect of new immigration, 67 million will be the immigrants themselves, 47 million will be their children and 3 million will be their grandchildren.

The Center’s projections indicate that nearly one in five Americans (19%) will be foreign born in 2050, well above the 2005 level of 12%, and also surpassing the historic peaks for immigrants as a share of the U.S. population—14.8% in 1890 and 14.7% in 1910. (Figure 2)

By 2050, the nation’s racial and ethnic mix will look quite different than it does now. Non-Hispanic whites, who made up 67% of the population in 2005, will be 47% in 2050. Hispanics will rise from 14% of the population in 2005 to 29% in 2050. Blacks were 13% of the population in 2005 and will be roughly the same proportion in 2050. Asians, who were 5% of the population in 2005, will be 9% in 2050.

Immigration is projected to be the key driver of national population growth in the coming half century, but it is important to note that possible future changes in immigration policy or other events could substantially alter the projected totals. These projections are based on trends over the past half century, during which immigration, both authorized and unauthorized, has played an escalating role in U.S. population growth. From 1960 to 2005, new immigrants and their U.S.-born descendants accounted for 51% of population increase. In the later part of that period, from 1980 to 2005, new immigration accounted for 58% of the 68 million additional people.

For the period from 2005-2050, new immigrants and their descendants will account for 82% of population increase. The contribution of new immigration to population change was derived by comparing our main projection with an alternative projection that assumes no new immigrants arrive after 2005. (See Contribution of Immigration to Population Change in Appendix.)

The heightening role of immigration contrasts with a decrease in fertility in recent decades. The average number of births per woman has declined markedly since the late 1950s, from more than 3.5 then to about 2 now. Also, a smaller proportion of women are of childbearing age now, compared with earlier decades. These two changes have made immigration a more prominent factor in population growth.

Methods and Assumptions

The Center’s projections for the period from 2005 to 2050 are based on detailed assumptions about patterns in births, deaths and immigration—the three building blocks of population change. All population projections have built in uncertainties, especially for years further in the future, because they are based on assumptions about future behavior. In addition, these uncertainties can multiply because key aspects of population change are often interrelated—for example, a decline in immigration could also lead to a decline in the birthrate because immigrants tend to have larger families than do native born residents.
The Center has developed three different population projections for 2050, but the body of this report presents findings from the main projection (figures from projections based on lower or higher immigration levels are set forth in a section that starts on page 23). These projections consolidate and build upon past trends, present conditions, and factors affecting future behavior.

None of the projections should be treated as predictions. The country’s policies may change, as may the factors that influence birth, death and immigration rates. Even given these caveats, however, population projections are an important analytical tool for planners. A rise or decline in the overall population—as well as in particular age groups—will have important impacts on the nation’s tax base and workforce. Demographic change has major implications for government spending in key areas such as schools, health programs, community services, infrastructure and Social Security. Projections also provide business with a basis upon which to make judgments about future markets. And they are of increasing interest because of the role that population may play in climate change and other environmental concerns.

When incorporating birth estimates into the projections, the Center has assumed that the overall fertility rate will remain near the level it has been for the past three decades, with differing rates by race and ethnicity (Appendix, Figure A2). Birthrates are assumed to be well above average for immigrants, slightly above average overall for the second generation (U.S. natives with at least one immigrant parent) and below average for subsequent generations born in the United States. The Center’s fertility rate projections are virtually identical to those of the Social Security trustees and slightly lower than those of the Census Bureau. (Census Bureau, 2004; Social Security Administration, 2007).

As for death rates, life expectancy is assumed to improve somewhat for all groups throughout the period covered by these projections. (Appendix, Figure A3) The Center’s projections employ Census Bureau assumptions about life expectancy gains, which are higher than those of the Social Security Trustees.

Immigration to the United States has risen rapidly and steadily for decades as a result of increasing globalization and population movements, changes in U.S. immigration laws, the growing linkages of immigrant families within this country to communities abroad and labor market factors. Not only have the numbers of new U.S. immigrants increased over recent periods, but the rate of immigration also has risen steadily, whether measured from 1930, 1960 or 1980. In the face of these strong and persistent trends, most U.S. government projections, whether done by the Census Bureau (1996, 2000, 2004) or the Social Security Administration (2007) have assumed constant or even decreasing numbers of immigrants, implying sharp and sometimes immediate declines in the rate of immigration. As a result, official projections over the last several decades have consistently underestimated actual population growth.

The Pew Research Center projections have assumed that the annual immigration level, now about 1.4 million people, will increase slowly by 1% per year, reaching 2.1 million immigrants in 2050. (Figure 3)

This rate of growth is in line with, but somewhat slower than, the growth trends of the last several decades. These immigration levels are slightly higher than those projected by either the Census Bureau or the Social Security Trustees in the short run and substantially higher toward the end of the projection horizon.

With the Center’s immigration assumption, the rate of immigration remains roughly constant over the 45-year projection horizon at 0.48% per year, or just under five immigrants per 1,000 population for each year. The rate is slightly below the rate for the first half of this decade and equal to the average for the last 35 years. (Figure 4)

The decades-long pattern of steady increases has been interrupted recently by year-to-year variation, including a spike in 1998-2001, a sharp decline in 2002-2004 (Passel and Suro, 2005), followed by a return to the long-term average in the last several years. The relatively steady growth of the last 70 years contrasts with substantial fluctuations that occurred in the 19th and early 20th centuries. (Figure 5)

Short-term fluctuations are likely to continue to occur, but the Center’s projections assume that those variations will be less important than the long-term trends. Although the rate is held steady in the Center’s projections, the number of new immigrants rises as the nation’s total population goes up. The projections also assume that several hundred thousand foreign-born residents will leave each year, which is in keeping with trends of the past several decades. Reflecting recent trends, the Center’s projections include a mix of new arrivals of legal permanent residents and unauthorized, or illegal, migrants, although the two groups are not broken out separately.

The issue of illegal immigration has become highly contentious in recent years. Last summer, Congress tried but failed to pass a comprehensive reform bill, and the debate over how to change immigration policies has become a major topic of the current presidential campaign. It is possible that a future Congress will enact laws that would sharply cut immigration flows. This has happened before. The Immigration Act of 1924 (along with an economic depression and a world war) drastically reduced immigrants as a share of the U.S. population from a 20th century peak of 14.7% in 1910 to a low of 4.7% in 1970.

This report offers two alternative population projections in addition to its main projection. The alternatives are based on immigration levels roughly 50% above and 50% below the baseline projection, but use the same fertility and mortality assumptions as the baseline projection. Under the lower-immigration scenario, the population would rise to 384 million in 2050, and new immigration would account for 71% of growth during the projections period. Under the higher-immigration scenario, the population would go up to 496 million, and new immigration would account for 87% of the increase. The baseline projection shows the population will rise to 438 million, and new immigration will account for 82% of the increase.

Population Trends

The Center’s main projection indicates that the nation’s population will grow 48% over the 2005–2050 period. That growth rate is lower than the 64% increase in the nation’s population from 1960 through 2005. The projected annual growth rate of 0.9% equals that of the 1980s but is lower than the pace of growth during other decades since the 1960s.

The Center’s projections also indicate that between 2005 and 2050 the number of elderly will increase more rapidly than either the number of children or working-age adults.² Immigration and births to immigrants in the United States will be responsible for all growth of other age groups but will have little impact on the number of elderly, which is affected mainly by the aging of the post-World War II baby-boom generation.

The relationship between the size of the working-age population, on the one hand, and the young and elderly on the other hand, is sometimes referred to as a nation’s “dependency ratio.” The Center’s projections show that the dependency ratio, which was 59 young and elderly for every 100 people of working age in 2005, will rise to 72 dependents per 100 people of working age in 2050. That means the costs per worker to support the young and elderly would go up. Under a lower-immigration scenario, the ratio would rise even higher, to 75 dependents per 100 people of working age. Under a higher-immigration scenario, it would be 69 dependents per 100 working-age people.

In terms of international comparisons, the United States will continue to be the world’s third most populous nation in 2050, behind India and China, each with more than a billion people (United Nations, 2007). The projected annual growth rate for the United States will continue to exceed that in most other developed nations, which are growing at a slower pace, if at all. European countries generally are growing at no more than 0.5% a year, and some are losing population.

As is true in the United States, the immigrant populations in many other developed nations have been growing rapidly in recent decades. The United States has a larger foreign-born population than any other country, but U.S. immigrants were a smaller share of the 2005 population (12%) than those in a few countries with long histories of receiving immigrants—including Australia at 20% or Canada with 19% (United Nations, 2006). Although immigrants are a larger share of the U.S. population than in most of Europe, some countries have a share of immigrants approaching that of the United States, including France (11%) and the United Kingdom (9%).

This report begins by presenting the baseline projection for the total population from 2005 to 2050. The next sections go into detail about the projected estimates for key segments of the population, including the foreign born, Hispanics, blacks, Asians, non-Hispanic whites, working-age adults, children and the elderly. This report then examines how these changes will affect the size of the potential workforce relative to the number of elderly and young people. A final section presents the results of two alternative projections. This overview concludes with a summary of major projections.

Racial and Ethnic Groups

In the Center’s projections, each person is included in only one race or Hispanic category. These projections assume that definitions of race and ethnic categories will remain fixed and that self-identification does not change over time. In reality, the growing numbers of births to parents of different racial and ethnic groups, as well as changing social norms about racial and ethnic self-identification, are serving to blur the boundaries of racial/ethnic categories. Consequently, the future sizes of race/ethnic groups could be higher or lower than the projection values even if the underlying demographic assumptions about the future prove to be correct.

Key Projections

Population and Immigration

• Between 2005 and 2050, the nation’s population will increase to 438 million from 296 million, a rise of 142 million people that represents growth of 48%.
• Immigrants who arrive after 2005, and their U.S.-born descendants, account for 82% of the projected national population increase during the 2005–2050 period.
• Of the 117 additional people attributable to the effect of new immigration, 67 million will be the immigrants themselves and 50 million will be their U.S.-born children and grandchildren
• The nation’s foreign-born population, 36 million in 2005, is projected to rise to 81 million in 2050, growth of 129%.
• In 2050, nearly one in five Americans (19%) will be an immigrant, compared with one in eight now (12% in 2005).
• The foreign-born share of the nation’s population will exceed historic highs sometime between 2020 and 2025, when it reaches 15%. The historic peak share was 14.7% in 1910 and 14.8% in 1890.
• Births in the United States will play a growing role in Hispanic and Asian population growth, so a diminishing proportion of both groups will be foreign-born.

Racial and Ethnic Groups

• The Hispanic population, 42 million in 2005, will rise to 128 million in 2050, tripling in size. Latinos will be 29% of the population, compared with 14% in 2005. (Figure 6) Latinos will account for 60% of the nation’s population growth from 2005 to 2050.
• The black population, 38 million in 2005, will grow to 59 million in 2050, a rise of 56%. In 2050, the nation’s population will be 13.4% black, compared with 12.8% in 2005.
• The Asian population, 14 million in 2005, will grow to 41 million in 2050, nearly tripling in size. In 2050, the nation’s population will be 9% Asian, compared with 5% in 2005. Most Asians in the United States were foreign born in 2005 (58%), but by 2050, fewer than half (47%) will be.
• The white, non-Hispanic population, 199 million in 2005, will grow to 207 million in 2050, a 4% increase. In 2050, 47% of the U.S. population will be non-Hispanic white, compared with 67% in 2005.

Age Groups

• The working-age population—adults ages 18 to 64—will reach 255 million in 2050, up from 186 million in 2005. This segment will grow more slowly over the projection period (37%) than the overall population. Future immigrants and their descendants will account for all growth in this group.
• Among working-age adults, the foreign-born share, 15% in 2005, will rise to 23% in 2050. The Hispanic share, 14% in 2005, will increase to 31% in 2050. The non-Hispanic white share, 68% in 2005, will decline to 45% in 2050.
• The nation’s population of children ages 17 and younger will rise to 102 million in 2050, up from 73 million in 2005. The child population will grow more slowly in future decades (39%) than will the overall population. Future immigrants and their descendants will account for all growth in this population segment.
• Among children, the share who are immigrants or who have an immigrant parent will rise to 34% in 2050 from 23% in 2005. The share of children who are Hispanic, 20% in 2005, will rise to 35% in 2050. Non-Hispanic whites, who make up 59% of today’s children, will be 40% of children in 2050.
• The nation’s elderly population—people ages 65 and older—will grow to 81 million in 2050, up from 37 million in 2005. This group will grow more rapidly than the overall population, so its share will increase to 19% in 2050, from 12% in 2005. (Figure 7) Immigration will account for only a small part of that growth.
• The dependency ratio—the number of people of working age, compared with the number of young and elderly—will rise sharply, mainly because of growth in the elderly population. There were 59 children and elderly people per 100 adults of working age in 2005. That will rise to 72 dependents per 100 adults of working age in 2050. (Figure 8)

Alternative Projection Scenarios

• Under a lower-immigration scenario, the total population would rise to 384 million, the foreign-born share would stabilize at 13% and the Hispanic share would go up to 26% in 2050.
• Under a higher-immigration scenario, the total population would rise to 496 million, the foreign-born share would rise to 23% and the Hispanic share would go up to 32% in 2050.
• Under a lower- or higher-immigration scenario, the dependency ratio would range from 75 dependents per 100 people of working age to 69 dependents per 100 people of working age. Both of these ratios are well above the current value of 59 dependents per 100 people of working age.

A Note on Terminology

The terms “Hispanic” and “Latino” are used interchangeably in this report.

The terms “whites” “blacks” and “Asians” are used to refer to the non-Hispanic components of each population.

“Children” refers to people ages 17 and younger.

“Working-age” refers to people ages 18 64.

“Elderly” refers to people ages 65 and older.

“Foreign-born” refers to an individual who is not a U.S. citizen at birth or, in other words, who is born outside the U.S., Puerto Rico or other U.S. territories and whose parents are not U.S. citizens. The terms “foreign-born” and “immigrant” are used interchangeably.

The terms “unauthorized immigrants,” “undocumented immigrants” and “illegal immigrants” are used interchangeably.

This report uses the following definitions of the first, second and third-and-higher generations:

• First: Foreign-born or immigrant.
• Second: U.S. native (born in the United States or territories), with at least one first-generation parent.
• Third-and-higher: U.S. native (born in the United States or territories), with both parents native-born.

Americans are increasingly concerned about immigration. A growing number believe that immigrants are a burden to the country, taking jobs and housing and creating strains on the health care system. Many people also worry about the cultural impact of the expanding number of newcomers in the U.S.

Yet the public remains largely divided in its views of the overall effect of immigration. Roughly as many believe that newcomers to the U.S. strengthen American society as say they threaten traditional American values, and over the longer term, positive views of Latin American immigrants, in particular, have improved dramatically.

Reflecting this ambivalence, the public is split over many of the policy proposals aimed at dealing with the estimated 11.5 million-12 million unauthorized migrants in the U.S. Overall, 53% say people who are in the U.S. illegally should be required to go home, while 40% say they should be granted some kind of legal status that allows them to stay here.

But nearly half of those who believe illegal immigrants should be required to leave nonetheless say that some could stay under a temporary work program. Overall, the public divides about evenly among three main approaches for dealing with people who are in this country illegally: 32% think it should be possible for them to stay permanently; 32% believe some should be allowed to stay under a temporary worker program under the condition that they leave eventually; and 27% think that all illegal immigrants should be required to go home.

There is also a division of opinion over how to stem the flow of illegal immigrants across the Mexican border. When asked to choose among three options, roughly half of Americans (49%) say increasing the penalties for employers who hire illegal immigrants would be most effective in reducing illegal cross-border immigration, while a third prefer boosting the number of border patrol agents. Just 9% of the public says the construction of more fences along the Mexican border would be most effective.

In general, however, the issue of immigration is not a top-tier problem for most Americans. Just 4% volunteer it as the most important problem facing the country, far fewer than the number mentioning the war in Iraq, dissatisfaction with the government, terrorism, and several other issues.

Nor does immigration loom particularly large as a local community issue. The new survey by the Pew Research Center for the People & the Press and the Pew Hispanic Center, conducted Feb. 8-March 7 among 2,000 adults nationally, includes separate surveys of an additional 800 adults in each of five metropolitan areas that have experienced differing rates of immigration in recent years: Phoenix, Las Vegas, Chicago, Raleigh-Durham and Washington DC.

Immigration emerges as a dominant local concern only in Phoenix, near a major entry point for illegal immigrants, where 55% say it is a very big problem. In the four other metropolitan areas, traffic congestion rates as a bigger problem than immigration.

The Bases of Ambivalence

The survey finds a number of opinions about immigrants that may well contribute to ambivalent attitudes toward immigration, especially in areas where immigrants are most numerous. First, attitudes toward both Latin American and Asian immigrants are more positive now than in the 1990s, even as concern over the problems associated with immigration has increased. Both groups are overwhelmingly seen as very hard working and having strong family values. Impressions of Latin American immigrants, in particular, have grown much more positive, with 80% describing them as very hard working compared with 63% nearly a decade ago.

Moreover, native-born Americans who live in areas with the highest concentration of immigrants hold more positive opinions of them. Analysis of the survey indicates that their more favorable views do not merely reflect their demographics or political composition, but suggests that exposure to and experience with immigrants results in a better impression of them. However, Americans living in areas with more immigrants rank immigration as a bigger community problem.

And while there is concern about the impact of immigration on the availability of jobs, nearly two-thirds (65%) say that immigrants coming to the country mostly take jobs that Americans do not want, rather than take jobs away from Americans. In this regard, the recent influx of immigrants into such metropolitan areas as Phoenix, Las Vegas and Raleigh-Durham has not undermined the generally positive perceptions residents have of the local job market.

Yet at the same time, a sizable minority (16%) says they or a family member have lost a job to an immigrant worker. And the perception of being passed over – more common among those with less education and lower incomes – is strongly associated with negative views of immigrants and high levels of support for strong measures to deal with the problem. For example, 75% of those who say they or a family member has lost a job to an immigrant view them as a burden compared with 47% of those who do not think this has happened.

Policy Solutions

The public’s divisions over illegal immigration are mirrored in views of legal immigration; 40% say the current level should be decreased, but almost the same number (37%) believe it should be kept at its present level, while 17% prefer to see it increased.

But it is illegal immigration, far more than legal immigration, that stirs public anxiety. Six-in-ten say illegal immigration represents a bigger problem than legal immigration. Just 4% say the opposite – that legal immigration is a bigger problem – though nearly a quarter (22%) says both forms of immigration are equally problematic.

Besides economic concerns, many express worries that illegal immigrants contribute to crime and increase the danger of terrorism. Yet fewer see tougher border controls, relative to employer sanctions, as the most effective way to reduce illegal immigration along the Mexican border. Even those who are most worried about the threat of terrorism associated with illegal immigration favor employer fines over border fences and more agents.

In line with these attitudes, two-thirds of the public favors the creation of a new government database for all of those eligible to work – citizens and legal immigrants alike – and a requirement that employers check this database before hiring new workers. Even more Americans support a de facto national identification card – either a Social Security card or new form of driver’s license – that job applicants would be required to show before obtaining a job.

Like policymakers, the public is conflicted about what to do with immigrants who are here illegally. Beyond questions of their legal status, Americans express very different opinions about providing government services for such people – and their children. By a wide margin (67%- 29%), Americans believe that illegal immigrants should be ineligible for social services provided by state and local governments. Yet by an equally lopsided margin (71%- 26%), most feel that the children of illegal immigrants should be permitted to attend public schools.

The Proximity Factor

The survey finds a complex relationship between exposure to immigrants and opinions about them and the immigration problem, more generally. People who live in areas that have high concentrations of immigrants are less likely to see them as a burden to society and a threat to traditional American customs and values. However, they are more apt than others to see immigration as an important problem for their local community.

In sharp contrast, native-born Americans who live in areas with few immigrants understandably are less inclined to see immigration as a local problem. However, many more of those in areas with relatively low concentrations of foreign-born people see immigrants as a burden to the nation and as a threat to American customs. People living in areas with few immigrants have a considerably more negative opinion of Hispanics and a slightly more negative view of Asians.

In general, the survey shows broad public recognition of the increasing level of immigration in recent years. Significantly more Americans than in the 1990s think that there are “many” recent immigrants living in their communities (35% currently vs. 17% in 1997). In each of the metropolitan areas surveyed separately, with the exception of Chicago, nearly half say there are many recent immigrants in their area.

Similarly, as many as 49% nationwide say they often come in contact with people who speak little or no English, up from 28% in 1997. This experience is very common in Las Vegas and Phoenix – 68% of Las Vegas residents and 66% of Phoenix residents say they often encounter people who speak little or no English. Most Americans who come in contact with people with little English say it does not bother them (61%), compared with 38% who say that it does. The balance of opinion is similar in the five metropolitan area surveys.

Politics in Washington and at Home

The American public is not particularly confident in its political leadership to deal with immigration. President Bush and the Republicans get especially anemic grades. Only 42% have a lot or some confidence in President Bush to do the right thing with regard to the issue. The Republican Party gets a similar rating (45%).

The Democratic Party achieves an only somewhat better evaluation (53%) as do governors (54%) and local leaders (56%). Residents of the five metropolitan areas surveyed separately evaluate their political leadership on the immigration issue about the same way citizens do nationwide. The exception is Governor Janet Napolitano of Arizona and local officials in Phoenix who achieve better ratings than other governors and local leaders.

Hispanics are more critical of all political leaders than are other citizens, but especially with respect to the Republican Party. However, they give President Bush a somewhat better grade than they do the GOP (41% vs. 33%).

For the most part, partisanship has only a modest impact on attitudes toward the severity of the problems associated with immigration and possible solutions. On basic attitudes as to how to reduce illegal immigration from Mexico, roughly half of Republicans, Democrats and independents prefer tougher employer sanctions; only about one-in-ten in each group thinks the construction of more border fences would be the most effective measure. About the same number of Republicans and Democrats also say illegal immigrants in the U.S. must go home. However, it is noteworthy that while Republicans express somewhat more concern about immigration overall, a plurality favors a temporary worker program for immigrants, a position President Bush has championed.

Opinions Marked by Many Divisions

Concerns about immigration, and views of what to do about it, divide the public in many different ways. Significant disagreement exists between college graduates and those who did not attend or complete college, between people who are struggling financially and those who are doing well, between liberals and conservatives, and along ethnic lines. While African Americans differ little from whites in their views about most of these issues, Latinos hold consistently more favorable views of immigrants and the impact of immigration on American society.

As a result, even when Republicans and Democrats do not differ overall, there are often deep divides within the political parties along ideological and socioeconomic lines. Generally financially struggling and less educated people hold more negative views of immigrants and favor more strict policies than do the financially secure and college graduates, and this is the case within both party coalitions.

For example, Republicans who rate their financial situation as “only fair” or “poor” are 20 points more likely than those who say they are in “excellent” or “good” shape to say immigrants are a burden on the country because they take jobs, housing and health care, and the gap between secure and insecure Democrats is comparably large. Within each party, education also plays a major factor – Democrats without a college degree are more than twice as likely to want to see legal immigration decreased compared with those who have a four-year degree.

The Metropolitan Area Surveys

In addition, the survey looked at five metropolitan areas that have experienced a significant increase in the foreign-born population. While respondents in Phoenix, Chicago, Las Vegas, Raleigh-Durham and Washington DC had similar views on some aspects of immigration and immigrants, there were also significant differences.

• Phoenix is the only metropolitan area where immigration is cited as the most important local problem.
• In Las Vegas, a majority says that immigrants from Latin America keep to themselves and do not try to fit in, the highest among the metro areas and significantly higher than the national result.
• Chicago, a historically diverse city, has seen recent population gains primarily from Hispanics. Residents of the area are generally more tolerant of immigrants and less inclined to support punitive measures for illegal immigrants.
• In Raleigh-Durham, a sizable majority believes that recent immigrants do not pay their share of taxes.
• Washington DC has a generally more welcoming view of immigrants compared with the other metropolitan areas.

Roadmap to the Report

The report that follows provides detailed analysis and discussion of findings from the national poll and the five metropolitan areas surveys. It begins with a description of how immigration ranks as a problem nationally and in the respondents’ communities. This section also addresses the distinctions the public makes between legal and illegal immigration. Next is a review of concerns about immigration and immigrants, including concerns about immigration’s impact on America’s culture and economy.

The third section of the report turns to the broad range of policy proposals being considered to deal with the issue of immigration. In addition to reviewing the shape of public opinion on each one, divisions in the public and within the parties are explored in detail.

Public perceptions about immigrants are examined in the report’s fourth section. In particular, trends in views of immigrants from Asian and Latin American nations are tracked, along with views about the willingness of recent immigrants to assimilate. Public perceptions about the size of the legal and illegal immigrant populations are described.

In the final section of the report, survey results from each of the five metropolitan areas are summarized, with notable differences highlighted among the cities and between each region and the nation as a whole.

The survey’s questionnaire and results for the nation and the five metropolitan areas are at the back of the report.

• The Hispanic population is growing in most metropolitan areas, but the rate and location of increase varies widely. Four distinct patterns of growth can be discerned. Established Latino metros such as New York, Los Angeles, Miami, and Chicago posted the largest absolute increases in Latinos between 1980 and 2000. However, new Latino destinations like Atlanta and Orlando charted the fastest growth rates, despite their historically smaller Hispanic bases. Metros with relatively larger Latino bases, such as Houston, Phoenix and San Diego, meanwhile, became fast-growing Latino hubs during the past 20 years, with population growth averaging 235 percent. Small Latino places, such as Baton Rouge, posted much lower absolute and relative growth than the other locales.

• Fifty-four percent of all U.S. Latinos now reside in the suburbs; the Latino suburban population grew 71 percent in the 1990s. In 1990 the central-city and suburban Hispanic populations in the 100 largest metros were nearly identical, but during the next decade suburban growth so outpaced central-city growth that by 2000 the suburban Hispanic population exceeded the central-city population by 18 percent. New Latino destinations saw the fastest growth of Latino suburbanites.

• Hispanic men outnumber Hispanic women by 17 percent in new Latino destination metros where the Latino population grew fastest. By contrast, in slower-growing metros with large and well-established Latino communities, more Hispanics live in family households and gender ratios are more balanced.