February 11, 2008

U.S. Population Projections: 2005-2050

II. Population Projections

Between 2005 and 2050, the nation’s population is projected to rise by 142 million, an increase of 48%. During those 45 years, it will expand from 296 million to 438 million. By contrast, the U.S. population rose by 116 million people between 1960 and 2005, which was a 64% increase.

Nearly all of the increase from 2005 to 2050 will be due to new immigrants and their U.S.-born descendants. They will account for 82% of the nation’s population growth, or 117 million additional people by 2050. Of those new residents, 67 million will be the immigrants themselves, 47 million will be their U.S.-born children and 3 million will be their U.S.-born grandchildren. That means new immigrants themselves will account for 47% of population growth during the projections period.

Only 25 million of the growth over the 2005–2050 period, or 18%, can be attributed to the resident population in 2005 and its descendants.

The Center’s population totals, calculated in five-year increments, were based on underlying demographic assumptions about births, deaths and immigration. They combine to yield an average annual growth rate of 0.9% over the 45-year horizon of the projections.

The nation’s annualized growth rate has been somewhat higher than that in recent years, with upturns in the 1960s that resulted from high fertility and in the 1990s that mainly reflected elevated immigration levels. The Center’s projections show somewhat less variation in future growth rates, because of underlying assumptions that immigration levels will rise slowly and that fertility will be stable. Under this scenario, the nation’s population would double in 80 years. (Figure 9)

Foreign-born Population

The nation’s foreign-born population, 36 million in 2005, is projected to rise to 81 million in 2050. The 129% rise during the 2005–2050 period is a sharper increase than for the population overall. In 2050, nearly one in five Americans (19%) will be an immigrant, compared with one in eight now. (Figure 10)

The number of foreign-born residents in the United States already is at a record number but is a somewhat smaller share of the population than a century ago. The foreign-born population was 12% of the total in 2005.3 At its previous peak, the percentage of foreign-born residents fluctuated from 13% to 15% for 60 years between 1860 and 1920 (Gibson and Jung, 2006 and Figure 2).

Pew Research Center projections indicate that the proportion of immigrants will exceed historical highs (14.7% in 1910 and 14.8% in 1890) sometime between 2020 and 2025, when 15% of the population will be foreign born.

Because most people immigrate as working-age adults, the foreign-born population is more heavily concentrated in the 18–64 age group than is the native population. In this country, about four out of five immigrants (81%) are ages 18–64. That share will decline slowly as the foreign-born population ages, reaching 73% in 2050. By contrast, only 60% of native-born residents were ages 18–64 in 2005, a share that will decrease to 55% by 2050. Thus, the foreign-born share of the working-age population will grow to 23% in 2050 from 15% in 2005.

Immigrants’ children and grandchildren born in the United States will account for all growth in the population ages 17 and younger. Most children of immigrants are born in the United States (about four of five) and therefore are U.S. natives. For this reason, the number of foreign-born children will remain relatively low throughout the projection period.

Very few people immigrate at older ages, so new immigration will have little immediate impact on the size of the elderly population. However, as younger immigrants age into the elderly group, they will make up a growing share of that population. In 2050, 16 million of the projected 81 million elderly will be foreign born, about 20%; in 2005, only 10% of the 37 million elderly were foreign born. (Figure 11)

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 2005.4 As the fastest growing major race or ethnic group, the Hispanic population will account for 60% of the nation’s growth during the 2005–20period. (Figure 12)

New immigrants and their descendants account for most of the projected Latino growth (74%), but the growth is mainly due to births in the United States. However, the Latino population is relatively young and has a higher than average fertility rate, so its growth would continue to outpace that of other groups even without new immigration. Growth of the current Hispanic population will add 22 million new U.S. residents by 2050, a 52% increase.

Although the Hispanic population will grow more quickly than the total number of U.S. residents, its growth rate will moderate—from 3.1% per year for 2005–10 to 2.0% per year for 2045–50. The reduced growth rate will be due, in part, to the increasing proportion of the Latino population born in the United States, because U.S.-born Hispanics have lower fertility rates than do first-generation Hispanics.

The Hispanic population already is mainly U.S. born (60%), and that proportion is projected to rise (to 67% in 2050) because of changes in the sources of growth. In a reversal of past trends, the number of births to Latino women will grow more rapidly than the number of new Latino immigrants. For a period of three decades, beginning in the early 1970s, new Hispanic immigrants had considerably outnumbered births to Latino mothers. As a result, the percent foreign-born among Hispanics increased from only 14% in 1960 to 40% in 2005.

However, this pattern of Latino births and immigration shifted during the current decade: From 2000 to 2005, there were more Hispanic births than new immigrants. The Center’s projections are that Hispanic births will grow much more rapidly than Hispanic immigration, so that by 2045–50 there will be almost twice as many Latino births as new Latino immigrants.

Accompanying this change will be a substantial shift in the generational composition of the Hispanic population. By 2050, the foreign-born share will drop to 33%. The second generation, which represented 23% of Hispanics in 1960 and has grown to 28% in 2005, will continue to increase. In 2050, 34% of Hispanics will be U.S.-born children with at least one immigrant parent.

The third-and-higher generations had accounted for a majority of Hispanics in 1960 (63%), but this had dropped to 32% in 2005 due to the large influx of immigrants. But as the share of births to U.S.-born Latinos increases, the third-and-higher generations will continue to grow. By 2050, all three generational groups of Hispanics will be roughly the same size. (Figure 13)

Non-Hispanic Race Groups

From 2005 to 2050, the black population will grow by about 56%, but as a share of the nation’s population it will be stable.5 By 2050, the nation’s population will 13% black, about the same share as in 2005. In 2000the Hispanic and black populations were nearly the same size; by 2005, the number of Hispanics (42 million) exceeded the size of the black population (38 million). The gap will continue to grow because of sustained Latino immigration. In 2050, there will be more than twice as many Hispanics as blacks (128 million compared with 59 million). (Figure 14)

The Asian population will grow almost as fast as the Hispanic population in percentage terms, almost tripling from 14 million in 2005 to 41 million in 2050. (Figure 15) In 1960, Asians represented a minuscule 0.6% of the total population. By 2005, the share of Asians had grown to 5%; by 2050, it will be 9%. Nearly all of the future growth in the Asian population (94%) will be due to immigrants arriving after 2005 and their descendants. But arrivals of new immigrants will play a declining role in Asian population change, and births in the United States to immigrants and their descendants will play a growing role. In 2005, most Asians in the United States (58%) were foreign born; by 2050, fewer than half (47%) will be foreign born. (Figure 16)

The white non-Hispanic population will grow much more slowly than any of the other groups because of low fertility rates and relatively low immigration. It is projected to rise by only 8 million, or 4%, over the 45-year period, with little change after 2025. As a result, the non-Hispanic white population’s share of the total will continue the decrease that has occurred since 1960, when 85% of the population was white, non-Hispanic. By 2005, the share had dropped to 67% and it will decrease steadily to 51% in 2040 and 47% in 2050. (Figure 17)

Age Groups

Working-age Adults

The working-age population—adults ages 18 to 64—will reach 255 million in 2050, up from 186 million in 2005. That is a 37% increase, a rate somewhat lower than for the population as a whole. At that time, working-age adults will be 58% of the population, down from 63% in 2005. (Figure 18)

Future immigrants and their descendants born in the United States account for all growth in the working-age population over this period, adding 76 million people to the 2050 size of this group. Absent new immigration, there would be a decline of 7 million people in this group.

Because immigration plays such a prominent role in future growth of the working-age population, the share of foreign-born residents in this segment will rise to 23% in 2050, compared with 15% in 2005. The Hispanic share of working-age adults, 14% in 2005, will more than double, to 31% in 2050. The non-Hispanic white share, 68% in 2005, will decline to 45% in 2050. (Figure 19)

It is important to note that not all working-age adults are in the labor force. Currently, more than three-quarters in the 18-to-64-age group are, but that could change depending on many economic and demographic factors. For example, a greater share of workers ages 50 and older may choose to stay in the labor force. Also, foreign-born Hispanic women now are much less likely to be in the work force than either U.S.-born Hispanic women or other U.S.-born women. If education levels of foreign-born Hispanic women continue to increase and their fertility continues to fall, their low labor force participation rates could increase, as has happened with other groups.


The nation’s child population will rise to 102 million in 2050, from 73 million in 2005. The 39% increase is somewhat slower than for the population as a whole. Children will make up 23% of the population in 2050, compared with 25% in 2005.

Even though the projected increase in the child population is relatively modest, it stands in marked contrast to the much slower growth during the previous 45-year period (1960–2005). In 1960, there were 65 million children ages 17 and younger, making up more than a third (36%) of the population. That number grew by only 14%, to 73 million (or 25% of the population), in 2005. The slow growth during those years reflects the aging of the post-World War II baby boom, a period of high birthrates that lasted from 1946 to 1964. The baby-boom babies are now middle-aged adults, and the first of them will turn 65 years old in 2011.

All of the projected increase in the population ages 17 and younger through 2050 will be due to the arrival of new immigrants and the children born to them in the United States. By 2050, 36 million children will be descendants of immigrants who arrived after 2005. Absent new immigration, there would be a decline of 8 million people in this age group.

Immigrants do not generally arrive as children, nor do immigrant families tend to bring large numbers of children with them from other countries. Instead, the pattern is that young adults immigrate and have children after they arrive. Consequently, the vast majority (94%) of these 102 million young people in 2050 will be born in the United States.

Because immigrants on average have more children than native-born U.S. residents, the share of U.S. children who are children of immigrants—the first and second generations combined—will increase. In 2050, one child in three (34%) will be an immigrant or the son or daughter of an immigrant, compared with almost one in four (23%) in 2005.

The share of children who are Hispanic will rise markedly, from 20% in 2005 to 35% in 2050. Non-Hispanic whites, who made up 59% of children in 2005, are projected to be 40% of children in 2050. (Figure 20)


The nation’s elderly population—people ages 65 and older—is projected grow to 81 million in 2050, from 37 million in 2005. This group will grow more sharply than other segments because of the aging of the generation born during the post-World War II baby boom. The last of the baby boomers turns 65 in 2029. Among all age groups, projections about the size of the elderly population are least uncertain because the vast majority already are part of the U.S. population.

The elderly have accounted for 12% to 13% of the population since 1990; this range is higher than at any time in the nation’s history. With each successive decade, the percentages will increase so that by 2050 the elderly will represent 19% of the population. By way of comparison, in Florida, the state with the highest proportion of elderly residents, 17% are in that age group today (Census Bureau 2006).

The proportion of Hispanics in the elderly population will almost triple, from 6% to 17%, between 2005 and 2050 but will be substantially less than the proportion of Hispanics in younger age groups.

The racial/ethnic composition of the elderly population will change as the overall population does, but the changes lag behind those of younger age groups. Thus, non-Hispanic whites will remain a majority of the elderly, but their share will drop markedly to 63% in 2050, compared with 82% in 2005. The proportions of other race groups in the elderly population will increase over the projections horizon. The share of blacks will grow from 8% in 2005 to 12% in 2050; the share of Asians will grow from 3% to 8%. (Figure 21)

Dependency Ratio

The dependency ratio is a demographic and economic indicator that compares the size of non-working-age groups—children and the elderly—with that of the working-age population.6 A higher number of elderly or children relative to the number of workers translates into higher costs per worker to pay for all government programs, including those targeted at the young and old such as schools and Social Security.

In 2005, there were 59 elderly people and children for every 100 Americans of working ages. In 2050, assuming current trends continue, that dependency ratio will rise to 72. The main reason for the increase is that the elderly population will grow more rapidly over the next four decades than the working-age population. Most of the increase in the number of elderly will occur by 2030 as the baby-boom generation enters the retirement years. The ratio of children to working-age people, on the other hand, will change little.

The elderly dependency ratio was 20 people ages 65 and over for every 100 people ages 18 to 64 in 2005. That ratio has risen slightly since 1960 (from 16 per 100) and will increase rapidly (to 32 per 100) until 2030, when the youngest of the baby boomers turns 66. The elderly dependency ratio will level off at 32 elderly per 100 people of working age through 2050.

A very different pattern emerges with regard to the child dependency ratio. It is projected to be stable in coming decades, at about 40 children for every 100 people of working age. The major driver of this ratio is not the level of immigration but the average number of children per woman, which is projected to change little in the future.

To understand how fertility affects the child dependency ratio, consider that in the late 1950s, when women had an average 3.7 children, there were 65 children per 100 working-age Americans. By the mid 1970s fertility had dropped, and it has remained at about two children per woman for the past two decades.7 As a consequence the child dependency ratio dropped to 42 childper 100 people of working age by 1990.

The current dependency ratio is lower than it has been for decades. It was 81 in 1960, during the post-World War II baby boom, and has been declining since then. (Figure 22)

Alternative Scenarios

This section examines how the population would change if immigration levels were lower or higher than current trends. The baseline projection for the overall population is 438 million in 2050. That could range from 384 million under a lower-immigration scenario, to 496 million if higher immigration levels prevailed. (Figure 23)

This baseline projection assumes annual immigration, authorized and undocumented, of about 1.4 million people a year, rising to 2.1 million by 2050; annual immigration would average 1.7 million per year over the entire 2005–2050 period. An alternative lower-immigration scenario assumes that levels would average about half that amount, or 900,000 per year; a higher-immigration scenario assumes that levels would average about 50% higher, or 2.6 million a year over the 45-year projection period.

These three immigration scenarios use the same mix of birth and death rates by age, race and ethnicity, but the projected numbers of births and deaths is a function of the projected population size.

Foreign Born. The nation’s foreign-born population, which numbered 36 million in 2005, would grow to 49 million in 2050 even under a lower-immigration scenario. The proportion foreign born in the total population would stabilize at roughly the current level (12% to 13%).

Under a higher-immigration scenario, the foreign-born population would rise to 115 million by 2050, compared with 81 million under the baseline scenario. Instead of making up nearly one in five Americans (19%), as the baseline projection envisions, immigrants would be nearly one in four (23%). Under the baseline scenario, the foreign-born proportion of the population would exceed the historic high by 2025; under the higher-immigration scenario, that would happen by 2015, when it would reach 15%.

Hispanic. The nation’s Hispanic population, numbering 42 million and accounting for 14% of U.S. residents in 2005, would grow sharply under all immigration scenarios. Under the baseline projection, it would more than triple by 2050, reaching 128 million, or 29% of the total population.

Even under a lower-immigration scenario, the Hispanic population would more than double, reaching 98 million in 2050, or 26% of all U.S. residents. Under the higher-immigration alternative, the Latino population would increase to 159 million in 2050, or 32% of total residents.

Working Ages. Turning to the impact of immigration on different age groups, the scenarios diverge markedly after 2030 for working-age adults. (Figure 24) The baseline projection is that the number of adults ages 18 to 64 will rise from 186 million in 2005 to 255 million in 2050. The working-age population, now 63% of the total, would decline to 58%.

Under a lower-immigration scenario, the size of the potential labor force would grow to 219 million in 2050. Under a higher-immigration scenario, the population ages 18 to 64 would rise to 293 million. The working-age share of the population will not be affected significantly by the assumed level of future immigration. This group would account for about the same share of the total population (57% to 59%) under all three projections.

Children. For the child population, the baseline projection shows that there will be 102 million Americans younger than 18 in 2050, making up 23% of the population. That compares with 73 million, and 25% of the population, in 2005.

Under a lower-immigration scenario, the number of children would rise to 85 million and the proportion would be 22%. Under a higher-immigration scenario, the number would rise to 120 million and the share would be 24%.

Elderly. The size of the elderly population, 37 million in 2005, would be the least affected by differing immigration scenarios because few people migrate at older ages and some older immigrants leave. The number of elderly, 81 million in 2050 under the baseline scenario, would be 80 million under the lower-immigration alternative and 83 million under the higher-immigration alternative.

Dependency Ratios. The overall dependency ratio, which was 59 youths or elderly for every 100 people of working age in 2005, will rise to 72 in 2050 under the baseline projection. It would be slightly higher, 75, under the lower-immigration scenario, and slightly lower, 69, under the higher-immigration scenario. (Figure 25)

The projected youth dependency ratio—that is, the number of children ages 17 and younger per 100 people of working age—would be relatively stable and change only slightly under differing immigration scenarios. It was 40 in 2005 and will stay in the range of 39 to 41 under all three scenarios.

The elderly dependency ratio is little affected by the assumed level of future immigration. As a result, the number of elderly people compared with the number of working-age people would rise substantially under all scenarios. The ratio would go up sharply until the youngest baby boomer reached 66 in 2030. In 2005, there were 20 elderly per 100 people of working age, which will rise to 32 under the baseline projection. Under the lower-immigration scenario, the ratio would reach 36 by 2050. Under the higher-immigration scenario, the ratio would be only slightly lower, 29 in 2050. (Figure 26)

  1. The foreign born population was 12.7% of the total in 2007, according to the Census Bureau’s Current Population Survey.
  2. The Latino population in 2007 was 47 million, based on the Census Bureau’s October Current Population Survey, and Hispanics were 15.5% of the total population.
  3. All race groups are non-Hispanics only.
  4. A true “dependency” ratio would compare workers with non-workers. Projections of this kind of ratio would require additional assumptions about future labor force participation levels.
  5. That rose to 2.1 in 2006, but it remains to be seen whether that increase will persist. The projections allow for that possibility.