Rising immigration levels, coupled with a declining rate of natural increase (births minus deaths), means that immigration accounts for a larger share of U.S. population growth now than in any decade since 1900-09:
The chart shows that in the 1960s, first-generation immigrants (annual new arrivals) accounted for 15.6 percent of the increase in U.S. population.By the 1990s, that figure rose to 35.8 percent.1
Immigration accounts for a larger share of population growth today than before the 1920s restrictions. U.S. population is about 3 times larger now than it was then. The negative consequences of population growth — global warming, urban sprawl, lower wages, higher taxes, deterioration of public infrastructure, school overcrowding, water shortages, catastrophic off-shore oil spills — are well documented. Yet unlike the 1920s, it is regarded as poor form to mention immigration control as a means to address the current crisis.
In reality, the impact of immigration on population growth is larger than the above figures suggest. These percentages understate immigration’s total impact because they ignore the children born to immigrant mothers in the U.S. This omission is significant.
To see why, consider the components of U.S. population growth (see table left).
The Census Bureau estimates that U.S. population grew by about 2.6 million in 2009. There were about 4.3 million births in the U.S. that year and 2.5 million deaths. This means that 1.8 million, or roughly two-thirds of total population growth, was from “natural increase,” and the rest was from immigration.
However, an estimated 950,000 births a year are to foreign-born mothers — legal and illegal.2 Almost all of these women arrived in the U.S. over the last three decades. These births, plus net immigration, accounted for nearly 70 percent of all population growth in 2009.
Trends portend an even larger role for immigration. As we report below, the Census Bureau’s most recent projections find that immigrants and their U.S.-born children will account for 79.5 percent of population growth between 2010 and 2050. Another large-scale study by Pew Research, estimates that fully 82 percent of the population increase we can expect from now to 2050 will be immigrant-based — immigrants and their children born here — as against an equivalent figure of 51 percent for the years 1960-2005; and that the bulk of that population increase will be Latino in origin.3
Why the disproportionate impact? It’s not because the number of new arrivals is expected to soar. Most of the population growth related to immigration stems from children born in the U.S. While native-born women are having fewer babies, fertility among their foreign-born counterparts has generally continued to increase. Equally important: the share of foreign-born women in their prime child-bearing years (15 to 44 years of age) is rising — it went from 53 percent to 56 percent between 1990 and 2002, while it fell for natives from 45 percent to 41 percent.4
These trends mean that the share of births to foreign-born mothers is rising faster than the foreign-born share of the U.S. population:
Over nearly four decades, the difference between immigrants’ share of total population and their share of births has increased significantly. In 1970 immigrants accounted for 4.7 percent of the total population and 6.1 percent of births — a 1.4 percent point difference. But by 2004, they were 12 percent of total population and 24 percent of total births — a 12 percent point difference.
The figures cited above make clear that births to immigrant mothers are at levels not seen in recent decades. But how do these decades stack up against the “Great Wave” early in the twentieth century? Many researchers feel that the current situation falls short of that epochal event.
In at least one important metric, they are right: in 1910 the foreign-born population reached 14.7 percent of the total population. That is still a record. In subsequent years, the number of new immigrants declined precipitously — first due to World War I, and a few years later from the restrictive legislation of the early 1920s. The most recent figures show immigrants account for about 13 percent of the U.S. population.
But an analysis of the 1910 Census also shows that the share of births to immigrants that year was lower than it is today. Using unpublished files from the 1910 Census, demographer Steven Camarota estimates that 21.9 percent of all births that year were to immigrant mothers. This is considerably below the 24 percent figure for 2004 reported by Pew Research.5
Conclusion: Births to immigrants account for a larger share of births today than during the “Great Wave.” The reason, of course, is that today’s native-born woman has far fewer children, on average, than her 1910 counterparts. (Fertility rates among non-Hispanic whites are so low that this segment of the population is expected to see their numbers decline by mid-century.)
At no time in American history have foreign-born mothers accounted for a larger share of the “natural increase” of U.S. population.
It is worth ruminating on the fact that the problems associated with population growth were a major reason for the post-1910 restrictions. But no such reduction seems in the cards today. Absent such a policy, births to immigrants will inevitably account for ever higher shares of all births.
Thus, in a very real sense, our demographic destiny is in the hands of people who were not born here. As a nation, we are headed into uncharted territory.
A moratorium and U.S. population growth
What is a moratorium? In its most extreme form, it is a total cessation of both legal and illegal immigration into the U.S. No one is seriously proposing this, and, indeed, implementing such an extreme measure would probably be impossible. But for research purposes, a zero-immigration scenario is useful.6 It provides an upper bound to the impact that a restrictionist policy can have on future population growth.
The Census Bureau examined the implications of a moratorium in a series of population projections released in late 2009. The Bureau projects U.S. population from 2010 to 2050 under several immigration scenarios, two of which are relevant to our study: a moratorium and current immigration policy.
Population projections for each scenario are shown in the chart below.
Under current U.S. immigration policy, the nation’s population, currently at 310 million, would grow to 439 million by mid-century, according to the Census Bureau projections.7 Under this scenario, annual net immigration (the difference between people entering and leaving the U.S.) would rise steadily from 1.3 million in 2010 to 2.0 million in 2050.
Under a moratorium, U.S. population would peak at 323.0 million in 2047, before descending slowly to 322.9 million, in 2050.
Population growth is a significant cause of most of the economic, fiscal, and environmental problems facing the United States. For this reason, it is worth summarizing the population implications of the two immigration scenarios:
A 40 year moratorium on new immigrants would reduce 2050’s population by 102 million, or 26.4 percent below the level that would have been reached under current immigration policy, according to the Census Bureau’s most recent projection. Instead of growing by 129 million, U.S. population would expand by just 26.4 million over that period.
In quantifying the environmental benefits of an immigration moratorium, these two numbers — 102.3 million fewer people; a 26.4 percent reduction in U.S. population by mid-century — are a useful point of departure.
The nation’s foreign population, which numbered 38 million in 2008, will grow to 81.3 million by mid-century, according to the Pew Research Center’s “Main Projection” immigration scenario. (This corresponds to our “Current Policy” projection, so we use this terminology below.)
Pew’s “Higher Immigration” scenario sees the foreign-born population rising to 115 million by 2050. Instead of making up nearly one in five Americans (18.6 percent) as the Main Projection envisions, immigrants would be nearly one in four (23 percent).8
Even under a “Lower Immigration” scenario, the nation’s foreign-born population would grow to 49 million in 2050, and the foreign-born share of U.S. population would stabilize at the current level (around 13 percent).
The Pew report is somewhat dated. Released in February 2008, it has been superseded by the Census population projections released in December 2009.
Unfortunately the Census, while projecting total U.S. population under different immigration scenarios, does not project the foreign-born population per se. Undeterred, we have constructed our own projections by applying Pew’s foreign-born population shares to Census’s newer U.S. population projections.
How would a moratorium impact the foreign-born population? Obviously, this population will decline as immigrant deaths will no longer be offset by new arrivals. For some strange reason, however, neither Pew nor the Census projects foreign-born population under a zero-immigration moratorium scenario.
We have filled this information gap. Using age- specific death rates for the Hispanic population, we have estimated the rate at which the foreign-born population will fall under a long-term moratorium. The following table shows the foreign-born population under two immigration scenarios: current policy and moratorium.
By mid-century, there will be 20.2 million foreign-born residents under a moratorium — fully 60 million less than under current immigration policy. Instead of doubling, the immigrant population will fall by nearly 50 percent under a 40 year moratorium.
In terms of the economic impact on native-born Americans, the foreign-born share of the population is more important than the head count. A larger share means a larger per capita transfer of income from natives to foreign-born beneficiaries. Similarly, the average wage loss suffered by native workers rises in tandem with the foreign-born share of the workforce — not the number of such workers per se.
There is good news to report: A 40-year moratorium will reduce the foreign-born share of U.S. population by more than half (see table, below left).
Under current policy, immigrants will account for nearly 19 percent of U.S. population in 2050. By contrast, a moratorium will reduce this by two-thirds — to 6.3 percent.
A tough, unprecedented policy? Hardly. The foreign-born share of U.S. population was 4.7 percent in 1970, and 6.2 percent as recently as 1980.
Bottom line: A 40-year moratorium merely reduces the immigrant share of U.S. population to levels reached in the decades immediately following the 1965 immigration act.
Impact on states
As things stand now, immigration will be the major driver of U.S. population growth to mid century. However, the degree to which immigration impacts population growth will vary among the states. While traditional immigration gateways such as New York and California are home to the largest immigrant populations, their immigrant populations are not growing as fast as in less traditional venues. From 1990 to 2000, for example, North Carolina’s immigrant population nearly tripled, while foreign-born populations in New York and California grew at less than the national average over that time.
Of the top 10 states (ranked on the percent increase in immigrant population growth in the 1990s), only one — Arizona — is a traditional immigrant destination (see first table, page 55).
More recent data — covering 2000 to 2008 — make it still clearer that immigrants are shunning traditional gateways in favor of the less crowded hinterland:
Eight of the top 10 immigrant destinations are in the South. None is a traditional gateway.
The Census Bureau does not project foreign-born population. It does, however, project total state populations. The most recent such report, released in April 2005, finds that three states — Florida, California, and Texas — will account for nearly one-half (48.5 percent) of U.S. population growth between 2010 and 2030. Florida’s population will grow the most, rising by 9.4 million. Now the fourth most populous state, Florida is expected to edge past New York, into third place by 2011. California and Texas would continue to rank first and second, respectively, in 2030.
The top five fastest-growing states between 2010 and 2030 will be Arizona (61.4 percent), Nevada (59.2 percent), Florida (59.2 percent), Texas (35.2 percent), and Utah (34.7 percent.) Only four states — Ohio, Iowa, North Dakota, and West Virginia — and the District of Columbia are expected to lose population over this period.
The Census projections are in 5-year intervals, and extend to 2030. They are based on rates of change estimated from historical data. They are not straight-line projections, however. Thus, New York’s population is projected to: rise by 0.53 percent over the five years 2010 to 2015; rise by 0.15 percent over the next five years (2015-20); fall 0.19 percent between 2020 and 2025; and fall by another 0.32 percent in the last five years of the projection period, 2025 to 2030.
Clearly, the Census Bureau uses a sophisticated, complex mathematical model to project what will happen if current state-specific trends in fertility, mortality, internal migration, and international immigration continue.
Moratorium vs. current policy
Our goal is to project future state populations under two immigration scenarios: current policy and a 40 year (2010 to 2050) moratorium on legal and illegal immigration to the U.S. Unfortunately, the Census Bureau does not do this analysis for individual states. Census implicitly uses only one immigration scenario — a continuation of current policy — in its state population projections. We must, therefore, develop our own moratorium population projections for the states.
We do this by making a heroic, but not implausible, assumption: the share of U.S. population residing in each state will be the same under both current policy and moratorium scenarios. Put differently, our methodology assumes that, in the long run, immigrants and natives respond to the same factors — jobs, economic growth, climate, etc. — when deciding which state to live in. To project a state’s population under a moratorium for, say, 2050, we simply multiply the Census Bureau’s moratorium projection for the U.S. population that year by the state’s share of U.S. population under current policy for that year.
Another problem: While the Census Bureau’s national population projections extend to 2050, their state population projections only go to 2030. We resolve this by extrapolating state populations to 2050 using the average annual percentage growth rates for 2025 to 2030 — the two future-most years of Census state population projections.
Population projections for the fifty states and the District of Columbia under current immigration policy and a 40-year moratorium are shown in the table on page 57.
Thirty-two states lose population under a moratorium, with losses ranging from a 34.1 percent decline in West Virginia to 1.8 percent in Arkansas. Five of these states would have lost population even under current immigration policy: West Virginia, North Dakota, Iowa, Wyoming, and Ohio. For the other 27 states a moratorium converts a population increase over the 2010 to 2050 period to a decline.
In three states — Florida, Texas, and California — a moratorium will reduce 2050 population by more than 10 million below levels that would have been reached under current immigration policy. At the other extreme, a moratorium will reduce 2050 population in the District of Columbia, Wyoming, and North Dakota by 100,000, 130,000, and 150,000 below current policy levels, respectively.
Keep in mind that these are projections, not forecasts. Population forecasts are predictions of future population levels. A good forecast of state population takes into account many different factors, including future birthrates, cost of living, whether the state’s economy will grow faster or slower than that of nearby states — as well as national immigration policy.
Our projections, on the other hand, assume that immigration policy is the only variable that changes over time. Everything else is “held constant.” Our goal is to compare state population growth under two scenarios: a continuation of current federal immigration policy and a moratorium (zero-immigration policy) — with everything else held constant.
Key states are highlighted below:
California’s population, less than 34 million in 2000, recently topped 38 million. By 2050, we project state population will increase by another 19 million, largely due to the new (post-2009) immigrants and their U.S.-born children. By contrast, under a 40-year zero-immigration moratorium, California’s population would grow by only 5.4 million, to 42 million.
Bottom line: a moratorium would reduce 2010-50 population growth by 13.4 million, or 71 percent, below the expansion projected under current immigration policy.
While immigration is and will remain a major driver of California’s future population growth, its role has declined. In 2008 the number of foreign-born Californians dropped 165,000 to 9.9 million. The reversal in the state was driven by several southern California counties with sharp declines, such as Los Angeles, with a slide of 3 percent, San Bernardino, down 3.6 percent, and Ventura, down 4.1 percent. Orange and Riverside counties showed smaller decreases.9
While the economy is a factor, the slowing of the increase in California’s foreign-born population began well before the latest recession. In the 1980s, for example, many immigrants targeted California because of their family and cultural ties to the state’s already established immigrant communities. But during the aerospace-led downturn of the early 1990s, immigrants began moving throughout the country, where they found employment more plentiful and housing more affordable.
Native-born Californians have joined the exodus. Their outflow — last seen during the economic and social struggles of the 1990s — started when it became too expensive for most people to buy homes in the state and accelerated in the Great Recession.
Though more births and rising international immigration helped boost California’s population a modest 1 percent per annum in recent years, the state continues to see a steady stream of outmigration — the movement to other states.10
Our projections indicate that California will be home to 13.0 percent of U.S. population in 2050, up from 12.3 percent in 2010. That is a far more modest increase than would have been projected during the state’s peak growth years.
Arizona recently overtook Nevada as the most rapidly growing state in the union, with an annual growth rate of 3.3 percent between 2000 and 2008. Combining the increase in the foreign-born population with estimated immigrant births suggests that immigration accounted for more than two-fifths (43.8 percent) of the state’s overall population increase during this period. Much of the rest came from outmigration from California. Retirees from the Upper Midwest, Great Lakes, and Northern Plains have also added significantly to Arizona’s population over the past two decades.
The Census Bureau estimates Arizona’s population in July 2008 had increased to 6,500,180 residents, i.e., an annual average increase of about 165,000 residents since 2000.
While the state’s population growth rate is expected to decline slightly, the Census projects Arizona’s population will grow at three times the national rate through 2030. If current immigration policies remain in place, we project state population will reach 10.8 million in 2030 and 17.4 million in 2050. As a share of U.S. population, this implies nearly a doubling — from 2.1 percent in 2010 to 4.0 percent in 2050.
A 40-year moratorium would cut Arizona’s 2050 population by 4.6 million below the level that would have been reached under current immigration policy:
Under a moratorium, state population is projected to grow by 6.4 million between 2010 and 2050, or 40 percent less than the 10.7 million rise expected under current policy. By comparison, a moratorium would lower total U.S. population growth by nearly 80 percent over that period.
Even with a moratorium, the state’s population is projected to double by mid-century. Immigration is relatively less important to Arizona’s population growth than elsewhere, reflecting its attractiveness to native-born retirees and ex-Californians.
Immigration-driven population growth is taking its toll on Florida, the seventh fastest growing state in the U.S. In the last ten years, over three million new residents settled in Florida — an increase larger than the entire population of the state in 1950. One-third of these new residents were immigrants. Adding the U.S.-born children of immigrant mothers to the increase in the foreign-born population suggests that immigration may account for more than seven-tenths (71.6 percent) of the state’s overall population increase since 2000.
In 1995 a report commissioned by then-Governor Lawton Chiles warned that “rapid population growth and sprawling development patterns are leading southern Florida down a path toward wall-to-wall suburbanization.” Governor Jeb Bush’s Growth Management Commission agreed that traffic congestion, crowded classrooms, water shortages, and pollution are serious and growing problems in the state. Yet Florida continues to add the third largest number of immigrants of any state.
Since 2000, the foreign-born population has increased by 34.9 percent compared to a 10.6 percent increase in the native-born population. The U.S. Census Bureau estimates that Florida’s resident population was 18,328,340 in July 2008, i.e., an annual average increase of about 282,645 residents since 2000.
If U.S. immigration policy remains unchanged, Florida’s population is projected to be 29.5 million in 2030 and 43.8 million in 2050. Under a 40-year moratorium, state population in 2050 would be 32.2 million — 11.6 million less than under current policy.
Only California and Texas will experience larger 2050 population reductions under a 40-year moratorium.
The U.S. Census Bureau estimates New York’s population was 19,490,297 in July 2008, having increased by about 61,910 residents per year since 2000. That is a rate of increase of about 0.3 percent per year — well below the national average.
New York State’s foreign-born population has grown by about 46,420 per year since 2000, which is three-fourths of the state’s annual average population increase over that period. Since 2000, the foreign-born population has increased by 10 percent compared to a 0.9 percent increase in the native-born population.
Moreover, immigrant mothers give birth to an estimated 106,000 babies per year. Adding the births to the increase in the foreign-born population suggests that immigration adds nearly 152,850 persons to the New York State’s population annually, i.e., more than double the state’s overall population increase.
The exodus of native New Yorkers is the largest factor in the state’s stagnant population trend. Outmigration rates are especially high among young adults (aged 20-34), who often leave the state for the economic opportunities elsewhere.11
State population would have declined without the population increase attributed to immigration. Not surprisingly, New York’s resident population is projected to decline substantially under a 40-year moratorium:
New York state population is projected to decline by 4.3 million, or 23 percent, under a moratorium. Under current immigration policy, population will remain virtually unchanged.
Once again it should be remembered that these are projections, not forecasts. By curtailing the influx of low-wage immigrants, a moratorium could slow or even reverse the outflow of native-born New Yorkers. There would be no need to seek greener pastures in other states. Over time, state population could actually be larger under a moratorium than under current immigration policy.
Michigan’s resident population was 10,003,422 in July 2008, having increased at an average rate of about 7,830 per year since 2000 according to the Census Bureau. Over that period, net international migration (more immigrants coming in than going out) added an average 18,935 residents per year — more than 2.4 times the state’s total increase. Adding in the estimated births to immigrant mothers suggests that immigration accounted for nearly 20,475 persons added to the state’s population annually.
As a share of total population, however, immigrants are far less prominent in Michigan than in the nation as a whole. The state’s foreign-born population was about 639,035 in July 2008. This means a foreign-born population share of 6.4 percent compared to 12.5 percent for the nation.
The most critical assumption for Michigan — and the one that is subject to the highest level of uncertainty — is the assumption about migration to and from other states. Michigan had high outmigration to other states in the late 1970s, extremely high outmigration in the early 1980s, and low outmigration in the 1990s.
Since 2000 — and especially during the Great Recession — several negative factors have re-appeared: a burgeoning gap in unemployment rates between Michigan and other states; the accelerated decline of the auto industry; outmigration of Baby Boomers; and rapid population decline in the city of Detroit.
Since 2000, Michigan’s foreign-born population has increased by 22 percent while its native-born population has declined by 0.5 percent. Clearly, Michigan natives are “restless” and many have moved to other states.
If current immigration policy remains in effect, Michigan will continue along its slow growth path. State population will increase by 3.1 percent from 2010 to 2050. By contrast, a moratorium will reduce state population from its current 10.0 million to 7.9 million in 2050.
Michigan’s 2050 population will be 2.9 million lower under a moratorium than under current immigration policy. Only twelve states are projected to experience larger population reductions with a zero-immigration policy.
The Census Bureau estimates North Carolina’s resident population was 9,222,414 in July 2008, having increased by an average 41,335 per year since 2000. That is a growth rate of about 1.8 percent per year.
Net immigration contributed about one-sixth (16.4 percent) of the state’s total population increase during this time. Adding in the estimated births to immigrant mothers suggests that immigration accounted for nearly one-third (32 percent) of the state’s overall population increase since 2000.
North Carolina is a relatively new immigration destination. The 2000 Census found that 62.4 percent of North Carolina’s foreign-born population had arrived in the state since 1990 — a much higher share than the national average (43.7 percent). Only 7.2 percent of the state’s population is foreign born, about half the national average. Yet, as revealed above, the state’s foreign-born population is growing faster than that of all but a handful of states.
If current immigration policies remain in place, immigration will account for an ever larger share of North Carolina’s population growth. A moratorium will make a profound difference:
State population is on course to hit 16.2 million in 2050, an increase of 72.3 percent from 2010. By contrast, 40 years of a zero-immigration policy would lower this to 11.9 million, a rise of 32.2 percent.
A moratorium will thus reduce state population by 4.3 million below the level that would have been reached under current policy in 2050. The only states to enjoy larger reductions from a moratorium are traditional immigrant gateways — California, Texas, Florida, New York, and Arizona.
District of Columbia
The District’s population peaked in 1950, when 802,178 persons resided in its 68.5 square miles. Over the next five decades, D.C. population fell by nearly one-third, to 572,069 in 2000. Since 2000, the city’s population has been growing in fits and spurts. The Census Bureau estimates D.C.’s population was 591,833 in July 2008, having increased by about 2,380 residents per year since 2000. That is a rate of increase of about 0.4 percent per year.
An estimated 74,420 D.C. residents — 12.6 percent of the population — were immigrants in July 2008. An analysis of data collected by the American Community Survey indicates an average annual increase of only 105 foreign-born residents per year — which is a small fraction (4.4 percent) of D.C.’s recent (post-2000) annual population growth. Adding in births to immigrant mothers suggests that immigration may account for nearly 2,080 persons added to D.C.’s population annually, i.e., more than seven-eighths (87.4 percent) of the District’s overall population increase.
Neither immigrants nor their U.S.-born children can offset the drag on D.C.’s population caused by the exodus of residents to other states and localities. Indeed, the District joins only a few states — North Dakota, West Virginia, Iowa, Ohio, and Wyoming — projected to lose population even under current immigration policy:
Under current immigration policy, D.C.’s population is projected to shrink by another 32 percent, to 362,500 in 2050. A moratorium would trigger a 48 percent decline between now and mid-century, to 266,700.
Again it must be stated that these are projections based on a continuation of historical relationships. As with New York, an immigration moratorium may reduce the economic displacement of U.S.-born D.C. residents by low-wage immigrants. This happy circumstance could slow or even reverse domestic outmigration that has decimated D.C.’s population since 1950. ■
1. Mark Nowak,
Immigration and U.S.
Population: An Environmental Perspective, Negative Population Growth, 1997.
2. Steven A. Camarota, 100 Million More: Projecting the Impact of Immigration on the U.S. Population, 2007 to 2060, CIS Backgrounder, August 2007.
3. Jeffrey S. Passel and D’Vera Cohn,
Population Projections: 2005-2050, Pew Research Center, February 2008.
4. Steven A. Camarota, Births to Immigrants in America, 1970 to 2001, CIS Backgrounder, July 2005. Page 6. (1970 to 1990); National Center for Health Statistics.
5. Gretchen Livingston and D’Vera Cohn, “The New Demography of American Motherhood,” Pew Research Center, May 6, 2010.
6. The zero immigration scenario pertains to both native and foreign-born populations. Specifically, it assumes that no foreigners enter the U.S. and no U.S. residents (foreign- or native-born) leave.
7. Census Bureau website.
8. Passel and Cohn, op. cit.
9. Don Lee and Alana Semuels , Survey data point to a dramatic shift in the nation and show the recession’s effect on foreign-born residents, LA Times, September 22, 2009 .
10. David Pierson, A State in the rearview
mirror: More are leaving California to live elsewhere in the U.S. than are
LA Times, December 18, 2008.
11. Empire Foundation for Policy Research, cited in http://www.fairus.org/site/PageServer?pagename=research_research05b8