As Immigrants Move In, Americans Move Up

  • Downloads
  • Related Content

Congress and President Obama may tackle the controversial issue of immigration reform as soon as the fall of 2009. If past congressional debates are any guide, one point of contention will be the impact of reform on the American underclass.

In 2006, and again in 2007, the U.S. Senate debated “comprehensive immigration reform” designed to curb illegal immigration by ramping up enforcement while providing expanded opportunities for legal immigration. Both bills would have legalized several million immigrants currently in the United States illegally and created a temporary visa program to allow more low‐​skilled workers to enter the country legally in future years.

One argument raised against expanded legal immigration has been that allowing more low‐​skilled foreign‐​born workers to enter the United States will swell the ranks of the underclass. The critics warn that by “importing poverty,” immigration reform would bring in its wake rising rates of poverty, higher government welfare expenditures, and a rise in crime. The argument resonates with many Americans concerned about the expanding size of government and a perceived breakdown in social order.1

As plausible as the argument sounds, it is not supported by the social and economic trends of the past 15 years. Even though the number of legal and illegal immigrants in the United States has risen strongly since the early 1990s, the size of the economic underclass has not. In fact, by several measures the number of Americans living on the bottom rungs of the economic ladder has been in a long‐​term decline, even as the number of immigrants continues to climb. Other indicators associated with the underclass, such as the crime rate, have also shown improvement. The inflow of low‐​skilled immigrants may even be playing a positive role in pushing nativeborn Americans up the skills and income ladder.

Measuring the Size and Composition of the Underclass

“Underclass” is not a precise term, but it is generally understood to mean those who live below or near the poverty line and who lack the education or jobs skills to join the middle class. If we define the underclass to be the number of people in the United States living below the poverty line, in households earning less than $25,000 a year or without a high school diploma, and then examine the changing size and composition of each of those categories by either race or citizenship status, a consistent pattern emerges.

By all three measures, the size of the underclass has been shrinking since the early 1990s‐​during a period of large‐​scale legal and illegal immigration. The composition of the underclass has also been changing, with the number of immigrants and Hispanics growing, while the number of native‐​born and non‐​Hispanics has declined at an even more rapid rate.

Families and individuals below the poverty level. If we define the underclass as families living below the official poverty level, the recent trend has been downward. Between 1995 and 2004, the number of family households living below the poverty level fell by half a million, from 8.1 million to 7.6 million. The number of immigrant households in poverty did indeed rise‐​by 194,000-but that increase was more than offset by a drop of 675,000 in native‐​born households living in poverty. In other words, for every poor immigrant family we “imported” during that time, more than three native‐​born families were “exported” from poverty.2

Poverty figures by race span a longer period, 1993 though 2007, but they tell the same story. The total number of family households living in poverty fell by 770,000 during that period, from 8.4 million to 7.6 million. The number of Hispanic families living in poverty increased by 420,000-providing evidence of a growing Hispanic/​immigrant underclass‐​but over those same years, the number of non‐​Hispanic families in poverty dropped by 1.1 million, including a decline of 408,000 in the number of poor black families.3

The trend is no different when we look at individuals in poverty. From 1993 through 2007, the number of individuals in our society subsisting below the poverty line declined by 2 million, from 39.3 million to 37.3 million. A 1.8 million increase in the number of Hispanics living in poverty was swamped by a 3.8 million decline in non‐​Hispanics, including a 1.6 million decline in black poverty. Similarly, a 1 million increase in immigrants living in poverty was more than matched by a 3 million drop in native‐​born Americans under the poverty line.4 Measured by the official poverty numbers, the American underclass has been shrinking as it has become composed of more immigrants and more Hispanics.

Households with income less than $25,000. Measuring the underclass by household income reveals the same underlying trend. The number of households earning less than $25,000 in a given year dropped by 5.6 million from 1995 to 2004, according to the most recent numbers that disaggregate the underclass by citizenship status. Almost all the drop was accounted for by a decline in non‐​immigrant households earning less than $25,000, which dropped from 20.6 million in 1995 to 15.0 million in 2004. (All incomes were measured in inflation‐​adjusted dollars.) The number of immigrant families under that income threshold also dropped, but only by 80,000. As a result, the immigrant share of the underclass grew from 15 percent to 20 percent, even as the size of the underclass was shrinking.5

The same picture emerges when we examine the number of low‐​income households by race and ethnicity. From 1994 through 2007, the number of households in America getting by on less than $25,000 fell by almost 10 million (with incomes measured across the years in real dollars). The share of total households living under that threshold dropped from 40 percent to 25 percent. Again, the entire decline was accounted for by non‐​Hispanic households, including a drop of 900,000 in black households, while the number of Hispanic households surviving on less than $25,000 was virtually unchanged.

Although the underclass became increasingly more Hispanic during the period, the share of all households living on less than $25,000 fell for every ethnic group. In fact, the steepest decline in percentage terms was among Hispanic households, with the share of households living below $25,000 dropping from 53 percent to 31 percent.6

Householders and Individuals without a High‐​School Diploma. A third way of measuring the underclass is by householders or individuals without a high‐​school diploma. In America today, a worker or head of household without a high‐​school education is almost invariably confined to lowerproductivity, and thus, lower‐​wage occupations, with limited prospects for advancement.

As with the poverty and income measures, here, too, the story is basically positive. Between 1993 and 2006, the number of households headed by someone 18 and older without a high‐​school diploma dropped by 3.7 million, from 19.9 million to 16.2 million. The number of such “low‐​skilled households” headed by a Hispanic did indeed increase by 1.8 million during that period, undoubtedly driven in significant part by large inflows of low‐​skilled immigrants from Mexico and Central America. The rest of the story, however, is that during those same years, the number of non‐​Hispanic households headed by a high‐​school dropout fell by a hefty 5.5 million. That means that for every net addition of one Hispanic‐​headed, low‐​skilled household to the ranks of the underclass, the number of such non‐​Hispanic households dropped by three. Meanwhile, the share of total U.S. households headed by a high‐​school dropout declined steadily, from one in five to one in seven.7

The number of individuals 25 years and older without a high‐​school diploma has also been in steady decline. From 1993 through 2006, the number of adults who were highschool dropouts declined from 32.1 million to 27.9 million, a fall of 4.2 million (see Figure 1). The number of adult Hispanics in the United States without a high‐​school education swelled by 3.9 million, much of that growth driven by the influx of low‐​skilled illegal immigrants. But during that same period, the number of non‐​Hispanic adults toiling in life without a high‐​school diploma plunged by 8.1 million, including a drop of 1 million in the number of adult black dropouts. For every additional Hispanic dropout added to the pool, the number of non‐​Hispanic dropouts fell by two. The share of adults without a diploma dropped in every racial and ethnic group, although the decline was less rapid among Hispanics.8

Educational attainment by citizenship status covers a slightly different period but also confirms the trend. From 1995 to 2004, the number of adults without a high school diploma declined by 2.9 million. An increase of 2.4 million in the number of immigrant dropouts was overwhelmed by a decline of 5.3 million in native‐​born dropouts. As that measure of the underclass shrank, the share represented by immigrants grew from 22 percent to 32 percent. By this and the other measures above, “the underclass” in our society has been shrinking as its face has become more Hispanic and foreign‐​born.9

Immigrants Move In, Americans Move Up

Multiple causes lie behind the shrinking of the underclass in the past 15 years. The single biggest factor is probably economic growth. Despite the current recession, the U.S. economy enjoyed healthy growth during most of the period, lifting median household incomes and real compensation earned by U.S. workers, which ushered millions of families into the middle class and beyond. Welfare reform in the 1990s, and rising levels of education, may also be contributing factors.

Another factor may be immigration itself. The arrival of low‐​skilled, foreign‐​born workers in the labor force increases the incentives for younger native‐​born Americans to stay in school and for older workers to upgrade their skills. Because they compete directly with the lowest‐​skilled Americans, low‐​skilled immigrants do exert mild downward pressure on the wages of the lowest‐​paid American workers. But the addition of low‐​skilled immigrants also expands the size of the overall economy, creating openings in higher‐​paid occupations such as managers, skilled craftsmen, and accountants. The result is a greater financial reward for finishing high school and for acquiring additional job skills. Immigration of low‐​skilled workers motivates Americans, who might otherwise languish in the underclass, to acquire the education and skills necessary so they are not competing directly with foreign‐​born workers.

The shrinking of the native‐​born underclass contradicts the argument that low‐​skilled immigration is particularly harmful to African‐​Americans, who are disproportionately represented in the underclass. By each of the three measures above‐​poverty, income, and educational attainment‐​the number of black American households and individuals in the underclass has been declining. Native‐​born blacks have been moving up along with other native‐​born Americans as immigrants have been moving in.

That same win‐​win dynamic may have been at work a century ago during the “great migration” of immigrants from eastern and southern Europe. Most of those immigrants were lower‐​skilled compared with Americans, and their influx also exerted downward pressure on the wages of lower‐​skilled Americans. It was probably not a coincidence that during that same period the number of Americans staying in school to earn a high‐​school diploma increased dramatically in what is called “the high‐​school movement.” From 1910 to 1940, the share of American 18‐​year‐​olds graduating from high school rose from less than 10 percent to 50 percent in a generation.10 Today’s immigrants are arguably contributing to the same positive dynamic.

America’s experience with immigration contradicts the simplistic argument that the arrival of a certain number of low‐​skilled immigrants increases the underclass by that very same amount. That approach ignores the dynamic and positive effects of immigration on native‐​born American workers. The common calculation that every low‐​skilled immigrant simply adds to the underclass betrays a static and inaccurate view of American society.

A Less Dysfunctional Underclass

Another contribution of immigration has been that it has changed the character of the American underclass for the better. Years of low‐​skilled immigration have created an underclass that is not only smaller than it was 15 years ago, but also more functional. Members of today’s more immigrant and Hispanic underclass are more likely to work and less likely to live in poverty or commit crimes than members of the more native‐​born underclass of past decades.

One striking fact about low‐​skilled immigrants in America, both legal and illegal, is their propensity to work. In 2008, the labor‐​force participation rate of foreign‐​born Hispanics was 70.7 percent‐​compared to an overall rate of 65.6 percent for native‐​born Americans. Immigrants 25 years of age or older, without a high‐​school diploma, were half again more likely to be participating in the labor force than native‐​born dropouts (61.1 percent vs. 38.4 percent).11 According to estimates by the Pew Hispanic Center, male illegal immigrants, ages 18–64, had a labor force participation rate in 2004 of an incredible 92 percent.12 Illegal immigrants are typically poor, but they are almost all working poor.

Nowhere is the contrast between the immigrant and native‐​born underclass more striking than in their propensity to commit crimes. Across all ethnicities and educational levels, immigrants are less prone to commit crimes and land in prison than their native‐​born counterparts.

The reasons behind this phenomenon are several. Legal immigrants can be screened for criminal records, reducing the odds that they will engage in criminal behavior once in the United States. Illegal immigrants have the incentive to avoid committing crimes to minimize their chances of being caught and deported. Legal or illegal, immigrants come to America to realize the opportunities of working in a more free‐​market, open, and prosperous economy; committing a crime puts that opportunity in jeopardy.

Strong empirical evidence points to the fact that immigrants are less likely to commit crimes than native‐​born Americans. In testimony before Congress in 2007, Anne Morrison Piehl, a professor of criminal justice at Rutgers University, addressed the question of “The Connection between Immigration and Crime.” Using census data from 1980, 1990, and 2000, she told the House Judiciary Committee that “immigrants have much lower institutionalization rates than the native born‐​on the order of one‐​fifth the rate of natives. More‐​recently arrived immigrants had the lowest relative institutionalization rates, and the gap with natives increased from 1980 to 2000.” Piehl found no evidence that the immigrant crime rate was lower because of the deportation of illegal immigrants who might otherwise be held behind bars in the United States.13

Crime rates are even lower than average among the poorly educated and Hispanic immigrants that arouse the most concern from skeptics of immigration reform. Rubén Rumbaut of the University of California at Irvine, after examining the 2000 census data, found that incarceration rates among both legal and illegal immigrants from Mexico, El Salvador, and Guatemala were all less than half the rate of U.S.-born whites.14 Immigrants without a high‐​school diploma had an incarceration rate that was one‐​fourth that of native‐​born high‐​school graduates, and one‐​seventh that of native‐​born dropouts.15

The reluctance of low‐​skilled immigrants to commit crimes helps to explain the lack of any noticeable connection between rising levels of illegal immigration and the overall national crime rate. As Professor Rumbaut explained in a recent essay:

Since the early 1990s, over the same time period as legal and especially illegal immigration was reaching and surpassing historic highs, crime rates have declined, both nationally and most notably in cities and regions of high immigrant concentrations (including cities with large numbers of undocumented immigrants, such as Los Angeles, and border cities like San Diego and El Paso, as well as New York, Chicago, and Miami).16

Ironically, illegal immigrants who break U.S. immigration laws to enter the United States appear much more likely than native‐​born Americans to respect our domestic criminal code once they are inside the country. Once here, low‐​skilled immigrants, as a rule, get down to the business of earning money, sending home remittances, and staying out of trouble. The wider benefit to our society is that, in comparison to 15 years ago, a member of today’s underclass, standing on a street corner, is more likely to be waiting for a job than a drug deal.

Contrary to popular notions, low‐​skilled immigration has not contributed to a swelling of the underclass, or any increase at all, nor has it contributed to a rise in crime or other antisocial behaviors. In fact, it would be more plausible to argue that low‐​skilled immigration has actually accelerated the upward mobility of Americans on the lower end of the socioeconomic ladder. At the same time, the influx of lowskilled immigrants has helped to transform the American underclass into a demographic group that is still poor‐​but more inclined to work and less prone to crime.


Members of Congress should not reject market‐​oriented immigration reform because of misguided fears about “importing poverty.” Based on recent experience, a policy that allows more low‐​skilled workers to enter the United States legally would not necessarily expand the number of people living in poverty or the number of low‐​skilled households demanding government services. It would not impose significant costs on American society in the form of welfare spending or crime abatement.

As Cato research has shown elsewhere, strong, positive arguments remain for pursuing a policy of expanding legal immigration for low‐​skilled workers. Comprehensive immigration reform that included a robust temporary worker program would boost economic output and create new middleclass job opportunities for native‐​born Americans. It would reduce the inflow of illegal workers across the border, allowing enforcement resources to be redeployed more effectively to interdict terrorists and real criminals. It would restore the rule of law to U.S. immigration policy, while reducing calls for enforcement measures such as a national ID card or a centralized employment verification system, which compromise the freedom and civil liberties of American citizens.17

Along with those major benefits, immigration reform would enhance the incentives for native‐​born Americans up and down the income ladder to acquire the education and skills they need to prosper in a dynamic economy.


1 See, for example, Robert Rector, “Importing Poverty: Immigration and Poverty in the United States: A Book of Charts,” Heritage Foundation, Special Report #9, October 25, 2006; Heather Mac Donald, “Surge in Birth Rate Among Unwed Hispanics Creating New U.S. Underclass,” Dallas Morning News, Jan. 21, 2007; and Robert J. Samuelson, “Importing Poverty,” The Washington Post, September 5, 2007.

2 U.S. Bureau of the Census, Current Population Survey, Foreign Born Population, 1995–2004, Survey Table 1.12, http://www. cen​sus​.gov/cps/.

3 U.S. Bureau of the Census, Current Population Survey, Poverty Status of Families, Table 4, http://​www​.cen​sus​.gov/cps/.

4 For poverty numbers by citizenships status, see U.S. Bureau of the Census, Current Population Survey, Historical Poverty Tables, People, Table 23; by race, see ibid., Table 2, http://www.census. gov/​cps/​.

5 U.S. Bureau of the Census, Current Population Survey, Foreign Born Population, 1995–2004, Survey Table 1.9, http://www.census. gov/​cps/​.

6 U.S. Bureau of the Census, Current Population Survey, Detailed Household Income, Table H-1, http://​www​.cen​sus​.gov/cps/.

7 U.S. Bureau of the Census, Current Population Survey, Educational Attainment, Table 4, http://​www​.cen​sus​.gov/cps/. Data is not available for householder’s education level by citizenship status.

8 Ibid.

9 U.S. Bureau of the Census, Current Population Survey, Foreign Born Population, 1995–1999, Survey Table 1.4, and 2000–2004, Survey Table 1.5, http://​www​.cen​sus​.gov/cps/.

10 See, for example, Claudia Goldin and Lawrence F. Katz, “Human Capital and Social Capital: The Rise of Secondary Schooling in America,” National Bureau of Economic Research, Working Paper no. 6439, March 1998.

11 U.S. Bureau of Labor Statistics, “Foreign‐​born Workers: Labor Force Characteristics in 2008,” Bureau of Labor Statitstics (news release, March 26, 2009), Table 1.

12 Jeffrey S. Passel, “Unauthorized Migrants: Numbers and Characteristics,” Pew Hispanic Center, June 14, 2005, p. 25.

13 Anne Morrison Piehl, “The Connection between Immigration and Crime” (testimony before the House Judiciary Subcommittee on Immigration, Citizenship, Refugees, Border Security, and International Law, hearing on “Comprehensive Immigration Reform: Impact of Immigration on States and Localities,” May 17, 2007).

14 Rubén Rumbaut, “Undocumented Immigration and Rates of Crime and Imprisonment: Popular Myths and Empirical Realities,” Appendix D, in The Role of Local Police: Striking a Balance Between Immigration Enforcement and Civil Liberties, ed. Anita Khashu (Washington: The Police Foundation, April 2009), p. 127.

15 Ibid., p. 129.

16 Ibid., p. 124.

17 See Jim Harper, “Electronic Employment Eligibility Verification: Franz Kafka’s Solution to Illegal Immigration,” Cato Policy Analysis no. 612, March 6, 2008; Daniel Griswold, “The Fiscal Impact of Immigration Reform: The Real Story,” Cato Free Trade Bulletin no. 30, May 21, 2007; Daniel Griswold, “Comprehensive Immigration Reform: Finally Getting It Right,” Cato Free Trade Bulletin no. 29, May 16, 2007; and Daniel Griswold, “Willing Workers: Fixing the Problem of Illegal Mexican Migration to the United States,” Cato Trade Policy Analysis no. 19, October 15, 2002.