Printer Friendly

How do immigrants fare in the U.S. labor market?

Recent immigrants earn less and have higher jobless rates than do earlier postwar immigrants and U.S. natives; educational attainment and English fluency are important factors in labor market success

Immigration to the United States has increased markedly in recent decades. In fact, the number of immigrants granted permanent legal residence during the 1980's was the highest since the 1910-19 decade.(1) (See chart 1.) Undocumented immigration, although difficult to measure precisely, also appears to have risen. (2) These developments have prompted policy-makers, employers, labor unions, social scientists, the news media, and the general public to focus anew on immigration and its impact on American society.

One area of interest is the way in which the labor market status of immigrants compares with that of U.S. natives. This issue can be examined using data from a survey conducted in November 1989. This survey showed, for example, that the unemployment rate for immigrants was somewhat higher than the rate for native-born workers, and that the weekly earnings of immigrants who worked full time were significantly lower than those of natives. The survey also pointed to differences in the level of schooling as a major reason for these disparities. Although immigrants and natives aged 25 and older were equally likely to have completed at least 4 years of college, the proportion of immigrants who had completed fewer than 12 years of school was nearly double the proportion of natives. Other factors affecting the labor market status of immigrants include the length of time they had lived in the United States and their fluency in English. This article presents an analysis of the relationship between these factors and immigrants' experiences in the U.S. labor market.

Data source and technical issues

The data used in this study are from the Current Population Survey (CPS), the monthly survey of about 60,000 households conducted by the Bureau of the Census for the Bureau of Labor Statistics, that provides a variety of information on the U.S. labor force. In November 1989, a set of supplemental questions about the characteristics of immigrants was included in the CPS.(3) One of these questions asked respondents to name their country of birth. Those who responded that they were born in the United States, Puerto Rico, or another U.S. territory, or that they were born abroad of an American parent or parents, are classified as U.S. natives.(4) Individuals who provided any other response were classified as immigrants? Respondents identified as immigrants were asked additional questions about the year they came to the United States to stay and the language they speak at home. Those who spoke a language other than English were also asked how well they spoke English.

At the time of the survey, there were 187.0 million people of working age (16 years and older) in the civilian noninstitutional population. Of that total, 87.5 percent (163.6 million) were identified as U.S. natives and 8.0 percent (14.9 million) as immigrants. The remaining 4.6 percent did not respond to the question on country of birth and are excluded from the analysis.6 As table 1 shows, two-thirds of Asians were born outside the United States, the largest foreign-born proportion of any race-ethnic group. A substantial share of the Hispanic population was foreign born as well, but more than half of the Hispanics residing in the 50 States or the District of Columbia were born in the United States (including Puerto Rico). Immigrants comprised only small proportions of the non-Hispanic white and black population groups (3.3 and 4.2 percent, respectively).

Labor force participation

A number of factors may influence a person' s decision to leave his or her native land and settle in a new country. These include family and cultural considerations, legal restrictions on exiting one country or entering another, the costs of travel to a new country, and differences between countries in political or religious environment. Perhaps the most important factor for many people is how their expected economic opportunities in their new country compare with those in their native country. The principal way in which immigrants (as others) can take advantage of the economic opportunities in the United States is to participate in the labor force. This section examines labor force participation of immigrants from three different, although related, perspectives: length of residence in the United States, number of years of schooling, and fluency in English.

Length of U.S. residence. Among men under age 55, there was little difference between immigrants and U.S. natives in the labor force participation rate--the proportion of each population group that is either employed or actively seeking employment. (See table 2.) Nor was there much variation in labor force participation among immigrants by the number of years they had lived in the United States. The notable exception to this pattern, in each age group, were immigrant men who had lived in this country for fewer than 3 years (arriving in the 1987-89 period). Specifically, these men were less likely to participate in the labor force than were immigrants who had lived in the United States for longer periods. (See table 3.) For example, among immigrant men aged 35 to 44, 85 percent of those residing here for fewer than 3 years were in the labor force. This compares with participation rates of 97 percent for those having lived in the United States for 3 to 8 years (arriving during 1982-86) and 95 percent for those having lived here for 8 to 15 years (arriving during 1975-81). The lower participation rates of the most recent immigrants suggest that it takes a few years after arrival for persons to be assimilated into the work force, as they gain information on the U.S. labor market, establish economic contacts, and, where necessary, improve their fluency in English.

In contrast to men, immigrant women under age 55 were much less likely than natives to participate in the labor force. However, in every age group, the participation rates of immigrants drew closer to those of natives the longer the immigrants had lived in the United States. For example, among 35- to 44-year-old women, 78 percent of natives and 68 percent of immigrants participated in the labor force. Women in this age group who had lived in the United States for fewer than 3 years had a participation rate of 49 percent, compared with 65 percent for those who had resided here for from 3 to 8 years and 71 percent for those who had arrived between 8 and 15 years before the survey reference date. The rate for those who had immigrated 30 years ago or more--that is, as children-was 77 percent, virtually the same as the rate for natives. It would appear that immigrant women who had lived in the United States the longest are more likely to have had experiences similar to those of natives, and this may explain why their labor force participation rates also were similar.

Educational attainment. Formal education is another means through which immigrants gain the skills and knowledge that are essential for success in the labor market. Schooling contributes to broader job opportunities, higher earnings, and increased likelihood of labor force participation for the native-born and immigrants alike.7 So, how do the educational levels of immigrants compare with those of natives?

This issue is examined by focusing on people aged 25 and older, thereby excluding 16- to 24-year-olds, many of whom have not yet completed their formal education. As table 4 shows, immigrants and U.S. natives in the central working ages of 25 to 54 were about equally likely to have completed 4 years of college or more. College graduates accounted for 28 percent of immigrant men and 27 percent of native men, and for 23 percent of immigrant and native women.

In stark contrast, nearly a quarter of 25- to 54-year-old immigrant men and women had completed fewer than 9 years of school, compared with only 4 percent of native men and 3 percent of native women. By comparison, the proportions of U.S. natives who had completed exactly 12 years of school or 1 to 3 years of college were about one-and-a-half times the percentages for immigrants. The educational levels of immigrants and natives aged 55 and older showed a similar contrast, although persons in both groups generally had completed less schooling than their younger counterparts.

Although the overall figures on educational attainment show that immigrants typically had completed less schooling than U.S. natives, very different patterns emerge within the race-ethnic groups. As table 5 shows, among 25- to 54-year-old men, non-Hispanic white, black, and Asian immigrants were considerably more likely than their native-born peers to have completed 4 years of college or more. In fact, the proportion of black immigrant men who were college graduates was more than double the share among black men born in the United States. Black immigrant men also were less likely than their native-born counterparts to have completed fewer than 12 years of school, while the proportions of Asian and white immigrants with less than 12 years of schooling were slightly larger than the proportions of Asian and white men born in this country.

Among 25- to 54-year-old women, the distribution of black immigrants across the educational spectrum was fairly similar to that of black women bom in the United States. Asian and white immigrant women, however, were somewhat more likely than their native-born counterparts to be college graduates. Immigrants in both of these racial groups also were more likely than U.S. natives to have completed fewer than 12 years of school, but the difference was much larger among Asians.

Among Hispanic men and women, the contrast in educational levels between immigrants and U.S. natives was considerably sharper than it was among non-Hispanic whites, blacks, and Asians. Hispanics born in this country generally had completed much less schooling than other U.S. natives, and the educational levels of Hispanic immigrants were even lower. For example, 30 percent of Hispanic men born here had completed fewer than 12 years of school, compared with 55 percent of immigrant Hispanic men.

How do the labor force participation rates of immigrants and natives compare at each educational level? Among 25- to 54-year-old immigrant men, the rates were virtually the same-- about 93 percent--across all educational levels. By contrast, the rates for native men rose at each successive educational level, ranging from a low of 75 percent for those with fewer than 9 years of school, to 95 percent for those with exactly 12 years of school, and 97 percent for college graduates. (See table 6.)

For most population groups, labor force participation rates rise with the level of education. This suggests that the investment of more time and money in schooling heightens people's expectations and realization of expanded job opportunities and higher earnings. The labor force participation rates of native men aged 25 to 54 follow this pattern. Why, then, do the rates for their immigrant counterparts not vary by their level of education?

The explanation could lie in the reason why many of these men immigrated to the United States in the first place: to improve their economic conditions. And to do so, they generally must participate in the labor force. Thus, it makes sense that their participation rates would be rather high regardless of their educational attainment.

Like the labor force participation rates for native-born men, the rates for both immigrant and native-born women generally rose at each successive level of education. The difference in participation rates between the least- and most-educated women was much smaller among immigrants, however. The rates for 25- to 54-year-old immigrant women ranged from 49 percent for those with fewer than 9 years of schooling to about 76 percent for those with 1 to 3 years of college. Among U.S. natives, the range was from 44 percent for those with fewer than 9 years of schooling to 85 percent for college graduates.

The participation rates of both immigrants and natives aged 55 and older also tended to be greater at successively higher levels of education. Among men, the participation rates for immigrants and natives were virtually the same in all but one educational attainment category. Older immigrant men who had completed fewer than 9 years of school participated in the labor force at a considerably higher rate (35 percent) than their native counterparts (22 percent). Immigrant women aged 55 and over participated at somewhat higher rates than did their native-born peers at each educational level.

Fluency in English. Work experience and education are often considered important "human capital" characteristics that enhance the productive capacity of workers. Persons with greater experience and higher levels of education can expect to realize broader job opportunities and higher earnings. The discussion of labor force participation so far has focused on two proxy measures of experience-age and duration of residence in this country--as well as on educational attainment. For U.S. immigrants, the traditional list of human capital characteristics needs to be expanded to include English fluency.

Two questions relating to English fluency were asked of immigrants in the November 1989 CPS. The first question was:
 Does...speak a language other than English at home?
 Yes ............................O
 No - Speaks only English .......O
 Immigrants answering "yes" to this question were
then asked:
 How well does... speak English?
 Very well ......................O
 Well ...........................O
 Not well .......................O
 Not at all .....................O

The responses to these questions are used to examine the English fluency of immigrants in each race-ethnic group, with an eye toward gauging the relationship between fluency and labor force participation. Before this is done, two points need to be made about the limitations of the data. First, the responses to the question on English fluency are based solely on the subjective opinions of respondents, rather than on any uniformly objective criteria. Thus, a household member described as speaking English "very well" may not necessarily speak as fluently as a member of another household who is described as speaking English "well." Because the distinctions among the four fluency categories are somewhat arbitrary, immigrants responding "very well" or "well" are referred to as fluent, and those responding "not well" or "not at all" are termed not fluent. Even within this scheme, however, survey respondents may report a higher or lower fluency level than that which might be indicated by some objective measure.

A second important limitation of the data on fluency is that the survey question asks only how well an immigrant speaks English, which may not perfectly correlate with how well the immigrant reads, writes, or understands English. Speaking ability, however, is perhaps the most apparent indicator available to an employer in assessing an immigrant's ability to communicate in English. Although self-reported speaking ability may not be the ideal measure of English fluency, the analysis presented below shows it to be closely related to the labor market success of immigrants.

As table 7 shows, 24 percent of immigrants aged 16 and older spoke only English at home. Another 45 percent spoke a language other than English at least some of the time and spoke English "very well" or "well." The remaining 32 percent spoke English "not well" or "not at all."

This pattern varied little by gender, but there were large differences across racial and ethnic groups. Among non-Hispanic black immigrants, many of whom came from English-speaking countries, 65 percent spoke only English at home. This compares with 45 percent of non-Hispanic whites, 14 percent of Asians, and 5 percent of Hispanics. Although the proportion of Asian immigrants who spoke only English at home was relatively small, an additional 57 percent who spoke another language at home spoke English fluently. Still, 29 percent of Asian immigrants spoke little or no English, a much larger proportion than the 10 percent of black and 13 percent of white immigrants, but considerably smaller than the 53 percent of Hispanics.

What might explain why Hispanic immigrants are so much less likely than non-Hispanic black, white, and Asian immigrants to be fluent in English? One factor is that many non-Hispanic immigrants emigrated from countries in which English is the primary language, such as the United Kingdom, Ireland, and Jamaica. Still others came from countries such as India and the Philippines, where English, although not the primary language, is commonly spoken, particularly among more highly educated persons. Hispanic immigrants, by contrast, largely came from countries in which English is not often spoken.

The gap in English fluency between Hispanic and non-Hispanic immigrants might also be explained by differences in length of U.S. residence. The following tabulation shows the percentage distribution of immigrants in each racial or ethnic category by the year they came to the United States to stay. (This tabulation excludes the few immigrants whose year of immigration is unknown.)
Year of
immigratian White Black Asian Hispanic
Total ....... 100.0 100.0 100.0 100.0
1982-89 ..... 16.0 32.9 36.8 32.3
1975-81 ..... 12.5 29.1 34.9 29.4
1965-74 ..... 19.0 27.9 19.2 22.7
Before 1965... 52.5 10.1 9.0 15.6

Non-Hispanic white immigrants generally had lived in the United States much longer than had Hispanic immigrants. This suggests that white immigrants who could not speak English at the time of their arrival have had more time to learn the language. Although the longer U.S. residence of whites partially explains their greater English fluency, the tabulation shows little difference between Hispanics and non-Hispanic blacks and Asians in the number of years they had lived in the United States.

The existence of large Spanish-speaking enclaves in this country also may help to explain the lower level of English fluency among Hispanic immigrants. Throughout much of U.S. history, immigrants have settled in neighborhoods with others who spoke their native languages. In the late 19th and early 20th centuries, for example, there were large enclaves of immigrants from Germany, Italy, Russia, Poland, and elsewhere. In time, these groups dispersed as they learned to speak English and began to assimilate into the multi-ethnic U.S. culture.

Enclaves of a variety of ethnic groups still exist today. Among them, Hispanic enclaves are probably the largest and most numerous. One reason is that the working-age population of Hispanic immigrants--about 6 million--is nearly twice the size of the Asian immigrant population and more than 6 times the size of the black immigrant population. In addition, Spanish is the native language of virtually all Hispanic immigrants. Among Asian and black immigrants, however, there is considerable language diversity, which limits their potential to form large enclaves in which any particular non-English language will dominate. Furthermore, many of the 7.2 million working-age Hispanics born in the United States also speak Spanish, whereas few U.S. natives speak the native languages of non-Hispanic immigrants.

A recent study sponsored by the National Center for Health Statistics (NCHS) used census data to examine racial and ethnic residential segregation. Although the NCHS study did not include country of birth as a variable, it did provide some important insights about the residential patterns of Asians and Hispanics, large proportions of whom are immigrants. This study found that Asians were the minority group most integrated into the U.S. population in 1990. The Hispanic population, in contrast, was heavily concentrated within enclaves, and, in fact, residential segregation of Hispanics had increased since 1980. This increase was most likely due to the massive influx of Hispanic immigrants during the 1980's, many of whom settled in areas where there were large numbers of Spanish speakers.(8)

The fact that there are so many Spanish speakers in the United States may enable Hispanic immigrants to create labor markets that are somewhat independent of the larger, English-speaking labor market. Thus, for Hispanic immigrants in enclaves, English fluency may be less vital for labor market success than it is for Asian and black immigrants.(9) This may be true not just for Hispanic immigrants in residential enclaves but for those in broader economic enclaves as well. For example, groups of Hispanic immigrants not fluent in English work as migrant farmworkers in some rural areas. Similarly, in a number of metropolitan areas, there are groups of Hispanic immigrants who work for English-speaking employers in construction or service occupations. In such instances, a bilingual intermediary often enables English-speaking employers and immigrant workers not fluent in English to communicate. Although economic enclaves do provide job opportunities for Hispanic immigrants who do not speak English well, these immigrants may experience difficulties if they seek work outside the enclaves.

Let us now examine the relationship between English fluency and the labor force participation rates of immigrant men and women in each race-ethnic group. These rates also are compared with those of U.S. natives. (The participation rates for non-Hispanic black immigrants are not analyzed because the population is too small to obtain reliable estimates of participation in each language fluency category.) To control for the different age distributions of immigrants and natives in each race-ethnic group, labor force participation rates are analyzed only for persons in the central working ages of 25 to 54.

Among non-Hispanic white men, 95 percent of U.S. natives and immigrants who spoke only English at home participated in the labor force. The participation rate for white immigrants who spoke another language but spoke English fluently was 93 percent, considerably higher than the 82-percent rate for those who did not speak English fluently. (See table 8.)

The labor force participation rate for men of Asian descent who were born in the United States was 92 percent. Nearly all Asian immigrant men who spoke only English at home were in the labor force (98 percent), as were 93 percent of those who spoke another language and were fluent in English. By comparison, only 74 percent of Asian immigrant men who spoke little or no English participated in the labor force.

Among Hispanic men, the participation rate for U.S. natives was 92 percent, compared with 95 percent for their immigrant counterparts. Unlike the rates for non-Hispanic white and Asian immigrants, participation rates among Hispanic immigrants varied little by degree of English fluency. This seems to support the hypothesis that, because of the size of the Spanish-speaking population and their concentration in Hispanic enclaves, immigrant Hispanic men who are not fluent in English have more abundant job opportunities than do non-Hispanic immigrants. The availability of jobs within enclaves may, in turn, reduce the economic need for Hispanic immigrants to become fluent in English.

However, the existence of Hispanic enclaves does not seem to have the same effect on the labor force participation of immigrant Hispanic women who do not speak English fluently. Only 47 percent of these women participated in the labor force. In contrast, the participation rate was 71 percent for immigrant Hispanic women who spoke only English at home, and 75 percent for those who spoke both another language and English fluently. Hispanic women who were born in the United States participated in the labor force at a 68-percent rate. The participation rates of non-Hispanic women followed essentially this same pattern. That is, the rates for U.S. natives were fairly similar to those for immigrants who spoke English fluently, and the rates were much lower for immigrants who did not speak English fluently.


The unemployment rate--the proportion of labor force participants who do not have jobs but are actively looking for work--was 5.7 percent for both immigrant men and immigrant women in November 1989. This was half a percentage point higher than the rates for men and women born in the United States. Although these differences appear rather small, they mask the fact that recent immigrants--specially those who came to this country during the 1982-89 period--were much more likely to be unemployed than were immigrants who arrived during earlier periods. The following tabulation shows the unemployment rates of immigrant men and women by the years in which they came to the United States to stay:
 Men Women
All years ........ 5.7 5.7
1982-89 ...... 7.5 8.4
1975-81 ...... 6.1 6.5
1965-74 ...... 4.1 3.4
Before 1965 .... 4.4 5.2

The higher unemployment rates of the most recent immigrants can be explained, in part, by the fact that a much larger proportion of their labor force was 16 to 24 years old--the age range for which unemployment rates tend to be highest among immigrants and natives alike. For example, 27 percent of male immigrants who arrived during 1982-89 were 16-to 24-year-olds. This age group accounts for 17 percent of U.S. natives, 15 percent of immigrants who arrived during 1975-81, and 8 percent of those who arrived in the 1965-74 period.

Even within particular age groups, the pattern of higher unemployment rates for recent immigrants and lower rates for earlier immigrants appears to prevail? This gap stems in part from differences between immigrant waves in the education and skills they possessed before they immigrated. (This phenomenon is examined more extensively in the section on occupational employment.) Differences in length of residence in the United States also play a role. This is because recent immigrants have had less time to learn the workings of the U.S. labor market and establish economic contacts than have earlier immigrants and U.S. natives. In addition, recent immigrants are less likely to be fluent in English than are those who have lived in the United States for longer periods. As table 8 shows, immigrants who did not speak English fluently generally experienced much higher unemployment rates than those who were fluent.(11)

Occupational employment

Among employed men, immigrants who moved to this country during the 1982-89 period were much more likely than natives to work in occupations that are generally lower paying. For example, 19 percent of recent immigrants--versus 9 percent of U.S. natives--worked in service occupations, which include such jobs as food preparation, child care, and janitorial services. One-fourth of recent immigrants, compared with one-fifth of natives, worked as operators, fabricators, and laborers. Farming, forestry, and fishing occupations accounted for 9 percent of employment among recent immigrants, but for only 4 percent among natives. U.S. natives were more concentrated in higher-paying jobs than were immigrants who arrived here during the 1982-89 period. For example, 26 percent of employed native men worked in managerial and professional occupations, nearly 1.6 times the proportion of recent immigrants. There was also a contrast between the groups within precision production, craft, and repair occupations, which accounted for 20 percent of employed U.S. natives, versus 16 percent of recent immigrants. (See table 9.)

Immigrant men who arrived during 1965-74 and 1975-81 were also more likely than U.S. natives to work in low-paying occupations and less likely to work in high-paying ones. These differences, however, were small when compared to the gap between natives and the most recent immigrants.

Immigrant men who arrived before 1965 were actually more likely to work in higher-paying jobs than were U.S. natives. Managers and professionals accounted for 36 percent of employment among these earlier immigrants. At the same time, only 11 percent of pre-1965 immigrants worked as operators, fabricators, and laborers.

Among women, the occupational distributions of U.S. natives and immigrants who arrived before 1965, although not identical, were fairly similar. By comparison, more recent immigrants held jobs very different from those held by natives. This was true even of immigrant women who had lived in this country for as long as 15 to 25 years (that is, persons who arrived during the 1965-74 period). These differences were generally smaller, however, than those between U.S. natives and immigrant women who arrived after 1974.

Service occupations, for example, accounted for 17 percent of employment among native women, and for 19 percent among women who immigrated before 1965. These jobs made up a somewhat larger share of employment among 1965-74 immigrants (23 percent), and a much larger share among immigrants who had arrived during the 1975-81 (28 percent) and 1982-89 (30 percent) periods. Them were also large representational differences among women in managerial and professional occupations and administrative support (including clerical) jobs. U.S. natives and immigrants who came to this country before 1965 were considerably more likely to work in these occupations than were women in the three more recent immigrant groups.

In summary, immigrant men and women who had lived in the United States the longest generally worked in higher paying occupations than did more recent immigrants. Furthermore, the occupational characteristics of the earlier immigrants were much more similar to those of natives than were those of newer arrivals. Although this pattern may suggest that length of U.S. residence goes hand in hand with labor force assimilation, it also results from factors other than assimilation.(12) The different age distributions of U.S. natives and the various immigrant cohorts clearly play a role. For example, among employed men, 26 percent of immigrants who came to the United States during the 1982-89 period were 16 to 24 years old. By comparison, 15 percent of U.S. natives and practically none of the pre-1965 immigrants fell into that age range. This might explain why, relative to U.S. natives, a smaller proportion of 1982-89 immigrants and a larger proportion of pre-1965 immigrants worked in managerial and professional jobs.

Differences between immigrant cohorts in education and skills obtained before they came to the United States might also explain their different occupational characteristics. Before enactment of the Immigration Act of 1965, U.S. policy gave preference to European immigrants and severely restricted immigration from Asia. The 1965 Act replaced this national-origin preference system with a policy that favored immigrants who have relatives in the United States. The Act also permitted greater immigration from Asia.(13)

In a study using data from the 1940-80 U.S. censuses, George J. Borjas found that the proportion of the immigrant flow that came from Europe declined during the period studied, and the proportions from Asia and Latin America rose.(14) This shift in the national-origin composition of immigrants resulted partly from the 1965 Immigration Act and subsequent amendments to U.S. immigration law. Just as important, however, were the changes in economic and political conditions around the world. Borjas noted that Western Europe became more economically and politically stable during the period studied, reducing the incentive for Western Europeans to immigrate to the United States. At the same time, many parts of Asia and Latin America became less stable, increasing the incentive for residents of these countries to emigrate. Borjas observed that this shift in the national-origin mix resulted in a deterioration in the educational and skill levels of successive immigrant waves, particularly after the 1960 census. This would partly explain why pre-1965 immigrants are more likely to work m higher-paying occupations than are more recent immigrants.

The phenomenon of return migration might also help to explain the differences in occupational characteristics between earlier and more recent immigrants. Borjas cited research showing that 20 to 30 percent of immigrants return to their native countries (or migrate elsewhere) within one to two decades after their arrival in the United States. If it could be shown that these outmigrants were people who generally did not fare well in the U.S. labor market, then immigrants from earlier waves who remained in the United States might appear to have been more "successful" than recent immigrants. That is, recent immigrants have not lived in this country long enough to evaluate their likelihood of labor market success and decide whether to remain or outmigrate. Borjas noted, however, that the studies do not reach a consensus on the reasons for return migration. Thus, it is uncertain how much return migration affects the occupational characteristics of each immigrant group.

Weekly earnings

One result of the difference in occupational characteristics between immigrants and U.S. natives is a rather substantial gap in earnings. The median weekly earnings of immigrant men who worked full time (35 hours or more per week) as wage and salary employees were $377 ([+ or -] $24).(15) This was 78 percent of the median for U.S. natives ($486 [+ or -] $6). This gap in earnings was noted for each age group under 55 years; in the older age groups, the median earnings of immigrants were not significantly different from those of natives. (See table 10.)

Among women, the median weekly earnings of immigrants ($296 [+ or -] $16) were 86 percent of the median for natives ($346 + $6). This earnings gap was concentrated primarily among 25- to 34-year-olds and 35- to 44-year-olds, the two largest age groups among both immigrants and natives. As with men, the median earnings of immigrant women in the 55-to-64 and 65-and-older age groups were not significantly different from those of their native counterparts. In addition, there were no statistically significant earnings differences between immigrant and native women aged 16 to 24 and 45 to 54.

The earnings of immigrants varied by the length of time they had lived in the United States. To examine the relationship between earnings and length of residence, immigrants have been divided into three year-of-entry groups: 1982-89, 1975-81, and before 197 5. Men and women aged 16 and older who immigrated during 1982-89 had somewhat lower median weekly earnings than did their counterparts who had immigrated during 1975-81. The 1975-81 immigrants, in turn, earned considerably less than immigrants who had arrived before 1975. Pre-1975 immigrants, in fact, had essentially the same median earnings as U.S. natives, and this equality prevailed in virtually all age-gender groups. (See table 11 .)

English fluency also affects the earnings of immigrants. Among immigrant men who spoke a language other than English at home at least some of the time, those who were not fluent in English earned only about half as much as those who were. The earnings difference between fluent and non-fluent women, although not so large as the gap among men, was also substantial. Immigrant women who spoke little or no English earned about three-fourths as much as those who spoke English "very well" or "well." These English-fluent immigrants, in turn, earned less than immigrants who spoke only English at home. In fact, immigrants who spoke only English at home actually earned somewhat more than U.S. natives. (See table 12.)

These data demonstrate the economic benefits of English fluency for immigrants, but they do not indicate whether these benefits vary across ethnic groups. For example, is the earnings gap between fluent and nonfluent Hispanic immigrants larger or smaller than the corresponding gap for Asian immigrants? The CPS sample size is not large enough to provide reliable earnings estimates by English fluency and ethnicity, but other studies have addressed this question using data from the Survey of income and Education, which was conducted in 1976 by the U.S. Bureau of the Census. These studies yield different conclusions, however, leaving the issues open to further investigation. (16)

THE QUESTION AS TO HOW immigrants fare relative to U.S. natives in the labor market cannot be answered in only a few words because immigrants are by no means a homogeneous group. Recent immigrants have much higher unemployment rates than natives, and their sharply different occupational characteristics result in much lower earnings. By comparison, immigrants who have lived in the United States the longest exhibit labor market characteristics that are similar to those of natives. This contrast between recent and earlier immigrants partly reflects differences between these groups in the skills and education they had acquired before their arrival in this country. It also results from the fact that earlier immigrants have had more time to assimilate into the U.S. labor market by gaining work experience and mastering the English language. The labor market performance of immigrants who speak English fluently compares much more favorably with that of natives than does the performance of immigrants who are not fluent. This fact is particularly important for Hispanic immigrants, more than half of whom speak English "not well" or "not at all." The existence of large Spanish-speaking enclaves, however, seems to enhance access to employment for some Hispanic immigrants.


1. 1989 Statistical Yearbook of the Immigration and Naturalization Service (Washington, U.S. Department of Justice, Immigration and Naturalization Service, 1990).

2. For an indepth discussion of techniques used to estimate the number of undocumented immigrants, see Frank D. Bean, Barry Edmonston, and Jeffrey S. Passel, eds., Undocumented Migration to the United States: InCA and the Experience of the 1980s (Washington, The RAND Corporation and The Urban Institute, 1990).

3. To the extent possible, CPS respondents who speak lime or no English are interviewed in their native languages, either by a bilingual interviewer or with the assistance of an interpreter. Thus, nonresponse resulting from language barriers is mimmal.

4. Although people living in the United States who were born in Puerto Rico or

another U.S. territory could be viewed as immigrants, they were not treated as such in the November 1989 CPS. For that reason, they were not asked questions regarding immigration that were asked of others born outside the United States.

5. This definition of immigrants differs from that used by the U.S. Immigration and Naturalization Service. As stated in the 1989 Statistical Yearbook of the Immigration and Naturalization Service, an immigrant is "an alien admitted to the United States as a lawful permanent resident. Immigrants are those persons lawfully accorded the privilege of residing in the United States. They may be issued immigrant visas by the Department of State overseas or adjusted to permanent resident status by the Immigration and Naturalization Service in the United States." In the Current Population Survey, foreign-born persons living in the United States are referred to as immigrants regardless of whether they have permanent legal resident status.

6. Occasionally, usable responses to some CPS questions are not obtained from some sample members. Reasons for nonresponses include error on the part of CPS interviewers or the reluctance of some sample members to provide answers to certain questions. When this happens, a procedure called imputation often is performed. Say, for example, that a sample member provides answers to all CPS questions except the one on how much he or she usually earns per week. Through the imputation procedure, this nonrespondent would be assigned the weekly earnings value of a sample member who provided a response on weekly earnings and who had demographic and employment characteristics similar to those of the nonrespondent. In the November 1989 CPS supplement on immigration, this imputation procedure was not used because there was insufficient information to assign accurately those whose country of birth was unknown to either the native- or foreign-born category. For the same reason, imputation also was not used for immigrants not responding to the other supplemental questions. Given the high response rates for all questions, excluding nonrespondents should not significantly affect the results of this study.

7. No distinction is made in the CPS between pre- and postimmigration schooling.

8. David Judkins, Joseph Waksberg, and James Massey, "Patterns of Residential Concentrations by Race and Hispanic origin," paper presented at the Mar. 17, 1992, meeting of the Washington Statistical Society.

9. See Waiter S. McManus, "Labor Market Effects of Language Enclaves: Hispanic

Men in the United States," Journal of Human Resources, Spring 1990, pp. 228-52. Using data from the 1980 census, McManus found that the larger the concentration of Hispanics in a particular county, the greater the job opportunities and earnings of Hispanic men who lacked fluency in English, This suggests that the labor market returns to English fluency and, thus, the incentive to learn English, are lower in larger enclaves.

10. It should be noted that tabulations of unemployment rates for immigrants by their sex, age, and year of immigration are based on small samples, and thus have large margins of sampling error. For this reason, these unemployment rates am not published in this article. Unpublished tabulations are available from the Bureau upon request.

11. The reader should be reminded that these data were collected in November 1989, 8 months prior to the official start of the most recent recession. If these data had been collected during the recession, the gap in unemployment rates between recent immigrants and earlier immigrants and U.S. natives might have been even larger. This is because workers with relatively low levels of education, experience, and, presumably, English fluency typically experience larger increases in unemployment when economic conditions deteriorate.

12. See, for example, Sherrie A. Kossoudji, "Immigrant Worker Assimilation: Is It a Labor Market Phenomenon?" The Journal of Human ReSources, Summer 1989, pp. 494-527.

13. For a more thorough discussion of the legislative history of the 1965 Immigration Act, see David M. Reimers, "Recent Immigration Policy: An Analysis,"in Barry R. Chiswick, ed., The Gateway: U.S. Immigration Issues and Policies (Washington and London, American Enterprise Institute for Public Policy Research, 1982), pp. 13-53.

14. George J. Borjas, National Origin and the Skills of Immigrants in the Postwar Period, Working Paper No. 3575 (Cambridge, MA, National Bureau of Economic Research, Inc., 1991).

15. Because the estimates used in this article are based on one of many possible samples of the working-age population, they are subject to sampling error--vanation that occurs by chance because a sample, rather than the whole population, is surveyed. The measure of this sampling variability is the standard error. The standard error and the sample estimate enable confidence intervals to be constructed. In this article, the 90-percent confidence interval--the sample estimate plus or minus 1.6 times the standard error--is used to analyze data on median weekly earnings. To illustrate, the median weekly earnings of immigrant men who work full time is $377, and the standard error of this estimate is $15.31. It can then be said with 90-percent confidence that the actual median for this population group is $353 to $402, approximately $24 ($15.31 times 1.6) less than or greater than the estimated median. In other words, if 1,000 independent samples had been selected, 900 of them would provide an estimate that falls between $353 and $402.

16. See, for example, Sheme A. Kossoudji, "English Language Ability and the Labor Market Opportunities of Hispanic and East Asian Immigrant Men," Journal of Labor Economics, April 1988, pp. 205-228; and Evelina M. Tainer, "English Language Proficiency and the Determination of Earnings among Foreign-Born Men," The Journal of Human Resources, Winter 1988, pp. 108-22.
COPYRIGHT 1992 U.S. Bureau of Labor Statistics
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Meisenheimer, Joseph R., II
Publication:Monthly Labor Review
Date:Dec 1, 1992
Previous Article:Industrial Relations Systems, rev. ed.
Next Article:Working and poor in 1990.

Related Articles
Economic effects of immigrants on native and foreign-born workers: complementarity, substitutability, and other channels of influence.
Immigration and wage changes of high school dropouts.
Immigrant earnings performance in Canadian cities: 1981 through 2001.
Homeownership rates lower for migrants moving to emerging gateway cities.
Lumina foundation seeks to identify new paths that lead Hispanic immigrants to colleges' doors.
Gay immigrants put the I before the G.
Immigration policies in the European Union: European policy towards immigration from third countries is subject to certain restrictions which must be...

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters