Immigration and wage changes of high school dropouts

Monthly Labor Review, Oct, 1997 by Maria E. Enchautegui

From 1979 to 1989, the earnings gap between immigrant and native high school dropouts remained virtually unchanged, but it increased substantially in areas of high immigration; the growing proportion of immigrant high school dropouts explains some of the earnings change

One of the most dramatic economic changes of the last 20 years has been a stagnation of earnings. In this regard, low-skilled workers fared worse than any other group. The hourly wage gap between white high school dropouts and white college graduates went from 23 percent in 1980to 36 percent in 1990.(1) Among the factors that have been held responsible for the drastic deterioration in the relative wages of low-skilled workers are technological change and international trade.(2) The wage trends in the 1980s are also characterized by a growing disparity by ethnicity, such as the growing wage gap between whites and African-Americans, and between Latinos and non-Latinos.(3)

During the period of decline in wages of low-skilled workers, immigration increased precipitously. The number of immigrants ages 18 to 55, not enrolled in school and with less than a high school diploma, increased from 2.8 million in 1980 to 4.5 million in 1990, leading some to suggest that immigration might be a factor behind the decline. Research on inequality shows that the largest increase in inequality took place in the West and that this differential can be attributed, in part, to the entrance of a large number of low-skilled immigrants into the labor force.(4) Outside the West, the large drop in the low-skilled labor supply partially offset the decline in wages resulting from structural change.(5) One source attributes 40 percent of the decline in the wages of low-skilled natives to immigration and trade combined.(6)

Immigrants can affect wages in two ways. One is compositional: low-skilled immigrants may decrease the wages of low-skilled workers because immigrants have, on average, lower wages than natives. The inability to speak English fluently, very low levels of education, and little familiarity with the U.S. labor market tend to depress the wages of immigrants. The second way immigrants affect wages is behavioral: by increasing the supply of low-skilled workers, immigrants may decrease the wages of other low-skilled workers. In trying to evaluate the effect of immigrants on the wages of low-skilled workers, these two factors are seldom differentiated. And there are other factors: obviously, wages of low-skilled workers could decline for reasons unrelated to immigration, such as industrial changes, technological changes, and international trade.

This article traces the 1979-89 earnings of foreign-born and native-born persons without a high school diploma, in order to identify trends in the wage standing of immigrants relative to natives, as well as assess the compositional impact of immigrants on the wages of low-skilled workers. Throughout, an eye is kept on how wage differentials and the compositional impact of immigration on wages vary across areas with different levels of immigration. Data are from the 1980 and 1990 Census of Population. A low-immigration area is composed of metropolitan areas in which the foreign-born population arriving between 1980 and 1990 accounts for up to 5 percent of the total population. In medium-immigration areas, the recent immigrant arrivals account for 6 percent to 9 percent of the total population, and in areas of high immigration, the foreign-born population makes up 10 percent or more of the population. Because the article examines wages for natives and immigrants, data are presented for selected metropolitan areas, including areas that have very few immigrants, but that are important labor markets for native-born workers, especially minorities.

The reason that metropolitan area data are used is because metropolitan areas are often utilized as a proxy for local labor markets. Due to the economic integration of the different units that compose the metropolitan area, metropolitan areas can serve as a proxy for local labor markets better than each separate component can. By looking at metropolitan areas, the article presents a comparative picture of immigrants across U.S. local labor markets that would not be possible if aggregate figures were used.

National and local wage changes

Table 1 shows percent changes in real annual earnings from 1979 to 1989 for natives and immigrants without high school diplomas and who worked during the previous year. The information is presented for workers inside and outside of the largest 25 metropolitan areas, as well as for those in areas of low, medium, and high immigration. The last column of the table shows the 1980 unweighted sample size for each area. The 1979 earnings are constant at 1989 prices, and the Consumer Price Index was used to convert earnings into real figures.(7) The decline in wages of low-skilled workers is evident in these data. Between 1979 and 1989, the annual earnings of those without a high school diploma fell by 13 percent nationally. Workers outside the largest 25 metropolitan areas fared worse than those residing in these areas. Areas of high immigration showed better wage performance than did areas of low and medium immigration: wages declined by less than 8 percent in the former and by more than 14 percent in the latter.