Changing Patterns in the Assimilation of Immigrants

Most immigrants arriving after the 1980s had a smaller rate of economic assimilation than those who arrived earlier.

In The Slowdown in the Economic Assimilation of Immigrants: Aging and Cohort Effects Revisited Again (NBER Working Paper No. 19116), George Borjas finds a cohort effect not only in the level of immigrant earnings, with more recent immigrants having generally lower entry wages than immigrants did before the 1980s, but also in their rate of earnings growth, with more recent immigrants having a smaller rate of economic assimilation compared to earlier immigrants. He suggests that this slowdown in wage convergence reflects a decline in "human capital accumulation" tied to a decline in the rate at which more recent immigrants are acquiring English language skills.

The issues involving immigrant assimilation, human-capital accumulation, and wage convergence have been widely studied for decades. Initial studies used cross-sectional data to compare the age and earnings profiles of immigrants and natives, and often found a rapid rate of wage convergence with native workers. The difficulty with this approach is that cross-sectional comparisons may mis-state the true rate of assimilation when there are substantial differences in earnings potential across immigrant cohorts. The current study uses data from the 1970 through 2010 Censuses to examine the evolution of immigrant earnings, and it focuses on immigrants who were 18 or older when they arrived in the United States.

The analysis finds different levels of earnings between those immigrants who arrived in the United States before and after the 1980s, but it also reveals that most immigrants arriving after the 1980s had a smaller rate of economic assimilation than those who arrived earlier. For immigrants who entered the country before the 1980s, their initial wage disadvantage compared to natives typically narrowed by around 15 percentage points during their first two decades in the United States. In contrast, the immigrants who entered the country after the 1980s have a much lower rate of wage convergence, and the evidence suggests there has not been any economic assimilation at all for immigrants who entered the United States in the 1990s.

The author considers three factors that might have contributed to the slowing assimilation of immigrants over time: changes in U.S. macroeconomic conditions that affected immigrant and native wage structures differently; changes in the national origin composition of the immigrant population; and changes in the geographic settlement pattern within the United States of more recent versus past immigrants. He concludes that "the data convincingly show that none of these factors can account for the severe decline in the rate of assimilation." The data instead suggest that part of the decline in assimilation appears to be connected to a discernible decline in the rate of human capital accumulation among recent immigrants. Specifically, immigrants who entered the country prior to the 1980s typically experienced a 15 percentage point increase in their fluency rate during their first two decades, while the cohorts who entered the country after the 1980s show only a 7 percentage point increase.

The study focuses on one factor that seems to explain part of the decline in the rate of economic assimilation and human capital accumulation: the growth in the size of the national origin groups from which recent, as opposed to historical, immigrants are drawn. The rate of increase in English language proficiency is significantly slower for larger national origin groups. This effect accounts for about a quarter of the concurrent declines in the rate of economic assimilation and the rate of human capital acquisition. Data from a number of immigrant groups -- including Chinese, Filipinos, Cubans, Mexicans, and Indians -- suggest that all of these large national origin groups exhibited a decline in the rate of assimilation between the cohorts who entered the United States in the late 1980s and the late 1990s. The author notes that the payoff for immigrants to learn English is likely tied to the frequency with which they use their language skills in everyday activities, and that the incentive to learn English is likely to be lower when immigrants find a large, welcoming ethnic enclave in the United States.

--Jay Fitzgerald