Presumably, persons who know the language would have an easier time adapting in the United States (although this effect could be attenuated by residential segregation). The 1970 U.S. Census, however, does not contain any information on English language proficiency, so that we cannot observe the initial language skills of the immigrants who arrived in the 1960s. I used the 1980 Census to calculate the probability that immigrants who arrived between 1975 and 1979 spoke English well or very well.
This statistic was calculated for each cohort by country of origin and age-at-arrival. The last two columns of Table 7 report the regression results obtained when one includes this variable into the model (and when the regression is estimated in the subsample of immigrants who migrated in 1975-79). English language proficiency at the time of entry has an independent positive impact on the rate of wage growth, but does not change the impact of most of the other variables in the model. The last column in the table shows that the impact of English language proficiency becomes insignificant if we control for the educational attainment of the cohort.
In sum, the empirical evidence shows that source country characteristics matter in determining both the entry wage and the subsequent rate of wage growth. Moreover, the same underlying factors that tend to generate higher wages also tend to generate faster wage growth.
In effect, the empirical results confirm that there is a positive correlation between the economic performance of immigrants at the time of entry and the rate of economic progress in the United States, so that the human capital production function for immigrants exhibits weak relative complementarity.