THE IMMIGRANTS: Summary Statistics

Using an adaptation of the Roy Model, he argues that the correlation of country of-birth effects with various characteristics of the source countries is consistent with the view that the “quality of immigrants” has declined and that immigrants do not achieve parity with the native-born over time. He traces the declining cohort quality to the shift in immigrant source countries that occurred after entrance criteria were revised in the 1965 Immigration Act.

In this study, we re-evaluate the evidence on changes in immigrant earnings and immigrant characteristics over time. Our analysis is similar in spirit to LaLonde and Topel (1992) wrho note that most immigrant/native-born comparisons assume “constant period effects”; to w7it, if, for example, immigrants have always been concentrated in the bottom tail of the wage distribution, and the wage distribution widens, then the immigrant native-born comparisons will show that immigrant’s earnings have declined. This will be the case even with no underlying change in the characteristics of immigrants vis a vis the native-born.

LaLonde and Topel recommend comparing immigrants to native-born workers who would have experienced similar wage changes due to the overall change in the distribution of earnings, for example, native-born Hispanics. Losing this comparison group, they show substantial assimilation of immigrants. Yuengert (1994) makes a related point demonstrating that the point in the distribution where earning comparisons are made has implications for immigrant/native-born w’age comparisons that are different than those that arise from computing simple means.

Our paper is not specifically about assimilation of immigrants – the growth of wages with time in the United States. However, the literature on assimilation and the changes in immigrant skills are inextricably linked, since in cross-sectional data, immigrants from different cohorts will have been in the United States for different lengths of time. In order to avoid making “longitudinal’’ inferences from cross-sectional information, we focus on the wages of recent immigrants – those who have been in the U.S. for 5 or fewer years. We adapt techniques appropriate for analysis of the entire distribution of wages to disentangle two influences on comparisons of native-born and immigrant wages between I960 and 1990: 1) the enormous changes in the structure of wages 2) the striking changes in immigrant characteristics.

After describing the data we use, we summarize the average characteristics and average wages for native- born, immigrants, and recent immigrants for the period I960 to 1990. payday one loans