It turns out that there are also sizable differences in the rate of wage growth experienced by the different national origin groups in the United States (Boijas, 1995; Duleep and Regets, 1997a, 1997b; Schoeni, McCarthy, and Vemez, 1996; and Yuengert, 1994). Therefore, it is important to determine if the rate of wage convergence exhibits cohort effects: do the most recent immigrant cohorts experience either faster or slower wage growth than earlier cohorts? The existing evidence however, does not settle this issue conclusively. Duleep and Regets (1997b) argue that more recent waves, who have lower entry wages, will experience faster wage growth in the future, while Boijas (1995) and Schoeni, McCarthy, and Vemez (1996) do not find any evidence of cohort effects in the rate of wage growth.
This paper presents a theoretical and empirical study of the rate of economic progress experienced by immigrants. The study uses a human capital framework to motivate and guide the analysis. There seems to be some confusion about whether human capital theory implies wage convergence among the various immigrant groups, in the sense that immigrants who have high wages at the time of entry should experience slower subsequent wage growth. I show that a reasonable set of assumptions can easily generate investment behavior in the immigrant population that leads to wage divergence among groups, with the most skilled groups earning more at the time of entry and experiencing faster wage growth.

The empirical analysis uses the 1970, 1980, and 1990 Public Use Microdata Samples (PUMS) of the decennial Census. The empirical analysis of wage convergence in the immigrant population has much in common with the literature that estimates cross-country regressions to determine if there is convergence in per-capita income across countries (Barro, 1991, 1997; Mankiw, Romer, and Weil, 1992; and Quah, 1993). These studies typically find that the “raw” correlation between the growth rate in per-capita GDP and the initial level of per-capita GDP is positive, but weak. There is, however, a strong negative correlation between growth rates and initial levels of per-capita income when the regression controls for measures of the country’s human capital endowment. The data, therefore, reveal “conditional convergence” in per-capita income, in the sense that countries that start out with the same human capital endowments will tend to have the same per-capita income levels in the long run.