The Effects of Immigrants on African-American Earnings: A Jobs-Level Analysis of the New York City Labor Market, 1979-89

Abstract

The improvement in the relative economic status of African-American Workers in the 1960's and 1970's was reversed in the 1980's, a decade that also featured a collapse in the relative (and real) wages of the last skilled (Bound and Freeman, 1992; Blau and Kahn, 1992; Levy and Murnane, 1992). At the same time, the U.S. experienced the largest absolute and per capita levels of immigration since the early part of the century. Significantly, this recent wave of immigrants was far less skilled, at least in terms of educational attainment, than earlier waves of immigrants in the post-war period. Friedberg and Hunt (1995) report that 43% of new immigrants did not possess the equivalent of a high school degree. And according to a recent study by David Jaeger (1995), in the 50 largest metropolitan areas employed male immigrants were about 16% of the civilian workforce with less than a high school degree in 1980; by 1990 this figure was over 30%. For women, this figure rose from 17% to almost 28%. Not surprisingly, there is a concern that growing numbers of immigrant workers have negatively affected the standing of African-Americans in urban labor markets. But with the exception of Borjas, Freeman and Katz (1996) and Jaeger (1995), the consensus in the research community appears to be that there has been little if any negative wage effects (see the surveys by Borjas, 1994; Friedberg and Hunt, 1995; and DeFreitas, 1996; National Academy of Sciences 1997). This is a rather surprising finding, since it requires a nearly instantaneous adjustment to labor supply shocks in local labor markets. Borjas (1994) terms this an unresolved puzzle. Indeed, it is particularly puzzling since the sharp growth in the supply of low-skill immigrants took place during a decade in which the power of labor market institutions to shelter low-skill workers from intense wage competition was severely eroded. In our view, the failure to find earnings effects from sharply rising supplies of low-skill foreign-born workers in increasingly deregulated labor markets may reflect the dominant research methodology, which has been to explore these effects with across-metropolitan tests. Since immigrants are overwhelmingly concentrated in a small number of urban labor markets, such as Los Angeles, New York, Houston, San Francisco, and Miami, we would expect wage effects to be concentrated in these same cities. In addition, the effects of increasing supplies of low-skill workers on wage outcomes are likely to be strongest in jobs that are unsheltered by unions, civil service rules, or craft- and firm-specific skill requirements. Indeed, if wage setting for low-skill workers takes place mainly at the job level (see Thurow, 1975), then the effects of a large increase in labor supply should be explored at the level of detailed jobs in specific metropolitan labor markets.