Occupational Disparity in a Migrant Metropolis: A Case Study of Atlanta

As the populations of immigrant and migrant groups swell in Atlanta, Georgia, the nature of race and ethnic relations in the metropolis, one historically dominated by a black-white dichotomy, is changing. In the 2000 census nearly 40% of the Atlanta metropolitan area's working-age population was an immigrant, a return migrant, or a northern-born migrant. We use the 5% Public Use Microdata Sample (PUMS) from the 2000 census to examine over- and under-occupational representation in the metropolitan area among migrant and nonmigrant groups. The results indicate that there remains a racial and ethnic hierarchy that is related to occupational queues and segmented labor markets. By comparing different types of migrant groups, and by taking into account race, ethnicity, and nativity, we more clearly delineate economic stratification in the metropolitan region. This paper was partially supported by a grant to the Center for Social and Demographic Analysis at the University at Albany, SUNY, from NICHD (R24–HD0449430) and by a Georgia State University Research Initiation Grant. We thank Lesley Reid, Jim Baird, and Glenn Deane for their assistance, and Greta Gilbertson and three anonymous reviewers for their critiques of an earlier version of this paper. This paper was presented at the 2005 annual meeting of the Population Association of America. Notes For simplicity, we use the term “northern-born” to refer to migrants born in any nonsouthern region of the United States. While we recognize the importance of studying the enclave sector, we focus on the primary and secondary labor markets because we are analyzing the occupational distribution for a number of native-born groups as well. We do not include those foreign-born persons who were born outside of Latin America or Asia (e.g., foreign-born Europeans) because Latinos and Asians have dominated the international migratory streams into the U.S. since 1965 (Heer Citation1996). We could further delineate this group into intrasouthern migrants. That is, many “nonmigrants” may, in fact, have migrated to Atlanta from outside of Atlanta but from within the South; in fact, this would show an even greater demographic impact of migrants on the area. We choose to remain focused on broadly-defined regional distinctions for our migrant definitions. Similarly, there is variation within the immigrant groups (e.g., Mexican immigrants), but we prefer our more parsimonious grouping. We do not control for English language ability and length of residence in the U.S. because we are comparing multiple types of internal and international migrants. However, in supplementary analyses we did estimate a separate equation for the foreign-born groups, with foreign-born Asians as the omitted category. There are no surprising results: The coefficients for English language ability and length of residence are statistically significant, but, most importantly, including these controls does not change the result that Hispanics are much less likely to be found in a primary sector occupation compared to Asians. Source: Lewis Mumford Center for Comparative Urban and Regional Research, University at Albany, SUNY (www.albany.edu/mumford). Notes: All groups are nonHispanic with the exception of native-born and foreign-born Hispanics. Adjusted weights are used in this analysis (see Footnote #6 for more details). Source: 5% Public Use Microdata Sample, Census 2000. Authors' calculations. All tables, except for Table 1, include weighted figures based on the “person weight” variable which allows us to produce estimates of 100 percent characteristics from the PUMS files. However, in order to maintain the sample size but properly implement tests of statistical significance, we multiply the sample weight provided by the PUMS by the ratio of the sample size to the sum of weights (i.e., wi∗ (N/Σ(wi))) and use the adjusted weight in the analyses. Notes: All groups are non-Hispanic with the exception of native-born and foreign-born Hispanics. Adjusted weights are used in this analysis (see Footnote #6 for more details). Source: 5% Public Use Microdata Sample, Census 2000. Authors' calculations. Notes: All groups are non-Hispanic with the exception of native-born and foreign-born Hispanics. Adjusted weights are used in this analysis (see Footnote #6 for more details). Bolded odds ratios are over- (≥ 1.5) and under- (≤ .67) represented occupational categories, and statistically significant at p ≤ .05 or less. All other odds ratios are statistically significant at p ≤ .05 or less, except for those that are italicized. Source: 5% Public Use Microdata Sample, Census 2000. Authors' calculations. We also tested an equation in which multiplicative terms between each migrant group and gender, and between each migrant group and education (measured dichotomously: college or graduate degree versus noncollege degree), were computed. For education, not surprisingly, the significant terms show that college educated individuals are more likely to be found in a primary occupation than noncollege educated persons across the groups. No clear pattern emerges for gender, but for the most part, the significant terms suggest that women are more likely to work in a primary occupation than men. Even though we place “Sales and Related Services” and “Office and Administrative Support” jobs in the secondary occupational category, our results hold for women across most of the groups. Supplementary analyses suggest that this may be explained by the fact that jobs like teachers and nurses—in which women are overrepresented (Reskin and Roos Citation1990, p. 5)—are defined as in the primary sector in this analysis. Notes: The reference group for the race/ethnicity/nativity migrant group is white southern-born nonmigrant and, for education, the reference group is high school graduate or less. Adjusted weights are used in the logistic regression analysis (see Footnote #6 for more details). †Eleven observations have zero weight; therefore, the sample size is reduced from 85,450 to 85,439. The predicted probabilities were derived using the following formula: Px=1/(1+e−L), where P is the predicted probability of primary occupation for group “x,” “e” is the mathematical constant, and “L” is the predicted log-likelihood of primary occupation obtained from the regression equation in Table 5.