![Sample our Urban Studies journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110826/TOC-Subject-Sample-2021-Urban-Studies.jpg"})
Abstract
This article extends research on intermetropolitan and regional wage inequality through an investigation of the uneven geography of racial and ethnic wage inequality across metropolitan labor markets. Prior geographic studies largely restricted analysis of the source of intermetropolitan wage disparities to differences in industrial structure. The study described in this article further expands the analysis of labor market effects by conceptually describing and empirically analyzing the effects of three significant racial labor market institutions: public employment, unionization, and the penal system. I investigate these effects as part of a more extensive analysis of how local labor market structure—comprised of industrial mix, demographic composition, and institutional and regulatory arrangements—matters in mediating racial wage inequality. I use data from the 2000 U.S. Census and multilevel methods to analyze the wage differentials of African American and native- and foreign-born Latino men relative to whites across 186 U.S. metropolitan areas. Local labor market structure mediates different types of racial wage inequality in distinct ways: Regulatory context matters most for the relative wages of African Americans; both regulatory context and industrial mix influence the relative wages of native-born Latinos; and industrial composition matters most for the relative wages of foreign-born Latinos. Against these broad patterns of difference, not all effects—especially regulatory and institutional effects—are singularly group specific. Unionization shores up wages for both white and black high school–educated workers and mitigates racial wage inequality. Higher minimum wage rates boost the relative wages of high school–educated whites and native-born Latinos.
Este artículo amplía la investigación sobre la inequidad del salario inter metropolitano y regional mediante una investigación de la geografía irregular de la inequidad del salario racial y étnico en los mercados de trabajo metropolitanos. Estudios geográficos anteriores mayormente restringieron sus análisis de las disparidades de salario inter metropolitano a las diferencias en la estructura industrial. El estudio descrito en este artículo profundiza el análisis de los efectos del mercado de fuerza laboral describiendo conceptualmente y analizando empíricamente los efectos en tres instituciones significativas en el mercado laboral racial: empleo público, unionización, y el sistema penal. Investigo estos efectos como parte de un análisis más extensivo de cómo la estructura del mercado laboral local—formado por la mezcla industrial, composición demográfica, y arreglos institucionales y regulatorios- importa en la mediación de la desigualdad en el salario racial. Uso información del censo de EE.UU del 2000 y métodos multinivel para analizar las diferencias de salario de afroamericanos y nativos—y nacidos en el extranjero- hombres latinos en relación con los blancos en 186 áreas metropolitanas de EE.UU. La estructura del mercado laboral local media los diferentes tipos de desigualdad en los salarios raciales en distintas formas: el contexto regulatorio importa más para los salarios relativos de los africano americanos; el contexto regulatorio y la mezcla industrial influencian los salarios relativos de los nacidos nativos latinos; y una composición industrial importa más para los salarios relativos de los latinos nacidos en el extranjero. Contra estos grandes patrones de diferencia, no todos los efectos–especialmente regulatorios e institucionales- son singulares a grupos específicos. La unionización eleva los salarios tanto para los trabajadores blancos como para los negros que hayan terminado la educación secundaria y mitiga la desigualdad del salario racial. La tasa de salarios mínimos es más alta en los salarios de los blancos y de los latinos nativos que han terminado educación secundaria.
Acknowledgments
This research was based on work supported by the National Science Foundation under Grant No. 0525667. Special thanks to Nelson Lim for comments and to Benjamin Roth for assistance with maps.
Notes
1. I include men only to focus on differing patterns of race, ethnicity, and nativity. Because gender shapes the labor market to such a great degree (Parks Citation2010), it demands additional theoretical and empirical explication beyond the main focus of this article.
2. I utilize the largest metropolitan unit available in the census data to best capture metropolitan regional effects. If a metropolitan region is classified as a Consolidated Metropolitan Statistical Area (CMSA), I treat the CMSA as a single metropolitan area. All other metropolitan areas are metropolitan statistical areas.
3. The male wage earners included in my sample are in their prime working years, a standard sample selection procedure in wage analyses. Because the volatility of the youth labor market and more erratic nature of work among older workers can considerably skew wage distributions, these sample restrictions provide a more conservative estimate of levels of wage inequality.
4. Because I examine wages, not earnings, I do not include the self-employed. Self-employment operates under fundamentally different rules and regulations in the United States and represents an altogether different market segment from wage work. Employment regulations and labor market institutions, such as minimum wage rates, antidiscrimination laws, and unionization, do not apply to self-employed individuals. Further, theoretical accounts of the labor market, from neoclassical economics to Marxist theory, conceive of self-employment and non-self-employment in fundamentally different ways (e.g., self-employed individuals are also owners of their means of production and thus their earnings reflect a return on labor and capital; wages of non-self-employed workers reflect a return on labor only).
5. National origin is an important differentiating characteristic within the immigrant population yet might be less relevant than nativity and pan-ethnicity for wage studies if employers discriminate primarily on the basis of ascriptive characteristics and language. Given that the majority of immigrant Latinos in the United States is Mexican (65 percent of my sample) and the severe cell size restrictions that would be imposed by subsetting my sample by national origin, I limit my analysis to the pan-ethnic category of foreign-born Latinos.
6. A detailed description of map patterns is available on request from the author.
7. The unionization rate is the percentage of workers covered by a union contract. Unionization data come from the Current Population Survey and are compiled at the CMSA level by Hirsch and Macpherson (Citation2003). My labor market sample of 186 derives from this data set: I utilize the maximum number of labor markets from the Public Use Microdata Sample for which CPS data on unionization were available.
8. Following Johnson and Raphael (2009), I use the percentage of males residing in nonmilitary group quarters as a proxy for the male incarceration rate. See Raphael (2005) for a comparison of this census estimate with data from the Bureau of Justice Statistics that shows the suitability of this census measure as a proxy for the incarceration rate.
9. Several metropolitan areas cover states with different minimum wages or are located in states that changed their minimum wage part way through 1999. For these metropolitan areas, I calculate weighted averages to determine an annual minimum wage for the entire population. States with a minimum wage higher than the federal rate in 1999 include Alaska, California, Connecticut, the District of Columbia, Hawaii, Massachusetts, Oregon, Vermont, and Washington. For states that had lower minimum wages than the federal minimum, I applied the federal minimum wage. Information on state minimum wages comes from Nelson (Citation1999) and U.S. Department of Labor (2008).
10. FIRE employment includes census industry codes 687–726. New technology sectors include technical services such as engineering, computer systems design, and scientific research and design services (census industry codes = 729, 738, 739, 746). The following industries proxy for the degree to which the service sector within a particular metropolitan region is concentrated in lower paying sectors: administrative and support services, such as employment services, business support services, security services, landscaping, and waste management; child care; traveler accommodation and food services; other services except public administration, such as nail salons, dry cleaning, and private household services (census industry codes = 758–785, 847, 866–915, 929).
11. Full model results are available on request from the author.