Segmented Assimilation Among Mexicans in the Southwest

Abstract

This article examines segmented assimilation among foreign-born and U.S.-born Mexicans. Using the 2000 census, this article investigates how immigrants' length of residence in the United States and nativity affect the earnings and self-employment outcomes of low- and high-skilled Mexican men and women in the Southwest. Findings reveal that the earnings of low-skilled, foreign-born Mexicans decrease as immigrants reside in the United States longer and are generally lower among the U.S. born than the foreign born. In contrast, the earnings of high-skilled, foreign-born Mexicans increase as immigrants reside in the United States longer and are generally higher among U.S.-born Mexicans than foreign-born Mexicans. Moreover, self-employment participation decreases as immigrants reside in the United States longer and is lower among the U.S. born than the foreign born, regardless of skill. Since self-employment results in lower earnings, a decline in self-employment indicates economic progress. Furthermore, men are generally better off than women. Drawing from segmented assimilation theory, findings support the “downward assimilation” hypothesis among low-skilled Mexicans and the “Anglo-conformity” hypothesis among high-skilled Mexicans. Overall, this research provides evidence of intragroup differences in segmented assimilation among foreign-born and U.S.-born Mexicans in the Southwest.

ACKNOWLEDGMENTS

I would like to thank Sheldon Danziger and several anonymous reviewers for their helpful comments and Andrew Yinger for editorial assistance. This article was funded with fellowship support provided by the Ford Foundation, Research and Training Program on Poverty and Public Policy at the University of Michigan.

NOTES

Notes

¹ Five states are considered Southwestern states: California, Texas, Arizona, New Mexico, and Colorado (see Saenz and Davila 1992; Warren 1996).

² Segmented assimilation theory includes a third hypothesis that does not characterize the Mexican-origin population. The “ethnic cohesion” hypothesis (Portes and Zhou 1993) posits that the maintenance of the ethnic community provides an additional, non-assimilatory path of socioeconomic assimilation. This pattern is associated with entrepreneurial ethnic groups (Zhou 1997:979), such as Cubans and Koreans (Wilson and Portes 1980; Light and Bonacich 1988; Portes and Jensen 1992; Portes and Stepick 1993). Since Mexican self-employment is low, the assumptions of the “ethnic cohesion” hypothesis are not supported. That said, whether Mexican self-employment is associated with economic progress is assessed here.

³ In addition, recent immigrants may outperform their intermediate and long-term counterparts and U.S.-born Mexicans. Since the downward assimilation hypothesis posits a decline in socioeconomic outcomes over time, the logic here is that recent Mexican immigrants are less affected by the negative context of reception than those who have been in the United States longer.

⁴ The 2000 census allows for reporting more than one race. Therefore, Hispanics may identify not only as “non-Hispanic white,”“other race,” or “Spanish write-in,” but as one or more of these categories. If a respondent identified as one or more of these categories and also identified as “Mexican” in the Hispanic origin question, the respondent is captured in this sample.

⁵ In 1970, 1980, and 1990, the census asked respondents when they “first came to stay,” which might refer to the first or last migration to the United States or the cumulative number of years in the United States (Massey et al. 1987). In 2000, this question asks when the respondent came to live in the United States; this is more readily interpreted as the cumulative number of years residing in the United States.