Since the late 1980s, immigrants to the United States have increasingly settled outside of traditional gateways, at times in destinations with no recent experience of immigration (Singer 2004, Marrow 2005, Waters & Jiminez 2005).  In these new destinations, the lack of pre-existing co-ethnic social networks raises questions about how immigrants will incorporate.  Employing new questions from the Current Population Survey (CPS), this paper investigates how immigrant civic participation varies with co-ethnic concentration for the largest immigrant ethnic group in the United States.  That is, do Mexican immigrants participate more amidst greater concentrations of Mexicans? 

Previous analyses have offered different theories about the role of ethnic concentration in immigrant participation and produced conflicting findings.  While some studies suggest that ethnic concentration fosters immigrant participation by reducing linguistic and cultural barriers (Ramakrishnan & Espenshade 2001, Ramakrishnan 2005), others argue that it hinders participation by isolating immigrants among resource-poor peers (Cho 1999, DeSipio 1996, de la Garza 1996, Espenshade & Fu 1997, Pearson-Merkowitz 2008).   Drawing on the insights of spatial assimilation theories, I offer an explanation for these conflicting accounts.  Spatial assimilation theory posits that while new immigrants rely on the resources of the ethnic enclave, longer-term immigrants can access additional opportunities by moving beyond the enclave (Massey & Denton 1987, Alba & Nee 1997).  Thus, I hypothesize that longer-term immigrants will participate more outside of traditional immigrant gateways.  Recent immigrant arrivals, on the other hand, will participate more in gateways because they rely on co-ethnic networks and institutions to overcome linguistic and cultural barriers to participation.

I analyze four types of Mexican/Mexican-American civic participation using models that interact first-generation immigrant status with state-level Latino concentration.  I focus my analysis on civic participation because unlike naturalization and voting, activities such as voluntarism, charitable giving, public meeting attendance, and community project participation are not barred to recent immigrants (Ramakrishnan & Bloemraad 2008).  The size and scope of the 2004-2008 CPS, with 30,092 Mexicans/Mexican-Americans located across every U.S. state, enables a truly national investigation of the effect of ethnic concentration on co-ethnic participation.  In line with predictions, I find that for Mexican-Americans in the second-generation and beyond, ethnic concentration is associated with declines in civic participation.  In the same analyses, however, ethnic concentration is associated with a differential boost in participation for first-generation Mexican immigrants.  The results hold when controlling for a range of potentially confounding individual and contextual predictors of participation.

Of course, cross-sectional analyses of contextual effects must contend with the possibility of selection effects.  While selection effects present a potentially compelling explanation for why Mexican-Americans in the second-generation and beyond participate more outside of ethnic enclaves, they do not as convincingly explain why first-generation Mexicans would participate less in the same places.  For first-generation Mexican immigrants, settings with a greater concentration of co-ethnics appear to support civic participation.  Future analyses remain necessary to sort out whether ethnic concentration itself or factors associated with it, such as institutional and legal context, play the crucial role in supporting early immigrant civic participation.