*Names in bold indicate Presenter
As the Latino population has grown from 8 million to more than 45 million persons between 1970 and 2010, average levels of Latino isolation have risen substantially, surpassing average levels of African American isolation in 2010 (De la Roca, Ellen and O’Regan 2014). Research on the effects of segregation on African Americans has found that higher levels of metropolitan area segregation have a significant connection to low birth-weight outcomes among African-American women (Ellen 2000), to gaps in black-white educational achievement (Card and Rothstein 2007), and to disparities in high-school graduation rates, employment levels, and earnings (Cutler and Glaeser 1997). The effects of segregation on socioeconomic outcomes for Latinos, however, have not been as extensively studied.
There are reasons to believe that the effects of segregation may be different for Latinos than for African Americans. Existing research suggests that as compared to the drivers of black-white segregation, white avoidance generally plays a lesser role, while income differences and preferences for clustering with those who share the same language and customs appear to play a larger role (Bayer et al. 2004).
This paper tests this hypothesis by examining the effect of metropolitan area levels of Latino segregation on the educational attainment, English proficiency, employment rate, earnings, and single-motherhood of young Latino adults. To address the concern of Latinos sorting across neighborhoods, we examine segregation at the metropolitan area level rather than the neighborhood while we restrict the analysis to young, native-born Latinos. To address potential endogeneity problems, we compare the association between segregation and outcomes for Latinos and for whites within the same metropolitan area. To further address potential reverse causality problems, we instrument for Latino segregation in 2010 using metropolitan area black-white segregation in 1970, the spatial distribution of older/newer and single-family/multifamily housing in 1970, and a shift share approach predicting the share of the metropolitan population that is Latino in 2010 based on historic stocks of Latino ancestry groups (Card 2001).
Our preliminary results show that residential segregation is highly correlated with crucial measures of Latino wellbeing such as the likelihood of high school graduation, employment, and single motherhood, as well as earnings levels. Although the effect of segregation on high-school graduation rates and English proficiency has declined since 1980, the effect of segregation on college graduation rates has increased and the substantial impact on earnings has remained steady.
