Identifying Ethnic Enclaves Using Linked Employer-Household Data

Liliana Sousa, U.S. Census Bureau

Using the unique data available through the U.S. Census Bureau’s Longitudinal Employer-Household Dynamics (LEHD) program, I identify individuals as part of an ethnic enclave economy based on two dimensions: their neighbors and their coworkers. I create and analyze measurements of immigrant enclaving proclivity based on both residential and employment clustering behavior. These measures of immigrant clustering show that, even among the largest immigrant groups living in five of the biggest immigrant population centers in the U.S., few immigrants are part of ethnic enclaves. I also measure the proportion of immigrant segregation that is attributable to observable demographic characteristics found in public-use data sets. I find that these characteristics combined with geographic area controls explain about half of the variation in residential own-exposure rates and a quarter of workplace own-exposure rates.