Scholars have debated whether students' enrollment in private schools changes levels of racial segregation across urban school districts. The authors examine this issue by comparing the actual racial composition of schools with the racial composition of school-aged children living in the corresponding attendance areas. They do so by linking maps of school attendance boundaries with 2000 census data, the Common Core of Data, and the Private School Survey for the 22 largest U.S. school districts. The results show that public schools would be less racially segregated if all children living in a school district attended their neighborhood schools. In addition, private, magnet, and charter schools contribute to overall racial segregation within many school districts. The effects are particularly striking for segregation between white and Hispanic children. Finally, a few school districts with desegregation policies have succeeded in reducing racial segregation. The analyses contribute to debates regarding recent proposals to eliminate desegregation programs while simultaneously expanding "free-market" educational reforms that promote students' mobility across public, private, and charter schools.