In 2002, a court case ended decades of forced desegregation in Charlotte, N.C., schools. Districts were redrawn and, practically overnight, about half of the district’s students switched schools.
Some white students found themselves sitting beside more black and brown kids, while others ended up in super-white schools that reflected the residential segregation of North Carolina’s largest city.
The economists saw the forced desegregation as an opportunity to test two competing theories. Would white students’ views become more diverse as they came into contact with more diverse peers?Or would white students feel that their status was being threatened by their proximity to minorities and become more hostile? In this case, economists found, students’ views seemed to have evolved to be closer to those of their minority peers.
When the share of minorities grew in schools by 10 percentage points during the formative elementary and middle-school years, white students at those schools were 12 percent less likely to register as Republicans later in life.
What’s the link between exposure to minorities and falling Republican support? There’s no perfect link between race and party in the United States, but we can guess it’s driven by the same forces that lead black and Hispanic voters to prefer Democrats by large margins, as well as some Republicans to doubt the disadvantages minorities face in the United States.
For example, a 2017 Washington Post-Kaiser Family Foundation poll found that 63 percent of Democrats believed that black and Hispanic Americans “losing out” because of preferential treatment for whites was a bigger problem than vice versa. Two in 10 Republicans agreed, while more than 4 in 10 said that whites “losing out” due to preferences for black and Hispanic Americans was the bigger issue.
This link between educational segregation and political polarization has risen in importance as court-ordered busing has been overturned around the country and black-white segregation has, by many measures, increased.
“This re-segregation of the school system in the U.S. ties to current events, insofar as it is another form of structural racism faced by black people in our country,” said Carnegie Mellon University economist Kareem Haggag.
Students growing up in cities such as Charlotte now have fewer opportunities to interact with peers of a different color. That will likely accelerate white voters’ rightward drift.
“Partisanship as a phenomenon in U.S. political culture is an increasingly important dimension of understanding what’s going on,” said Eric Chyn, an economist at Dartmouth College.
Chyn and Haggag have known each other since they were research assistants at Yale University more than a decade ago. Now rising stars in academic economics, they seek new ways to detect how socioeconomic factors influence political behavior. Before this, they measured how people who were moved out of public housing in their youth showed stronger voter participation later on. Haggag was also part of a team that used smartphone data to find huge racial disparities in voter wait times in the 2016 election.
In this case, to measure how the end of integration changed the lives of North Carolina students in about 100 Charlotte-area schools, Chyn, Haggag and economist Stephen Billings of the University of Colorado used names and birth years to link the addresses and races of 35,988 students in third through eighth grades back in 2002 to their voter registration in elections through 2018, when students would have been, on average, in their late 20s. Their working paper was recently circulated by the National Bureau of Economic Research.
The team compared students who changed schools after the redistricting with equivalent peers within the same neighborhood (or census block group) who happened to live on the other side of a school boundary and likely didn’t transfer schools. They could then measure how many minority students each person would have encountered in school, then follow their voter registration to see what parties they supported.
Political affiliation is one of the few cultural outcomes the economists could measure using public records. They found that minority students saw no political impact from being assigned to schools with more minority students, while white students’ likelihood of becoming a registered Republican dropped 2 percentage points. That’s a 12 percent decrease from the 16.4 percent baseline.
They also found that the change in political support can’t be fully explained by changes in how likely a student is to attend college or register to vote.
A school’s racial composition could shape your future partisan identity through “friendships, peer and teacher role models, teacher and administrator instruction and discipline, as well as extracurricular activities and interaction outside school hours,” they write.
Mark Hoekstra, a Texas A&M University economist who was not involved in the study, said he was impressed by the authors’ use of voter-registration data to quantify hard-to-measure political outcomes.
In a related work published in the economic-policy edition of the American Economic Journal last year, Hoekstra and two collaborators analyzed random squadron assignments among young men in the U.S. Air Force Academy.
They found that for every additional black student who was assigned to a white student’s 30-person squadron in their freshman year, the white student became 23 percent more likely to choose a black roommate the following year.
The results held even though students were assigned to a different squadron each subsequent year, and even when results were explicitly limited to men who had never previously been in a squadron together.
Most interestingly, there was no significant change among white students from states with large black populations. A white student from a state with smaller black populations, on the other hand, was 30 percent more likely to want to room with a black student in the future, if he had more black students in his freshman squadron.