One of the original arguments for a government-run education system is that public schools are necessary for stable democratic society. After all, self-interested families might send their children to private schools that specialize in maximizing earnings rather than citizenship skills. But new evidence suggests that private schools are actually more conducive to maintaining social order than public schools. Here’s why.

The new study — coauthored by Dr. Patrick J. Wolf and me — used student-level data from the longitudinal evaluation of the longest-running private school voucher program in the United States. We found that students using a voucher to attend a private school in 8th or 9th grade were convicted of fewer crimes and were involved in fewer paternity suits than their public school peers by 25 to 28 years of age.

As shown in the figure below, students using the Milwaukee Parental Choice Program (MPCP) were found guilty of 60% fewer property damage crimes, 41% fewer drug-related crimes, and experienced 31% fewer paternity suits as young adults than their carefully matched peers in Milwaukee Public Schools (MPS).

But why?

Competition works in the market for education services. Families care about the civic and character educations their kids are getting at school each day. And, of course, private schools must cater to the needs of families if they want to stay in business. Public schools, on the other hand, hold significant monopoly power over their customers because of residential assignment and funding through property taxes. Private school choice programs could also reduce criminal activity by exposing students to peer groups and school cultures that discourage risky behaviors.

While our new study is the first to link a private school voucher program to adult paternity suits, four other studies have also examined the intersection between school choice and crime. As shown in the table below, all five studies on the topic have found that school choice reduces criminal activity. For example, an experimental study published in the Journal of Political Economy in 2015 found that winning a lottery to go to a charter school in Harlem Children’s Zone completely eliminated (a 100% reduction) the chance male students would be incarcerated and reduced the likelihood of teen pregnancy by 59% for female students.

The Effect of School Choice on Crime Reduction

Notes: Green cells indicate the study found statistically significant positive effects on crime reduction. “RCT” is “Randomized Controlled Trial.” “QED” is “Quasi-Experimental Design.”

The evidence on this subject is limited. Much more research needs to be done. But every existing study on the topic suggests that school choice has large positive effects on crime reduction. And two reviews of the evidence indicate that private school choice tends to lead to better civic outcomes. Maybe public schools aren’t necessary for a stable democracy after all?