On Tuesday, my colleague Daniel Pryor appeared as a witness on BBC Radio 4’s The Moral Maze to discuss the causes of gang violence. He made a persuasive case that ending the prohibition of recreational drugs such as cannabis would reduce gang violence just as the end of the prohibition of alcohol reduced gang violence in the US.

After Daniel deftly dealt with Melanie Phillips objections, another questioner Matthew Taylor, a former Head of the Number 10 Policy Unit under Tony Blair, suggested that it was competition between schools that caused gang violence. He said that schools were excluding difficult pupils and they were ending up in the hands of gangs as a result.

Putting aside whether this is a case of correlation and not causation, after all the same pupils might have been excluded in the first place because they were violent, the evidence simply doesn’t show school choice leads to more crime. In fact, it’s the opposite.

One study by Corey DeAngelis and Patrick Wolf looked at the Milwaukee School Voucher program. They matched 1,089 Milwaukee students who used the vouchers to attend private school with 1,089 similar public school students on the basis of grade, race, gender, English language learner status, baseline math and reading test scores and neighbourhood.

They then checked the Milwaukee criminal record database for the students names when they were between 22 and 25 years old. The results were impressive. They found “attending a private school through participation in the Milwaukee voucher program was associated with a 79 percent reduction in the likelihood that a given student would commit any felonies and a 66 percent reduction in the likelihood that a student would commit any misdemeanours.”

There’s evidence too that when voucher programs are tried that state-run schools tend to get better as well in response to the competition. According to a review by Dennis Epple, Richard E. Romano, and Miguel Urquiola “evidence on both small-scale and large-scale programs suggests that competition induced by vouchers leads public schools to improve”.

Another study looking at Charter Schools (the American equivalent of Free Schools) by Roland Fryer and Will Dobbie found that winning a lottery to attend a charter school in Harlem had massive crime-reduction benefits, reducing the risk that a male student would be incarcerated by 4.4 percentage points.

A further study by David Deming also looked at lottery winners who got to choose the school they wanted (this time in North Carolina). He too found that letting parents pick the school their children go to lead to significant falls in criminal activity.

It’s likely that gang violence has many causes, but forcing schools to compete by giving parents choice over where they study simply isn’t one.