Reading papers today on impact of school choice on crime and risky behavior. Not a huge number of well-designed empirical papers. But the good ones--the ones that examine lottery-based systems (and thus have natural experiments)--find significant reductions in crime, risky behavior, and teen pregnancy. One paper finds that winning the lottery and being sent to the first or second choice school predicts cutting students' expected days incarcerated and the expected social costs of their crimes in half.
https://www.facebook.com/jasonfbrennan/posts/10205427476881999?__xts__[0]=68.ARDWfpAnWgubtveqAIVVhjEI_3lNiLwx9iMSD4pY5_KY0cctTdFwNHxR6X99BFRpbFL7m-eT63x3jTp2o0rHxBYeIeyBsQeC_W_qAYbbRSoucoGIN7Z0q7gac38WnZ_l1t8zPzb5jJ7i_ZHvEv442_mg84LtCh6mWDSQYOwezFSVbYGfbq-a4PI3X6mS-67MG_HzWEZllzeB3Ep84MUvn2eaVyzlUtWOIY0xQeOK&__tn__=-R
Dobbie, Will, and Roland G. Fryer Jr. "The medium-term impacts of high-achieving charter schools." Journal of Political Economy 123.5 (2015): 985-1037.
