Aside from the other problems that have been pointed out, this would provide an incentive for prisons to favor groups that have been statistically linked to l

Aside from the other problems that have been pointed out, this would provide an incentive for prisons to favor groups that have been statistically linked to lower recidism rates -- not only could it create some kind of favoritism, it would also probably favor those who perpetrated more serious crimes (thieves have by far the highest recidivism rates, much higher than, say, murderers [1]). The American penitentiary system is fundamentally flawed and is in need of a deep reform; the real shocker to me is that stuff like this can happen before trial: > Nazario first went to Rikers at 16, after an arrest on an assault charge. Before leaving at 19, he says, he had spent more than 300 days in solitary confinement — all before being convicted of a crime. (on a side note, I was surprised to learn Rikers is actually a jail and not a prison) [1] http://www.bjs.gov/index.cfm?ty=pbdetail&iid=1134 http://www.bjs.gov/index.cfm?ty=pbdetail&iid=1134