Citation needed. Here, let me Google "prison recidivism non-violent" for you -- a 1997 study which suggests that only about 15% of nonviolent offenders are returned to prison within three years for violent offenses [PDF]:

http://www.ojp.usdoj.gov/bjs/pub/pdf/pnoesp.pdf

You're proving the parent's point?

Isn't 15% is about what you would expect if they were randomly convicted first of a nonviolent offense?

Maybe in America. In my corner of the world it would be way under 1%.

I wouldn't be surprised if it were lower, but I would love to see some real statistics in play.