We are getting slightly off the point, but I will respond to my hon. Friend’s intervention. The statistics do not suggest that. They suggest two things. The first is that people should perhaps have longer sentences, for which the reoffending rate is lower, not that they should have no sentences at all. The high reoffending rate for short sentences is an argument for longer sentences, not for no sentences.
The second point is that, in the main, someone has to have committed many offences to get to prison. If someone goes to court with more than 100 previous convictions they are more likely not to be sent to prison than to be sent there. People have community sentence after community sentence, and the only reason they go to prison is that those community sentences have not worked—they have not prevented them from reoffending. The reoffending rate for that cohort of people in prison, therefore, is lower than for those people when they were on community sentences.