The experience we've had in Canada, particularly as documented by the Correctional Service of Canada, is that the process of reintegration through supervision of federal offenders over the last 20 years has shown a continuous improvement in recidivism rates. There is good reason to believe that supervision in the community has much greater effect, in terms of reintegrating a person, than simply locking them up and releasing them.
In the data you've received, for instance, you'll notice that the recidivism rate for those on conditional sentences was 17% over the next year, whereas with prison it's 30%. But that's the year following the completion of the sentence. As you point out, many people on conditional sentences are serving 700 days in the community before their sentence is even completed. So there's good reason to believe that conditional sentences are very effective.
We're also aware of the fact that people being selected for conditional sentences are selected, in part, because they're considered to be low risk. So we have to put that into perspective.
But in the end, if we're trying to determine what is the best cost-benefit situation for reducing crime, then it's very difficult to make a case for imprisonment over conditional sentencing—for the cases that are receiving conditional sentences.