I understand that, and I think it's a prudent thing to do. If you were to assume as a hypothetical that the correctional investigator is correct and that other witnesses are correct and that the experience of every jurisdiction in the world that has tried this thing is correct and that we're going to see higher rates of recidivism, if you were to plug into your model higher rates of recidivism, would that result in a significantly higher cost figure?
On February 17th, 2011. See this statement in context.