With respect, that's not what the chart shows. What the chart shows is changes in unemployment rate and changes in robbery. What we're saying is, regardless of where you are in Canada, regardless of the overall wealth, where you see changes in unemployment rates you see—in fact, very quickly following—changes in robbery.
This is data that was produced by Statistics Canada. The point of putting it here is to show that factors such as these have a very clear and dramatic parallel—between these two factors—whereas you cannot find any similar parallel with sentencing.
I was not suggesting that unemployment was the sole cause of crime, however; it's much more complex than that. Within given communities, there are all sorts of other factors that can compensate for things such as unemployment. For instance, disparity of income appears to be far more important than rates of income. But overall, when you look at the trends, there seems to be a correlation between the two.