I would like to point out that I've looked at one of the articles that Professor Blais has published, and I believe it's the one he's using to base his claims on. There were serious methodological errors defining the independent variable, the covariate included, and a lack of trend lines, and this invalidates, I think, his claims.
If you look at the overall homicide rate, you see that it declines, as I pointed out in my testimony. The rate fell rapidly before the introduction of Bill C-68, that is to say, it was put into effect in 1998-2001, and the long-gun registry was completed in 2003. If you use those break points, you can see that the homicide rate fell a lot faster before it was introduced than afterwards. So how it could have increased that when it obviously decreased is not clear at all.