I have to tell you, as someone who tries to follow this stuff very closely, it is very difficult, because you're working with so many different data sources. I appreciate that question.
I would argue that the monitoring and assessment report is probably the best place for information on what's happening, because it's the only place you can try to get consistent data. When I've gone back and tried to look at the annual reports that the various provinces are submitting to Ottawa, it's frankly been hard to tell what the outcomes are.
I think the LMA evaluation that you received, or that you have probably seen, which was done last year on the LMAs—and it's a different population from the one you would see in the LMDAs—gives you a very good indication, which I think is a good starting point for analysis, about the effectiveness of different kinds of interventions. This is where we really don't know a lot about the long-term effectiveness of different kinds of programs and different kinds of services, because they vary.
The benefit of the way the LMDAs and the LMAs are structured is that you kind of have a choice from among five or six different kinds of interventions that you can use. Over time that should allow you to do some experimentation and analysis about what works and what doesn't. I think that LMA evaluation report is a very significant contribution to the literature in that it tells us, basically, that skill development is very important.
What B.C. is doing with their new centre for employment excellence, with respect to workplace practices or employment practices, is a very important starting point for qualitative analysis with stakeholders on how different populations find their way into work.
I think it would be really good for ESDC to continue the work they've started on longitudinal analysis on their data sets, because that's what's going to begin to tell us, for programming design across the country, what provinces should be focusing on.