We have a comprehensive ranking system that applies to many, but not all economic programs. That needs to be updated. There was meant to be a rolling update process, but when it was originally created, economists at StatsCan, I believe, looked at earnings of immigrants up to that point and asked, “What factors seem to affect earnings performance?”
You could go back and do that again. If you find that French language fluency raises earnings appreciably in Vancouver, that should be part of the process, but it should be a data-driven exercise.
What a skill is has become controversial. What economists think about human capital is usually measured in terms of earnings. I would suggest that we could take a lot of the politics out of it by agreeing on what the policy goals are—and then let the data drive it.