There's probably actual research on this, but I can't come up with the right search terms. It could be one of those situations where it's impossible to say for sure. You can't exactly design an experiment around it: the decline already happened. So you'd have to do some kind of comparative analysis to figure out how outcomes would have changed without a union. Unions were never a 100% thing, so it might be enough to look at how non-union shops in the same markets fared.
Yes and no you have to be better than the median in all categories some of which you have no control over. For example you need to work in a better than median group for a better than median boss in a better than median division with better than median experience.
