The libertarian’s freedom-coercion heuristic says that working at a sub-minimum wage cannot possibly be a bad thing if someone agrees to do so. For progressives, aside from the fact that this is not their preferred heuristic, the problem is that working at a sub-minimum wage is not, in Michael Munger’s terms, euvoluntary. That is, it is only voluntary because your best alternative (not working at all) is so bad.Caplan is asking a sophisticated version of Andrea's Question: we know we have an analytically rigorous case that minimum wages are on net harmful and that the (often hidden) consequences of pricing people out of the labor market should outweigh people's moral objections, so perhaps what we should seriously consider is the way we package our rhetoric, to tune our arguments more carefully to the actual audience we have instead of a room full of like-minded folks. This, I believe, is precisely what Munger was doing in his discussion of Locke's Venditio. It is, I believe, what anyone who is writing pop econ should be doing, and perhaps something I need to hone in my own writing.
Additionally, Kling's approach of using the three axes approach to moral intuition seems useful for euvoluntarists. I've been kicking around some responses, but I don't quite have anything ready yet. I think there might be a good fit for the exchange factory metaphor I posted about in my explanation of meta-euvoluntarity, so I suppose you might look forward to that.