Many of you say paternalism is to protect the very stupid. Would you support Would-Have-Banned stores with an [sic] min IQ rule? If so, what would that IQ be?The point I’m about to make is actually clear in the original post by Hanson, to which the above question was merely an addendum in response to comments. But since the IQ-threshold concept seems to be getting the most attention from other bloggers, the point is worth making again. The main reason new paternalists probably won’t support that proposal is they think we’re all stupid. Okay, that’s not really true. What they actually think is that all of us, stupid or not, are to some degree cognitively biased in ways that reduce our well-being. We all have problems with self-control, we all are subject to confirmation bias and hindsight bias, etc., etc. While the extent of these biases may be correlated with IQ, it’s still just a correlation, as even very smart people can be afflicted by cognitive bias. Since the new paternalists aim to “help” all of us who need it, high-IQ-only stores won’t do the job.
Still, Robin’s (possibly tongue-in-cheek) proposal is a good question to pose to the paternalists because it points toward a more serious problem with their agenda. High-IQ-only stores would fail to help certain people – specifically, cognitively biased but smart people. They would also harm stupider people who nonetheless have smaller cognitive biases. So the policy would err by both under- and over-inclusion, which is the objection that (I surmise) would enable new paternalists to reject it. But the same is true of all the paternalist policies allegedly supported by behavioral economics, once we recognize the heterogeneity of people – in the extent of the their biases, in their self-correction efforts, and so on. In the context of heterogeneity, virtually any paternalist policy will help some and harm others; for instance, any level of fat tax will assist some people who lack will power while burdening people who don’t (or whose mild will-power problems would justify a smaller tax). To justify such policies, the new paternalists must be willing to weigh the gains to those helped against the losses to those harmed and say the former outweigh the latter. And if that’s the standard, then why not support the high-IQ-only stores? If IQ is correlated with cognitive bias, even if imperfectly, the paternalists’ criterion might well be satisfied.
(To be clear, I have no problem with cost-benefit analysis per se. I just think the new paternalists have exaggerated the benefits of their policies, by ignoring self-correction efforts among other things, while downplaying a whole host of costs – including loss of autonomy, encouragement of dependence on the state, and stigmatization of certain people and lifestyles. I think the high-IQ-only store proposal could push the debate in the direction of taking those costs seriously.)