Tuesday, August 9, 2011

meta-(behavioral economics)

I'm not fond of the idea of libertarian paternalism.

Part of the reason why is that I'm partially just a stubborn classical economist who believes nearly dogmatically that people should be left alone to do what they want to do because they know best what is best for them, and the whole concept of sneakily coercing people into making different choices that they might not want to make is creepy.

And part of the reason is that I personally really want to be left alone to do what I want to do, and I really don't want to be sneakily manipulated into doing things I might not want to do.

And probably most of the reason is a purely irrational gut negative reaction to the word `paternalism' which I and most other economists are firmly opposed to for a host of perfectly reasonable reasons and some personal preferences too of course; and another negative reaction to terminology that taints such a wonderfully optimistic and freeing philosophy of libertarianism with such an ugly, pessimistic, patronizing thing as paternalism.

But beyond all that is another, very significant, reason. That is that people have consistent types of irrational behavior that will likely lead to abuse of libertarian paternalism as soon as it's accepted as an option. I worry that behavioral economists who propose behavioral solutions to behavioral problems are forgetting how other behavioral problems will lead those solutions to be counterproductive.

  • We're overconfident about what we believe to be true: when we pick up on some reason that people might make irrational decisions, we're then convinced that we know what's better for them.
  • We project our own preferences and values on others: if we see someone doing something that we wouldn't do, we overestimate the likelihood that they are mis-optimizing their own utility via some kind of irrationality. When a smoker starts begging me to tax him so they'll have an easier time quitting, I'll listen, but until then I'm inclined to trust their revealed preferences more than my own guess of their preferences.
  • We have delusions of control: we think we're capable of tweaking the world for the better without having to deal with endless fallout that often outweighs any benefit of the tweak.
  • We hate being passive: give us an option, any option, to actively do something to address an imperfection in the world, and we'll take it, whether or not we know how to do it well or what the side-effects might be or who might get hurt in the process.
  • And we pay more attention to salient information: it's easy to see that a default savings program helps many people increase their savings, but it's not obvious how many people are coerced through that program to save more than they should in their personal circumstances; a policy maker who isn't as used to searching for those hidden effects as economists who invent the policies are are going to be too keen on the policy. Not to mention, `small' costs or reductions in freedom are too easily written off as `no' costs. I for one really don't want to have to actively opt out of every silly nanny state policy the government wants to coerce me into. Government bureaucracy is never a trivial detail.
Obviously libertarian paternalism is a step in the right direction from paternalism. But that's only true when comparing two alternate policies of each type. But people react much more negatively to paternalistic policies, so it's hard to get them passed in the first place. Libertarian paternalistic policies are less damaging but also much easier to implement because moral libertarians can't really object and the downsides are much less obvious. So I worry that shifting towards this type of solution upsets the balance of volume in favor of the nanny state champions that I really don't want running the country.


Riley said...

Alright, I just followed your links to past blog posts about libertarian paternalism and got to your post about Nudge. I'm intrigued by the ideas that you say might be "very interesting, if not profoundly revelatory" for those of us who aren't behavioral economists, but your review of the book was... less than glowing. Do you have any suggestions for better resources? I have zero background knowledge - my interest stems almost entirely from your blog posts!

Vera said...

probably Nudge :) unfortunately... but to be fair most of my objection to the book is to the ridiculous libertarian paternalism terminology. I didnt like the organization either but you can partly get around that by only reading a couple chapters of it. and I know a few people without background knowledge who read it and loved it, so considering the fact that it's also really the definitive book on the subject, that's the best recommendation I can come up with...