The theoretical part of Nudge Notes for November 26

Main points

We discussed the theoretical part of Sunstein and Thaler’s book. This lays out the psychological research on some apparent flaws in our decision making. The policy part of the book will suggest ways that so-called libertarian paternalism might address these flaws.

Our discussion

Prof. Brown led us off by noting that economists have used the idea of bounded rationality in their models to accommodate some of the points about decision making that Sunstein and Thaler note. But, she said, the rest poses a challenge to the standard economic models of behavior.

We talked a bit about how the various distinctions introduced in this work (and others that we have or haven’t read) do or don’t line up with one another. These are:

  1. Automatic system vs. reflective system
  2. Calm state vs. aroused state
  3. Consumer vs. citizen (from Sen)

This discussion left me feeling that I really need more information to fully understand what they were talking about.

I did form an opinion, however: I don’t think it’s helpful to talk about these systems or states as persons or selves. Rather, in my opinion, a person contains these different parts, none of which would be enough to make a (psychologically normal) person on its own. But, again, I don’t feel confident that I know what these are, so take that opinion for what it’s worth.

We closed with some discussion of framing effects and following the herd at the 5C’s. Brendan in particular had a provocative theory about ideological uniformity in classes. I’m not sure about the suggestion that pre-tenure faculty are more likely to try to conform to their students’ views than tenured faculty are. He might be overthinking it: the faculty are just not ideologically diverse, prior to any concerns about student opinion. But one virtue of Brendan’s theory is that we can test it: compare the student evaluations and behavior of the two kinds of faculty! (I guess we can test mine too, come to think of it.)


Libertarian paternalism is a bit of an odd duck. Since it doesn’t involve limits on liberty, it’s unlike the kinds of paternalism that show up in, for instance, Dworkin and Mill’s definitions. But it does involve the use of social and political power to alter individual decisions for their own good. In that sense, it’s like the other instance of paternalism.** See Gerald Dworkin, “Paternalism,” Stanford Encyclopedia of Philosophy (2010).

I said that I wish they had spent more time explaining what objections to paternalism they meant to address. Professor Brown correctly said they identified two: paternalism limits liberty and it is often directed at making the target’s life better in ways that the target does not accept. Sunstein and Thaler, by contrast, do not advocate limiting liberty at all: people are always free to act contrary to their nudges. And they insist that nudges make life better by the lights of those who are nudged.

She’s right, of course. So why do I feel unsatisfied? I suppose I’m still frustrated by the fact that discussions of paternalism rarely give a convincing case against it. That paternalism is an undesirable thing is just assumed rather than thoroughly argued for. For instance, it’s true that paternalism usually limits liberty.†† This is so by definition according to Mill and Dworkin. But it’s not obvious that this is automatically a bad thing, especially given that it’s for the benefit of the person whose liberty is being limited. We limit liberty in lots of ways: that’s the difference between civilized society and the state of nature. The trick is to explain why particular limits on liberty are bad.

Of course, I have a paternalistic streak. As I said earlier, I have no problem with in-kind benefits (like health care or food) rather than cash transfers for paternalistic reasons: I think it’s more important that people have health care and food than that they move to their highest indifference curve. I don’t have any problem with a mandatory pension plan (e.g., in the US, Social Security). And I think there are many elements of the law that make sense only on paternalistic grounds. For example, the victim’s consent is not a defense against criminal prosecution and the state refuses to enforce harmful contracts even though they are freely entered into.

I also believe that Sunstein and Thaler have an interesting argument against a wide array of objections to paternalism: our choices are always shaped by outside influences. Since that is so, why not shape them in a way that is directed at our benefit rather than someone else’s profit? So the sort of thing that I wanted to see was a discussion of the scope of this kind of argument. But that’s just what I wanted. Authors aren’t obliged to satisfy me. I can work away at it on my own, thanks to the great material they gave me.

This page was written by Michael Green for Freedom, Markets, & Well-being, PPE 160, Fall 2012. It was posted November 26, 2012.
Name of website