Real Libertarians on Nudge

Notes for December 3

Main points

We discussed Rizzo and Whitman’s criticisms of what they call the “new paternalism.”

New and old paternalism

Rizzo and Whitman distinguish what they call new paternalism from old paternalism. New paternalism aims to influence people’s behavior for their own good as measured by their own preferences. Old paternalism aims to influence people’s behavior for their own good even if they do not see it that way.

To put it another way, the new paternalists are said to try to help people to achieve what they most want. The old paternalists are said to try to help people to achieve what the paternalists want.

Rizzo and Whitman argue that the new paternalists face a host of problems in determining exactly what people’s real preferences are. Since these are practically insuperable, their program will inevitably devolve into old paternalism.


Some of us didn’t find this conclusion so terrible. I think it’s fair to say that Maddie, Jessica, Ian (?), and I were reasonably comfortable with old paternalism. Some things just are better for you even if you don’t appreciate that. The psychology of addiction alone makes this plain.

Of course, there are obvious dangers in old paternalism. But that’s a different point from saying that it’s necessarily a terrible thing.

I also thought that while their points are well taken, they didn’t really undermine the strongest examples in Nudge. People really do leave money figuratively on the table and it’s just hard to see how that reflects their real desires as opposed to a psychological quirk. Plus, we can look at mandatory retirement plans like Social Security and see pretty good evidence that they have worked out well by the lights of those who have been treated paternalistically. Poverty among the elderly is significantly lower after Social Security than it was before. Does anyone really want to live in poverty? Really?

Now, there are two classes of exceptions, that is, people who are worse off with forced savings than they would have been without it. First, there are the very wealthy, who do not need forced savings and would probably do better investing their money on their own. Second, there are those who die after paying into the forced retirement system but before collecting on it. They would have been better off consuming rather than saving.

These examples show that it is very difficult to have paternalistic policies that are better for literally everyone. The most that we can reasonably aim for is that they improve the vast majority of people’s lives. I think it also points up some of the limits of libertarian paternalism. If Social Security were voluntary, the wealthy would overcome the behavioral nudges and withdraw from the system. I don’t think that Sunstein and Thaler meant to suggest that.

The case of people who die too young to collect gets more complicated because race is often added to the mix: black men die young and don’t cash out their Social Security benefits, therefore Social Security is a racially discriminatory program, therefore it should be abolished. (Professor Brown mentioned the first part of this argument but I don’t think she meant to suggest the rest; I’m just saying this is how it is often completed, not that this is what she meant.)

Race is thrown in there as a rhetorical device, in my opinion. Really, the point is just about people who die young. Social Security isn’t aimed at worsening the position black men in the way that, say, voting restrictions were.

More importantly, I think the argument willfully ignores the actual problem. The problem is that there is a class of people who die earlier than necessary. The proper policy response is to prevent the deaths. It’s perverse to think that canceling retirement programs is the first thing to do.

This page was written by Michael Green for Freedom, Markets, & Well-being, PPE 160, Fall 2013. It was posted December 5, 2013.
Freedom, Markets, & Well-being