Bryan Caplan  

Nudge, Policy, and the Endowment Effect

How to Fund All Your Favorite ... The new future of television?...
Last week, Maxim Lott solicited my thoughts on Obama's "nudge team."  Here's what I would have told him if I hadn't been on vacation:

"Nudging" is a great idea.  We should start by ending existing hard paternalism in favor of gentle (or even subliminal) persuasion.  Instead of prohibiting drugs, we should allow anyone who wants to use currently illegal drugs to go to a government website to request an Authorized Narcotics User Card.  Instead of requiring everyone to pay Social Security taxes to provide for their retirement, we should allow people to opt out of the system if they write a one-page essay explaining why they prefer to handle their retirement on their own.  Analogous opt-out rules should be devised for government health care programs, worker protection laws, consumer protection laws, and so on.

Of course, you might want to tweak these proposals a tad.  But if you really believe in the effectiveness of "nudging," such policies deserve serious consideration.  Why then do almost all real-world nudge proposals involve tacking new soft paternalism onto existing hard paternalism?  The endowment effect - humans' tendency to value our stuff because it's our stuff. 

How so?  Behavioral economists, like most people, think of existing hard paternalism as "their policies."  We own these policies, they're ours, and we're not going to casually toss them aside just because we've found a cheaper, more humane alternative like nudging.  Behavioral economists' sense of policy ownership is so strong that the thought of replacing existing paternalism scarcely comes to mind.  As a result, they ignore the revolutionary implications of their own insight, and dwell on moving the salad bar to the front of the cafeteria.

I believe in behavioral economics.  But behavioral economists deeply disappoint me.  They ought to be paragons of rationality - to puritanically avoid the foibles they so ably document.  In practice, however, behavioral economists are all too human.  They use nudge to rationalize a little extra paternalism, when they ought to use their insight to systematically rethink paternalism from the ground up.

That's what I would have told Maxim Lott.

COMMENTS (17 to date)
MikeP writes:


D writes:

Wonderful point Bryan.

Steve Reilly writes:

Doesn't status quo bias explain this better than the endowment effect?

A suggestion for soft paternalism: You are permitted to experiment with drugs only after explaining what a control group is.

MingoV writes:

I read, a few years ago, a web article in support of nudging people to make the "right" choice (such as sending 8% of pay to a 401k plan) the default choice or the "if you don't respond by ____" choice. The authors brazenly claimed that this approach was more libertarian than undiluted paternalism. They wouldn't consider a hands-off approach, because too many people would make "wrong" choices.

Rob writes:

I think endorsing nudging is inappropriate as it creates a false acceptance of paternalism in general.

It's like saying, "Okay, we'll endorse raping people once a week so that we can slowly get away from daily rapes."

The correct answer, of course, is to oppose the rapists by any means necessary.

Chris Koresko writes:

Bryan's point is very good, but it still leaves open the question of how these nudges are to be selected. More savings, or less? More drugs, or less? Who decides, and on what basis?

Perhaps the decisions about nudging should be made by the nudged themselves. Each of us could sit down and make selections from a list of possible personal policy 'defaults', much as we can choose the default behaviors of many of the software packages we use. Then when we're too busy to make a thoughtful choice in a particular situation, the preference we expressed when we were thinking would come into effect. And when we do have time to think it through, we can always choose differently in that particular case, or perhaps even change the default.

Mike H writes:


Really? Those are the same thing? Because that's not obvious to me in the least.

@Chris Koresko

I think the idea is it doesn't matter. Soft paternalistic nudging would be so preferable to our current situation that your questions are not terribly worth considering at this juncture.

Chris H writes:


Not endorsing nudging hasn't helped much now has it? We simply have a lot of hard paternalistic policies everywhere. Perhaps if soft nudges seem like an option, the majority of people who are OK with wide swaths of paternalism will be willing to consider the option. It's worth a shot as long as we aren't, as Bryan says, simply tacking on soft paternalism to the hard.

Tracy W writes:

It does strike me with cognitive scientists that the more they report finding irrationalities in a wide range of people's behaviour, the less reliable their results are.
After all, generally they report finding these irrationaliaties not just in what a 19th century writer might have called "The Great Unwashed" but in people with university degrees, even professors, and the like.
So it seems very likely that the cognitive scientists are irrational and subject to various cognitive biases of their own.
So how can we trust them to have conducted their experiments and analysed their results rationally?

Tom West writes:

Tracy W, it's not that we can't think deeply, it's that most of the time we don't, and when we're not, we exhibit systematic irrationalities.

If there's any book you're willing to countenance on the subject, I'd suggest Thinking Fast and Slow by Daniel Khaneman. His work on this got him a Nobel prize in economics, which implies that economists found is work relevant.

Leo writes:

I'd perfer the federal government not try to nudge me into doing anything in particular. What makes their opinions better than the ones I would form without their nudgings.

Floccina writes:

Great post I think that should one want to get a medicine without a prescription the state could allow you take a little test. Perhaps for antibiotics because if the externality it should be a little tougher and one should need to write an essay in which one explains why antibiotics should not be over used and why it is very important to take the full course.

There is a study of the cognitive biases of regulators. The next logical step is to examine the cognitive biases of people examing cognitive biases and then the cognitive biases of people examing the cognitive biases of people examing cognitive biases and so on up the ordinals ...

Daublin writes:

I have a hard time getting started on this discussion, because I simply don't respect the people who will inevitably control the "nudges".

I suspect many people would feel the same if they thought about these policies applying to them. Instead, people think of these policies as being nudges for *other* people. Even if you are a staunch supporter of Barrack Obama, do you really think he is only going to dish out nudges to Republicans? If you make less than a million a year, you can be sure that he holds you in disdain and will not hesitate to tell you how to live your life.

I find it frankly disturbing. The busy bodies in D.C. have nudged us to use a certain kind of light bulb, to drink high-fructose corn syrup, to use ethanol in our motors, and to build a food pyramid based on carbohydrates. What basis is there for thinking that their next round of nudges will be better?

Tom West writes:

I suspect many people would feel the same if they thought about these policies applying to them.

Daublin, I strongly disagree. I think the visceral reaction to being nudged is limited to a minority of the population that tends towards Libertarianism.

I remember Bryan hating the idea of locking the fridge door (in essence being nudged by his past self) and finding it odd.

I think lowering cognitive effort to "do the right thing" for themselves is something a *lot* of people strive for. Having the government nudge us is simply making it easier to make decisions we wanted to make in the first place.

I suspect distrust of the authority making the nudges is also higher among Libertarians. I think many of the rest figure we're being nudged by commercial interests in their favor dozens of times a day, so why not be nudged by our interests. (And yes, for many, the government = "us".)

Yes, rumour has it that White House considers establishing a nudge unit similar to the UK Behavioural Insights Team. In this post follows the US debate as it unfolds:

We also mention Caplan's post.

Comments for this entry have been closed
Return to top