Predictably Paternalistic

arielyWhen I had a class with Dan Ariely—author of the highly acclaimed Predictably Irrational—last semester he performed a dollar auction, where students could bid on a $20 bill. The highest bidder got the $20, but the second-highest bidder had to pay the second-highest bid. What generally results is a bidding war that goes above $20, causing both the highest and the second-highest bidder to lose money. Predictably, then, this led to an irrational escalation of the bids up to $30, at which point the two remaining bidders agreed to split the cost.

Now, the Freakonomics blog directs us to Swoopo, an auction website that takes full advantage of this behavioral economics insight:

“How can Swoopo, the online auction site, rake in $2,151 selling a laptop for $35.86? Easy: set an opening price of $0.01 (almost free!), then let each new bidder top the last by only a penny, and extend the auction each time someone places a bid in the final seconds. Oh, and collect $0.60 from each player for each bid they place. The winner of the auction might walk away with a good deal, but the losers will have racked up big fees chasing their sunk costs. The house always wins. Writer Mark Gimein calls the site “the evil bastard child of game theory and behavioral economics.”

So, for that laptop that sold for $35.86, there are 3,586 bids, and everyone but the winner—who gets a fantastic deal—goes home with significant sunk costs. Gimein does a great economic analysis of Swoopo, so I encourage readers to check out his article.

For me, Swoopo is reflective of my broader concern about the uses of behavioral economics. There are legitimate concerns about the external validity of behavioral economic experiments, but my concerns are more moral in nature. When a website like Swoopo utilizes behavioral economic insights, it is only affecting those people who choose to engage in the behavior on the website. When a government chooses to utilize behavioral economic insights, it has the potential to affect the behavior of everyone within its jurisdiction. Clearly, on the government level, the stakes are high.

Let’s take a look at one seemingly harmless proposal by Ariely based on a paper by Eric Johnson and Daniel Goldstein. He explains that 99.8% of Austrians and 98% of Belgians choose to donate organs upon death, while only 12% of Germans and 17.7% of British do so. Why the difference? Is it because of the egotism of the Brits and the Germans compared to the benevolence of the Austrians and the Belgians? While this is an amusing explanation, it is not true. The countries with high percentages have opt-out policies whereas the lower percentage ones are opt-in. As Ariely explains it, in both cases most people simply stick with the default option. The implication of Ariely’s commentary is that opt-out is the better default option, since it leads to higher donation percentages. Ariely rightly claims, “This is a hard emotional decision about what will happen to our bodies after we die and what effect it will have on those close to us. It is because of the difficulty and the emotionality of these decisions that they just don’t know what to do, so they adopt the default option….”

Ariely is right: This is a hard emotional decision for many people. Accordingly, the default should not be to perform a cognitive trick on people in order to get them to donate their organs. Ariely talks as if the default options are somehow morally equivalent, but they are not for most people. I personally have opted-in to donate my organs when I die, and I have no moral qualms about doing so. Many people do, however, and for them, the default option of removing their organs is not equivalent to the default option of keeping their organs intact. The former action may very well be more undesirable for the individual, but they go with it anyway because of a cognitive ploy.

And, this is what worries me about applications of behavioral economics: If something as simple as organ donation can be morally controversial, so can virtually every issue to which insights from behavioral economics are applied. Rather than engaging in those moral debates or allowing individuals to engage in these moral debates independently, the government is choosing a side when it decides to go with “opt-in” or “opt-out.” Behavioral economists like Ariely often assume that what they think is best is best. But, insightful psychologists and economists are not always wise philosophers and moralists; accordingly, we should be wary about outsourcing our personal decisions to them.

7 responses to this post.

  1. Posted by John S on July 10, 2009 at 11:41 AM

    “This is a hard emotional decision for many people. Accordingly, the default should not be to perform a cognitive trick on people in order to get them to donate their organs.” Well, why not? I understand what you’re saying about a “cognitive trick” being manipulative and a way for behavioral economists to impose certain standards on others, but the fact is, SOMETHING needs to be the default. And if the default has such an overwhelming impact on how people decide, why shouldn’t we make the default the option that is more socially beneficial? Granted, this sounds like an imposition of Ariely’s (and my own) morality, but it’s not like morality is purely a matter of taste; there are very compelling reasons for organ donations. I don’t begrudge people who are opposed to it, but for those who, essentially, cannot make up their minds, why not “nudge” them in the direction that benefits more people?

    I actually have a more practical issue here. Ariely says that the default option usually wins out because the decision is so difficult and most people simply do not have strong feelings. This accounts for a lot of it, likely, but I think there are also a fair share of people who stick with the default answer because they don’t know a question is being asked. In New York, for example, you decide on organ donation when filling out a form for your driver’s license…Is there some sort of intuitive connection between the ability to drive and the quality of my organs? I think it’ reasonable to assume some people just skip that box completely, not because of the “difficulty and the emotionality of these decisions”, but because, well, the box is small and they are worried about getting their license. In this situation, it’s not that people are deferring to the default option, but that the option is insinuating itself based on inattention, essentially like fine print.


    • Posted by Josh on July 10, 2009 at 1:03 PM

      I agree with Dan that something doesn’t need to be the default. It could be as simple as when you get your driver’s license, you only get it on a condition that you check one of the boxes, to donate or not to donate. So, if the individual could be left to make his own decision, I don’t think the government should be nudging him towards the decision that you think is more socially beneficial.

      John, I think you’re right on with the point that the default option sometimes wins out just because a. either people don’t realize or b. it’s just too hard to change–the transaction cost is too high. But, I actually think this practical issue is relevant to the broader discussion. This would mean that people are making this important decision to donate organs based on ignorance or transaction costs, which is obviously not ideal. So, the solution is not to look for ways to nudge people but to look for ways to maximize the ability of individuals to make autonomous decisions.


  2. Posted by Dan on July 10, 2009 at 12:55 PM

    I don’t think there necessarily HAS to be an opt-in or opt-out. You could have two boxes: donate organs and do not donate organs. The form will not be accepted until one of the boxes is checked. This will force people to make a decision and not impose any morality.

    Josh, you’re really trying to create a distinction between “positive” behavioral economics and “normative” behavioral economics: the former being the results of the experiments (outside of the external validity debate) and the latter (in the case of predictably irrational) being the couple of comments Ariely seems to add after discussing the experiments (which, to me, never seemed to have appropriate justification).


  3. Posted by Jane on July 14, 2009 at 5:26 AM

    “Behavioral economists like Ariely often assume that what they think is best is best.”

    You didn’t think I was going to let that slide, didja? I’d argue Dr. Ariely isn’t assuming any such thing. A careful reading of his work -specifically, his experiments, and not the plethora of studies that he skillfully weaves into his book and lecture in order to fully explain concepts to his readers -suggests that his focus is *not* on helping institutions manipulate the masses ‘for their own good.’ It’s true he makes normative remarks about his findings -but the goal seems to be helping people to 1) avoid manipulation by others, and 2) successfully *manipulate themselves* -to outmaneuver their own irrational patterns of behavior, as it were, by giving them insight into the thought processes (or lack thereof) behind them.

    To respond to “Dan” and his comment about Ariely’s lack of appropriate justification for his comments: I’m not sure I understand what you mean. I think that’s Ariely’s way of pointing us to the many interesting, and disturbing or inspiring, implications and applications for his outcomes.

    And finally: “this is what worries me about applications of behavioral economics.” I wonder what you mean by “applications.” For behavioral economics to turn out the data that it does, and for us to “do nothing” with the information we gather about how we behave -that would in itself be a kind of application, and one that seems to make little sense.


    • Posted by Josh on July 14, 2009 at 1:24 PM


      Let me be clear: I don’t think all or even most of Predictably Irrational falls prey to the critiques that I’m expressing. For instance, in the auction example I give at the beginning of the post, I think Ariely’s dollar auction does, like you say, help us to realize our own systematic irrational impulses. Hopefully, someone who read Predictably Irrational wouldn’t be partaking in too many Swoopo auctions. Moreover, his experiments on things like “hot” states versus “cold” states are very informative and potentially useful for being aware of your own irrational behavior.

      So, I don’t think we disagree too much. My problem is when he (and other behavioral economists) switches from experimental economist to politician and moral philosopher without seriously engaging in the moral and political questions that need to be considered (as he does when discussing the organ donation example, or the financial crisis). Nudge (by Thaler and Sunstein), one of the most successful behavioral econ. books, is all about “light” paternalistic ways of nudging us in the right direction. So, I don’t think it’s unreasonable to say that behavioral economics as a discipline has inherent inclinations towards paternalism.

      Jane, you claim: “For behavioral economics to turn out the data that it does, and for us to “do nothing” with the information we gather about how we behave -that would in itself be a kind of application, and one that seems to make little sense.” My argument is not to always do nothing. Take the organ donation example. I advocate shifting AWAY from a default system to a system where you give individuals a real meaningful choice rather than cognitively tricking them into choosing the default option that the behavioral economist prefers. You may agree with Ariely on many moral/political questions, but you might not agree with other behavioral economists. To condone their not considering moral issues seriously and just making policy prescriptions based on what they think is best, is not advisable.


  4. […] into economic models so that they offer more accurate and wise predictions and prescriptions. While I think we need to be wary of some of the applications of behavioral economics, there is little doubt that Predictably Irrational made these applications […]


  5. […] flaws, we should generally champion* behaviors and policies that help to minimize these flaws. I earlier discussed the tendency of individuals to be much more likely to say yes to something if it is opt-out rather […]


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: