[Ed. Note: On Friday, May 2 and Saturday, May 3, 2014, the Petrie-Flom Center hosted its 2014 annual conference: “Behavioral Economics, Law, and Health Policy.” This is the first installment in our series of live blog posts from the event; video will be available later in the summer on our website.]
Today’s sessions opened with a keynote address from Cass Sunstein, Robert Walmsley University Professor at Harvard. Sunstein is also the founder and director of the Program on Behavioral Economics and Public Policy at Harvard Law School.
Sunstein addressed a wide variety of in-depth issues in his talk on “Choosing Not to Choose.”
In the beginning of the talk, he provided three objections to the argument that active choice-making is a muscle and should not be interfered with. First, people often choose not to choose, and forcing them to do active choice-making is “part of the family” instead of an alternative to default rules. It is also a form of libertarian paternalism. Second, in thinking about active choosing and default rules, we should bear in mind a basic evaluation framework that helps to minimize sum of decision costs and error costs. Third, sometimes it is best to choose not to choose. In many case we should honor people’s choice not to choose when it minimizes decision costs and reduces the magnitude and number of errors (especially when people are forced to choose, they may go in the direction that is wrong or not in their best interest).
Sunstein went on to three examples on the ground to orient the audience:
- Case 1: A private company is deciding among three options: (1) automatic enrollment in insurance unless opt out, (2) opt in to insurance, or (3) as condition for starting work, forced choice of whether to be insured and which insurance plan (active choosing).
- Case 2: A utility company is deciding between a green default, a grey default for its consumers, or to force them to decide which source they prefer (no service until you decide).
- Case 3: A doctor is dealing with patient facing difficult medical situations, and could decide among: (1) present array of options, (2) default (“if it were me I would/most patients do”) with opt out.
All three cases have an institution considering requiring active choice instead of default rules. Sunstein played with the meaning of “requiring” in order to unsettle the opposition and to suggest that it is often illusory. He briefly reviewed claims such as that doctors and policymakers are prone to error as well (behavioral biases), that governments lack knowledge as well, that behavioral findings can compound the problem, or that even when people are likely to err, their autonomy to choose should be respected. He argued that the distinction between active choosing and default is rather illusory because people often want to choose not to choose (for reasons like limited bandwidth, find choosing unpleasant, don’t want to take responsibility or regret, or know they are biased). When people don’t want to choose and are forced to do active choosing, we forbid their choice not to choose. “Choice requiring paternalism is not an oxymoron.”
What does requiring active choosing mean? Sunstein considered three categories of possibilities:
- Punishment for those who refuse to make an active choice (it is very rare but not unheard of – the case in Australia).
- Required active choosing with respect to an ancillary matter, so people cannot obtain a good or service unless they make a choice (it is also coercion here if the type of good or service is essential).
- Active choosing among goods, services, or jobs in a free market structure (think about grocery shopping in a supermarket).
The last category of market inevitability is true for now, but as technology and big data develop, it may change (book of the month club, predictive shopping algorithms). At the moment, people don’t like the idea, why? Remarkably, the chief argument for liberty is the choice architect’s lack of knowledge (epistemic gap).
With regard to the relationship between leaning (developing individuality) and paternalism, Sunstein used GPS as an example. GPS route can be understood as a soft, means–oriented paternalism, as it helps you get somewhere but doesn’t override your choice. A study of taxi drivers showed that there is a neural change as they come to understand their route. Maybe that wouldn’t happen if they had GPS. This gets at John Mill’s “muscular power that is improved only by being used.” So the argument is that active choosing can develop a capital stock. Sunstein placed a Footnote in this regard: “People do and should learn about whether to choose actively or choose not to choose.”
As pointed out by Sunstein, choice can be a great benefit or an immense burden. It is of crucial importance to respect people’s desire to choose and also their frequent antipathy to choosing. Both default rules and requiring active choosing count as nonlibertarian paternalism in the sense that people’s own (first-order or second-order) choice is being rejected. The argument for active choosing or default rules largely depends on decision costs and error costs (heterogeneity, information, neutrality).
During the Q&A period, Sunstein spoke further about the possibility of abusive choice architecture by governments and private institutions, difference between active choosing and simplified active choosing, whether healthcare is a special case for second-order learning, types of choice architecture and information asymmetry, what default rules entail, and the (ir)relevance of the rationalist model.