Nudges and public policy. Get back to the basics of behavioral science

Cass Sunstein is busy working for the Obama administration and Richard Thaler is currently consulting to the coalition government in the UK. This post expresses the independent thoughts of the Nudge blog’s applied behavioralist John Balz.

Where behavioral science meets up with public policy, politics is not far behind. That is the implicit takeaway from the recent House of Lords report’s conclusion that effective policies to change behavior need to use a variety of interventions including traditional financial incentives, regulations, and yes, behavioral nudges.

This is an unsurprising conclusion. You would be hard pressed to read Nudge and conclude that the world’s most complex problems like climate change and obesity are to be solved with a single behavioral tool, or even a group of them. Together, preliminarily successfully behavioral interventions like Opower’s utility statement along with preliminarily less successful interventions like calorie count displays in New York City restaurants underscore the importance of social contexts, evidence-based policy, and further research. Applied behavioral science is an emerging field of innovation, which requires experimentation and failure. Policymakers who are looking for guarantees of success should, to use a favorite word from Nudge, opt-out of the arrangement.

Like other catchy phrases, “nudging” has taken on a life of its own in popular discourse, which is always a double-edged sword. This is evident in the differing accounts of the term’s meaning that government officials offered the committee. Ultimately, the committee settled on describing nudges (p.12 of the report) as interventions that “prompt choices without getting people to consider their options consciously, and therefore do not include openly persuasive interventions such as media campaigns and the straightforward provision of information” – a description that satisfies the committee, but not necessarily a room full of behavioral scientists.

In the blogosphere, Liam Delaney acutely points out that popularization has led many to focus on the term nudge “rather than the core ideas,” and in the process “obscures the really important set of ideas emerging in behavioral economics, very few of which are spoken about in the document.” To the House of Lords committee, nudging is the raw, narrow application of non-regulatory interventions of which there appear to be four (1. Provision of information – although it somehow can’t be “openly persuasive”; 2. Changes to physical environment; 3. Changes to the default policy; and 4. The use of social norms and salience). To an applied behavioral scientist, nudging is part of a broader application of behavioral science principles, which can be understood as an engagement with how a boundedly rational decision maker with a heterogeneous set of material and non-material preferences interacts with the world around them. At such an abstract level, you can understand a policymaker’s desire to quickly descend to the ground. Great. Now how can I apply this to my program?

Policymakers who work with behavioral scientists to identify particular behavioral problems (e.g. Why don’t people save more in their retirement accounts? Why do people fail to take advantage of a social program they’ve started to apply to?) and understand how and why they form are the ones most likely to benefit from behavioral insights – in combination, of course, with other traditional insights. So are those where considerable thought is given to questions about the scalability of a particular behavioral solution. Skipping ahead to technical solutions in the form of handy lists of nudges may be tempting, but is likely to lead to unsatisfactory results.

There is also a much bigger political debate going on in the background of these conversations. Is behavioral science a challenge to the regulatory state or a supporter of it? Is it ideologically left, right, or center? When is a nudge a shove? What constitutes “voluntary”? Are behavioral policies transformative or cute and tiny? Perhaps you think applied behavioral scientists should grapple with these questions. Or perhaps you think they are largely irrelevant to the particular problems and policymakers who work in partnership with them. Like other political debates, these are philosophical ones that can go in circles without reaching clear resolutions. That’s fine. Individual societies with differing social and cultural values will settle on answers to their own level of satisfaction. Either way, applying behavioral science is not a mandate, it is a choice. The applied behavioral scientists and those who would like to engage with them, in the public and private sectors, should (and hopefully will) continue to do so with an openness to scientific experimentation, a commitment to rigor, and an expectation of realism.