Home > Books, Discourse, Incentives, Information, Privatize > Book Review: Nudge

Book Review: Nudge

September 29, 2010 Leave a comment Go to comments

As promised, the following is my review of Nudge: Improving Decisions About Health, Wealth, and Happiness, by Richard H. Thaler and Cass R. Sunstein. This isn’t an especially timely post (the book originally was published in 2008, and I read the “Revised and Expanded Edition” from 2009), but, for its popularity and apparent influence, I do not think many of the readers of this site also have read this book.

The book is concerned with the way people deal with choices in their lives. The authors refer to an arrangement of options– for example, the layout of a food buffet– as choice architecture. Choice architecture is manipulable, and choice architects have the ability to control decision making in a meaningful way through the presentation and arrangement of options. “Humans predictably err,” and, by understanding these cognitive biases, choice architects can shape outcomes. Richard H. Thaler & Cass R. Sunstein, Nudge: Improving Decisions About Health, Wealth, and Happiness 7 (Penguin Books 2009) (2008). There is an entire set of studies and literature on cognitive biases that is emerging, particularly in the behavioral economics literature. Examples include confirmation bias, self-serving bias, and anchoring bias. A detailed exploration of these and other cognitive biases is beyond the scope of the book and this post. What is important for Thaler and Sunstein is that everyone has these systematic biases. Also important to Nudge is pointing out the “misconception…that it is possible to avoid influencing people’s choices”; in other words, there is no such thing as neutral choice architecture. Id. at 10; see also id. at 249-51 (discussing different types of neutrality and situations in which neutrality may be possible). Intended or not, every arrangement of options has an effect on those faced with the options.

The core of Nudge‘s contribution is an approach to choice architecture (which also could be thought of as system organization) the authors call “libertarian paternalism.” As the authors explain:

Libertarian paternalists urge that people should be free to choose. We strive to design policies that maintain or increase freedom of choice. When we use the term libertarian to modify the word paternalism, we simply mean liberty-preserving….Libertarian paternalists want to make it easy for people to go their own way; they do not want to burden those who want to exercise their freedom.

The paternalistic aspect lies in the claim that it is legitimate for choice architects to try to influence people’s behavior in order to make their lives longer, healthier, and better. In other words, we argue for self-conscious efforts, by institutions in the private sector and also by government, to steer people’s choices in directions that will improve their lives. In our understanding, a policy is “paternalistic” if it tries to influence choices in a way that will make choosers better off, as judged by themselves.

Id. at 5 (quotation marks omitted). Thaler and Sunstein call this influencing “nudging”:

A nudge…is any aspect of the choice architecture that alters people’s behavior in a predictable way without forbidding any options or significantly changing their economic incentives. To count as a mere nudge, the intervention must be easy and cheap to avoid. Nudges are not mandates. Putting the fruit at eye level counts as a nudge. Banning junk food does not.

Id. at 6. The authors acknowledge that “some of our nudges do, in a sense, impose cognitive (rather than material) costs, and in that sense alter incentives,” but argue that “nudges count as such, and qualify as libertarian paternalism, only if any costs are low.” Id. at 8.

Much of the book covers case studies showing nudging in action or how the greater use of effective choice architecture could improve outcomes for people. These examples include saving money, financial investment, the privatization of social security, prescription drug plans, organ donation policies, environmental conservation, school choice, health care, and marriage policy. While I’m not sure that Thaler and Sustein’s proposal for a “privatization” of marriage in which states only issue civil unions fits neatly within this book’s framework, I do think it could make a good topic for a future post here. If you are especially interested in these and the many other examples of nudges, the revised edition of the book or the authors’ regularly updated blog are good places to look.

Near the end of the book, the authors respond to certain objections to their approach. The first objection is that of the classic slippery slope: if we permit this limited paternalistic intervention, it won’t be long before government overreaches, and “highly intrusive interventions will surely follow.” Id. at 239. Thaler and Sunstein have substantive and structural responses to this objection: first, they would prefer this sort of critic to engage with them as to the merits of the chosen policy preferences rather than the structure itself, and second, this sort of structure is inevitable: “It is pointless to ask government simply to stand aside. Choice architects, whether private or public, must do something.” Id. at 240. They continue:

Those who make this argument sometimes speak as if government can be absent– as if the default terms that set the background come from nature or from the sky. This is a big mistake. To be sure, the default terms that now apply in any particular context might be best…. But that view must be defended, not assumed. And it would be odd for those who generally hold government in extremely low esteem to think that in all domains, past governments have somehow stumbled onto a set of ideal arrangements.

Id. at 241. Thaler and Sustein also do not think that a traditionalist, Burkean response based on the wisdom of longstanding social practices has much value here because “inertia, procrastination, and imitation often drive our behavior.” Id.

A second objection is that choice architects may have their own agendas, the implication being that these agendas could be “evil” or otherwise not in people’s best interest. Recognizing “that choice architects in all walks of life have incentives to nudge people in directions that benefit the architects…rather than the users,” the authors nevertheless believe that “lin[ing] up incentives when we can, and employ[ing] monitoring and transparency when we can’t” will be sufficient to overcome this issue. Id. at 242. Freedom of choice and transparency are important in this area. Related is John Rawls’ publicity principle: “In its simplest form, the publicity principle bans government from selecting a policy that it would not be able or willing to defend publicly to its own citizens.” Id. at 247. This would prohibit secrecy on the part of the government when it alters legal default rules, for example.

Thaler and Sunstein present and respond to other objections, but, at this point, I would like to both encourage readers to include their questions and critiques in the comment section below and offer my own thoughts.

First, despite the widespread, sweeping accolades for Nudge, I am not convinced that this is an especially groundbreaking book. Advertisers have been doing this sort of thing for years, grocery stores and casinos are consciously designed to make us spend more time and money in them, and there are entire fields of public relations, marketing, and advertising focused on driving individuals’ decision making towards particular ends by playing on the individuals’ desires, habits, interests, and incentives. Nudge does bring the idea of cognitive biases to the public in a way probably not present before, but it barely explains these biases, and there’s almost no discussion of the fascinating research in that area. The book also missed a chance to do a bit more (perhaps empirical) work to explore just how people respond to various nudges as a general matter that might allow readers to apply the conclusions that followed therefrom in leadership settings in their own lives. I appreciate that the book is geared toward a general audience (of which I may or may not be a member),  but I found it a bit too light.

Second, I think the second objection I mentioned above– a concern over the goodness of the intentions of choice architects– is a serious one that the authors fail to rebut sufficiently. Anyone who spends time studying cognitive biases and systematic human behavior should be able to take that knowledge and design a system that preferences any particular outcome, even ones that are not “beneficial” to users. Thaler and Sunstein recognize, through the myriad examples they cite throughout their book, that this happens all the time; they simply chalk it up to bad choice architecture and poor nudging. This might very well be the case in the examples they chose. The “bad” outcomes there could be a result of a lack of organization, planning, awareness, knowledge, or forethought. But what if similarly “bad” outcomes resulted from a knowledgeable, organized, aware choice architect and choice architecture? Freedom of choice and transparency might help here, but there seems to be no special reason why they should. An early example in the book is a school cafeteria in which a choice architect may organize food in a way that makes students more likely to choose fruits and vegetables and less likely to choose junk food. The reason this works has nothing to do with the nature of the fruits and vegetables (or that nature vis-a-vis the students) and everything to do with the systematic biases of the students themselves. So long as the libertarian constraint is in place, the students can choose whatever they want, but the paternalistic element permits the choice architect to select any option as the preferred option. Fruit, vegetables, and junk food are mere variables in this arrangement. (In political theory terms, the structure here is largely deontological, the right to define “the good” resting solely with the choice architect.) Additionally, Thaler and Sunstein place no duties on the choice architect in his or her selection of the “good” alternative. An obvious one might be a duty to be informed. The cafeteria administrator might have some idea that fruits and vegetables are healthier than junk food, but we also are told that dark chocolate and red wine (though not for children) have health benefits too. How should one evaluate this particular tradeoff, and how should one evaluate it against all of the dietary decisions that go into putting together a balanced meal? Does a “balanced meal” mean the same thing for every student? For a majority of students?

An architect

I think concerns about the qualifications and motivations of choice architects are greater than Thaler and Sunstein admit, and, if nothing else, deserved a more adequate defense. Overall, though, I am glad I read the book, and I think it does a good job of spurring discourse. Irrespective of whether you’ve read Nudge, I welcome your thoughts below.

  1. September 29, 2010 at 9:17 pm

    I admire the clarity of your writing. You obviously keep your eye on what you want to say and express it without ambiguity, something I could do more often in my own writing (although ambiguity has its place). Reading your review, I was struck by a vision of millions of choice architects (advertisers, government officials, lobbyists, politicians, families, individuals) all trying to nudge us in a million different ways. On the receiving end, it becomes a challenge to maintain your own beliefs and navigational system and meet all your daily responsibilities. But that’s part of living in a pluralistic society, and essentially what shapes our personalities–the give and take with other people on this planet.

    I also wondered how you feel about initiatives that go beyond nudging–take laws, for example. Is something more than a nudge necessarily a bad thing?

    • AD
      October 4, 2010 at 2:23 pm

      Thank you, and thanks for reading. To respond to your question, I think laws are, for the most part, a different animal. Under the Thaler & Sunstein approach, nudges are useful to governments or societies because they are tools to advance social policy in areas where command and control regulation is less desirable. Environmental policy is a good example because it lately tends to be a polarized area politically. For the authors, nudges must meet the libertarian paternalism requirements (costless ability to choose alternatives), so a nudge in this area toward greener practices would be a way to advance those aims that is less objectionable to opponents because they could opt out at low or no cost.

      Much of nudging has to do with the setting of default rules. Parties always can bargain around default rules (and libertarian paternalism requires the cost of that bargaining to be negligible), but the setting of the default rules is important because people tend to just accept whatever terms are presented to them, even when they are not what they would choose given a blank slate (basically an inertia notion). The crux of this line of thinking is that if you can create favorable defaults, the desired result will obtain in more instances because people who are indifferent or uninformed will select that result because they don’t know or don’t care, and at least some who are opposed might still select your result because it’s easier to do. A very simple example is a motion-activated paper towel dispenser in a restroom. If the establishment manager wanted to limit paper towel consumption, he could reduce the length of paper towel that spooled out after a hand wave. (Of course, successful default rule setting probably requires some research. For example, if he sets the length too short, too many people might wave for a second or third piece.) Back to the realm of controversial political areas, hospital administrators could require doctors to ask newly pregnant patients a question about the possibility of aborting the pregnancy. The choice always would belong with the woman, but the question, merely by being asked, is likely to influence that choice. (Notably, the influence could go either direction. Cf. “Are you aware of the physical, mental, and ethical dangers of having an abortion?” with “Have you considered having an abortion in light of the substantial responsibilities that come with raising a child?”)

      Getting back to your question, there are different sorts of laws. The laws I think you’re asking about are of a regulatory variety. In that case, nudges and laws could be understood as being situated along a spectrum. My interpretation of the Thaler & Sunstein view described in this comment is that nudges can advance policy objectives in areas where it is too politically difficult to do so with a legal edict. We might favor a legal approach where we think the nudge will be insufficient to result in meaningful levels of complying behavior with the chosen goal, or where the dangers of not complying are very high. We do this sort of command and control regulation with some frequency, and not everybody likes it, which probably is the impetus for the authors’ attempt at a compromise approach. (Another sort of laws work not at advancing social policies but toward protecting individual rights and liberties. I think one can have an opinion on criminal laws and civil rights laws mostly independent of one’s opinion of approaches to regulation.)

      Once we say that it’s appropriate to do some command and control regulating, we face a new set of questions, including who should do the regulating (e.g., federal vs. state government, public vs. private institutions) and on what basis. The book’s contribution comes in raising an alternative to the familiar approach. Some think that there is a certain amount of paternalism inherent even in the concept and existence of government. It is hard to see how a government could function without exercising power through means more forceful than nudges. After reading all of Nudge, I’m confident that Thaler & Sunstein believe that there are some, perhaps many, areas in which nudging is inappropriate because it would be insufficient. One of their points, as I read them, is that there are some, perhaps many, areas in which something less paternalistic will work just fine.

  1. No trackbacks yet.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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: