Nudge: Improving Decisions About Health, Wealth, and Happiness.

• Author: Schlag, Pierre
• Position: Book review

NUDGE: IMPROVING DECISIONS ABOUT HEALTH, WEALTH AND HAPPINESS. By Richard H. Thaler & Cass R. Sunstein. New Haven and London: Yale University Press. 2008. Pp. x, 282. $26.

INTRODUCTION

By all external appearances, Nudge is a single book--two covers, a single spine, one title. But put these deceptive appearances aside, read the thing, and you will actually find two books--Book One and Book Two.

Book One begins with the behavioral economist's view that sometimes individuals are not the best judges of their own welfare. Indeed, given the propensity of human beings for cognitive errors (e.g., the availability bias) and the complexity of decisions that need to be made (e.g., choosing prescription plans), individuals often make mistakes. Enter here the idea of the nudge--the deliberate effort to channel people into making the selections that are best for them. The key thing about a nudge is that it's noncoercive and there's always an opt out.

The nudge is an idea that ought to appeal to many people on the political spectrum. The basic idea is that institutions, including the government, inevitably structure the choices that people make (Cass Sunstein (1) and Richard Thaler (2) aptly call this a "choice architecture"). Their idea is that we might as well be thoughtful about how to structure our choice architectures. Choices are being driven by choice architecture anyway, so why not do a good job of it as opposed to a bad job or no job at all in structuring the thing?

This is the gist of the first book. Understood as an idea about how to improve government delivery of services, Sunstein and Thaler's theory is really hard to fault. (3) They have aptly described a way of thinking about government programs and regulation that could well improve the delivery of government services. "Choice architecture" could even become part of the new regulatory lexicon; it's evocative and it's smart. And they've operationalized it (which is an important thing given Cass Sunstein's current job description as head of OIRA--Office of Information and Regulatory Affairs). (4)

Then there's Book Two. This book is also about nudge. Nudge on steroids. This is the book where Nudge--the insight about how to improve delivery of services--turns into the full-fledged ideology that Sunstein and Thaler call "libertarian paternalism." In its libertarian moment, libertarian paternalism eschews command-and-control regulations in favor of regimes designed to allow individuals to "go their own way." In its paternalist aspect, libertarian paternalism nonetheless allows the government to structure the regime with a view to achieving set government goals. The defining characteristic of this ideology (besides its origins in Nudge) is a persistent drive to find the middle ground between libertarians and paternalists, Democrats and Republicans. This is what makes Book Two different: In Book Two, Sunstein and Thaler are not simply interested in correcting for cognitive error or improving the choice architecture of government programs, but in displacing ideological competitors on the right and the left. They are quite deliberately looking to flesh out that elusive "Third Way."

Book Two is a pretty far cry from Book One. True, Sunstein and Thaler are still aiming for moderation with libertarian paternalism--trying to find "the middle way," as it were. But they're pretty assertive about it. Call it assertive centrism. (5)

Book One and Book Two are not unrelated, but it will be part of my central point here that the not inconsiderable virtues of Book One (the Nudge) do not transpose easily into Book Two (Libertarian Paternalism). Part of the reason is that the spectrum on which libertarian paternalism registers--one defined by Republicans and Democrats--is just not all that interesting. Sunstein and Thaler are perhaps right to claim that we are tired of the needless polarization between Republicans and Democrats. But my guess is that we will also be tired of approaches that do little more than mediate between the two. The big problem then with libertarian paternalism is not that it is an oxymoron (something that Sunstein and Thaler explicitly acknowledge and address) but that it lies on the axis defined by its parent conceptions.

BOOK ONE: THE NUDGE

Imagine along with Sunstein and Thaler a tribe of "Econs." These are people who are unfailingly adept at making decisions for themselves. They are at the very least better at it than third parties (most notably, that third party known as "government"). Econs are incarnations of the familiar econtextbook persona known as the rational utility maximizer. According to Sunstein and Thaler, the problem with Econs is that, for the most part, they don't exist. Turns out that the world is populated mostly with "Humans." These are people who make predictable and recurring cognitive mistakes.

Humans are known to chronically underestimate the time it takes them to finish projects (the planning fallacy). They have a tendency to stick with the status quo even when it's not in their interest (inertia). They routinely assess risk based on the most recent, most salient, most accessible examples of those risks (availability bias). They estimate probabilities based on the degree to which the event is deemed as representative of the category (representativeness). In short, they make precisely the sorts of cognitive errors (there are many more) that hundreds of studies by psychologists and behavioral economists have diagnosed and documented over the last several decades. (6)

Based on this wealth of knowledge, Sunstein and Thaler recommend that government and other institutions take these errors into account in formulating programs and policies. Their idea is to nudge people (push them a little) into making the decisions that are right for them.

How should institutions do that? Enter here the idea of "choice architecture." With regard to many decisions that people make, they are offered a menu of options. This holds tree for choosing food in cafeteria lines as well as for the selection of savings plans and much more. Sunstein and Thaler's idea is that whoever is in charge of the choice architecture (the cafeteria supervisor or the government) should frame the choices so as to lead most of the Humans to make the choices that are right for them.

How does one do that? Well, start with their example of how not to do it. Imagine, for a moment, the perennially distracted law professor down the hall. Come this May, there is a damned good chance (you know this) that he will forget to choose a health plan. Being a good colleague, you will remind him of course, but chances are he will forget anyway. Now, if the default rule (the role that applies unless another option is deliberately chosen) is no plan, the law professor will be in trouble. The "no plan" default rule is plainly a bad one. Better to pick as a default some health plan (the cheapest one? the one last year?) than none at all. This is a nudge. In fact, this kind of nudge is the mother of all nudges--the default role. Picking smart default rules is a key way of nudging people.

But the default rule is hardly the only form of nudge available. There are a number of basic strategies for dealing with cognitive errors.

One approach is to bypass and even enlist cognitive errors. Take employment savings plans, for instance. People say they want to save, but then fail to follow through. They're subject to a loss-aversion error: they don't want to see their paychecks reduced. What to do? Well, Thaler and others have developed a "save more tomorrow" program, which enlists people to commit themselves to a series of savings contributions timed with pay raises in the future. In this way, people don't see their paychecks go down, and thus the "loss aversion error" is bypassed...