Commentary

Why Opting Out Is No “Third Way”

By Will Wilkinson
This article appeared on Reason Online on September 29, 2008

At first blush, “libertarian paternalism” seems a linguistic miscarriage, a self-crippling idea condemned to limp aimlessly in eternal darkness on the island of misfit creeds alongside “humanitarian sadism” and “color-blind racism.” But that hasn’t stopped Richard Thaler and Cass Sunstein, law and economics superstars at the University of Chicago, from pushing the catchphrase and concept as a solution to the nation’s problems for a half-decade now. And this year libertarian paternalism has achieved manifesto status with the new Thaler/Sunstein book, Nudge: Improving Decisions about Health, Wealth, and Happiness.

In Nudge, Thaler and Sunstein argue that new findings in psychology should be used to help people and thereby chart an exciting “third way” beyond the exhausted politics of left and right. The book offers a list of inventive policy tweaks, some with a welcome libertarian flavor. But the modesty of the proposals mocks the occasional grandeur of the rhetoric and should put to rest any hopes or fears that the authors’ brand of applied “behavioral economics” will soon transform the ideological landscape. Remember when that dork chariot, the Segway, was supposed to utterly reshape transportation? Libertarian paternalism is a lot like that: an innovative but overhyped dud.

Nudge is the highly anticipated book-length sequel to two 2003 papers, “Libertarian Paternalism” and the protests-too-much “Libertarian Paternalism Is Not an Oxymoron.” The papers and the book argue that we often make poor choices, and that unobtrusive fiddling with the context in which our choices occur (i.e., “nudges”) can improve our decisions and lives. If a cafeteria puts its key lime pie a bit out of the way, fewer people will succumb to delicious temptation. If employees are not required to fill out confusing paperwork to enroll in a savings plan, more of them will enroll. To grasp why these papers caused a bit of a sensation among policy wonks, despite their seemingly anodyne recommendations, you’ve got to understand the promise and threat of behavioral economics, a fast-growing discipline that enlivens the dismal science by probing the quirks of real-world decision making.

Begin with the influential Chicago school of free market economics and its Nobel-winning notables, such as Milton Friedman and Gary Becker. Suppose you assume that the best (or only) argument for something approaching social and economic laissez faire is the Chicago school argument, an argument that rests on the traditional Homo economicus model in which people are purely rational calculators. Now suppose you discover that Homo economicus is little more than a character in a math nerd’s fairy tale. You might well conclude that the case for laissez faire is doomed. That is indeed what many behavioral economists conclude, despite the fact that neither Adam Smith nor Friedrich Hayek relied on the existence of a hyper-rational economic calculator to make their cases for laissez faire. The behavioralists have now spent almost three decades enthusiastically enumerating the foibles of the hapless human mind in the service of the idea that we are, as Thaler and Sunstein put it, more Homer Simpson than Mr. Spock.

The great hope among many left-leaning behavioralists is that wider recognition of our earthbound limitations and self-defeating tendencies will loosen the grip of Chicago-style laissez-faire dogma in social and economic policy, clearing a little intellectual and political space for benign, welfare-promoting government regulation. The fear-shared by libertarians, liberals, and some of the behavioral economists themselves-is that exposing humans as “irrational” perpetrators of cognitive “anomalies” invites invasive control by paternalistic elites. Thaler and Sunstein’s libertarian paternalism is best understood as an attempt to hasten the hope-the death of laissez faire-while assuaging fears that our would-be rulers have been handed a dangerous intellectual weapon. “Emerging developments should strengthen, at once, the principled commitment to freedom of choice and the case for the gentle nudge,” they write. The “gentle nudge,” they assure us, is to be welcomed, not feared.

Most of the book goes something like this: First, tell a story about the lovably crazy things we “Humans” do-not at all like the choices made by those coolly calculating “Econs” of economics lore. Second, describe a clever reform in life strategy, business practice, or public policy that takes this foible into account in a way that encourages more optimal choices. Repeat.

Thaler and Sunstein’s best idea is the Save More Tomorrow Plan. Psychologists say folks really hate losing what they already have but don’t so much mind forgoing an equivalent gain. There is also a well-known tendency to postpone efforts at self-control; the diet always starts tomorrow. These findings, Thaler and Sunstein argue, recommend an ingenious scheme benevolent employers can use to help their workers put more money in the bank. Employees are offered the option to have a rising portion of future pay raises automatically deducted and placed into a savings or investment account. Since participants need do nothing now and are not required to cut back in order to save more, the percentage of total income saved increases more or less painlessly as pay goes up over time. Tests of the program show excellent results: Very few choose to stop their savings rate increases, and many end up maxing out their contributions to the plan. These results strongly support the idea that people would save more if savings programs were structured to work with the grain of our predictably crooked timber.

Rather than denying that behavioral economics implies some kind of paternalism, Thaler and Sunstein try to defang paternalism by arguing that it is unavoidable. Its inevitability allegedly follows from a core finding of behavioral psychology: Context matters. The way people envision their choices partly determines what they end up choosing. If you modify the context-the lights go down, the volume on Peaches and Herb goes up-you can subtly bias the choice, for good or ill. To refuse to intervene in the context of choice is simply to acquiesce to the accidental biases of the status quo. That’s no way to conduct a successful seduction, and it’s no way to increase the personal savings rate.

Thaler and Sunstein introduce the notion of “choice architecture,” the practice of manipulating the context of choice in order to manipulate (or nudge) choosers. Choice architecture, it turns out, includes anything you might do to alter someone’s environment in a way that affects what they do. “A good rule of thumb,” the authors say, “is that ‘everything matters.’ ” Which is to say, nothing doesn’t. If you have ever made a shopping list, adjusted a thermostat, pointed to the nearest rest room, or arranged your living room furniture, you’re a choice architect yourself. You may be a bit of a paternalist too, since paternalism is nothing more or less than the idea that it is OK to do something-choice architecture-that we cannot avoid.

“The paternalistic aspect” of libertarian paternalism, say Thaler and Sunstein, “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.” So a shopkeeper “putting fruit at eye level” is a paternalist. The designers who made the iPod especially user-friendly are paternalists. (These are the authors’ examples.) The trouble is, influencing people’s behavior to make their lives better isn’t called “paternalism.” Usually it’s called “helping.” If you happen to be some kind of actual designer, be it of urinals, user interfaces, or public policy, then it’s called “your job.” But that really doesn’t make you a professional paternalist.

Paternalism, in English, is interference (coercive or not) with the exercise of another’s freedom for his good, as judged by the one interfering (or his principal). But Thaler and Sunstein say “a policy is ‘paternalistic’ if it tries to influence choices in a way that will make choosers better off, as judged by themselves” (italics in the original)-a perplexing inversion. In the centuries-old debate over the justification of paternalism, the whole issue has been whether it is permissible to help someone by overriding her judgment. Since the authors show an otherwise fine feeling for English, the reader is left suspecting an unstated motive for their audacious redefinition. Perhaps these eager choice architects think we will be made better off if the meaning of “paternalism” is gutted and remodeled.

As for “libertarian,” what they mean is “choice-preserving.” This at least is a sense of the word. When applied specifically to government, it’s close to the main sense. But Thaler and Sunstein intend it to apply generally-to the policies not just of governments but of human relations directors and mothers. And yet a mom who keeps cookies in a high cupboard rather than not buying them at all is “libertarian” in about the same sense that a bossy roommate who bogarts the TV remote is “fascist.” Since this casual use of libertarian forces a confusing distinction between libertarian libertarians, like me, and nonlibertarian libertarians, like Thaler and Sunstein, we might be better off just leaving it out of serious social thought.

“Libertarian paternalism” turns out to translate into something like “choice-preserving helpfulness” or, more naturally, “helping people without taking away their options.” Hey, I’m for that. But who isn’t? Paternalists, that’s who! Indeed, Thaler and Sunstein’s philosophy when spelled out in nontendentious terms sounds a lot like plain old benevolent liberal anti-paternalism. The crucial conceptual difference is that “libertarian paternalism” pictures all manner of helpful guidance-meddlesome or imperceptible, persuasive or compelled-as falling along a common gradient of paternalism. Thaler and Sunstein argue powerfully for the desirability of staying on the choice-preserving end of that gradient, and this may succeed in making some of us less nervous about standing on the gradient at all. But the deeper point seems to be that we are on it, no matter what. Once that is accepted, we are left, as the old joke goes, “haggling over the price.”

Thaler and Sunstein go out of their way to build some libertarian cred. They come out in favor of increased school choice, offer extremely helpful advice about setting up the default investment options in a system of personal retirement accounts, and argue the state ought to get out of the business of deciding who can and can’t be legally married. And they advocate a principle of “one-click opt out” that keeps it easy to say “no thanks” to a nudge.

But run-of-the-mill libertarians would be more reassured had Thaler and Sunstein done more to recommend “choice-preserving” alternatives to existing hard-paternalist measures. As the George Mason University economist Bryan Caplan suggested to me, perhaps you could legally purchase cocaine if you wrote an essay for the Drug Enforcement Administration explaining that you grasp the risks but want to purchase some anyway. Maybe we could exempt ourselves from seat belt laws by watching an hour of bloody car crash videos. Surely there is some clever way to opt out of Food and Drug Administration restrictions in order to preserve dying patients’ ability to choose experimental drugs and treatments. A liberty-minded “libertarian paternalist” would focus more than Thaler and Sunstein do on reinstating choices that the state has already stripped away.

The authors make a great deal of the idea that there is no “neutral” default rule-any way you arrange the desserts, you are biasing the outcome. This point deserves more attention than they give it. Suppose President John McCain were to implement a policy of opt-out national service. We might reasonably object on the grounds that it would all too clearly communicate that individuals need reasons not to serve the state. If allowed to stand, such a policy could shape social expectations and individual preferences in a direction at odds with individual liberty. Soon enough we might find ourselves asking, “Why should you be able to opt out at all?” The initial paternalistic nudge may “leave the choice open,” but accepting the legitimacy of certain nudges may eventually diminish the value we place on liberty.

Thaler and Sunstein’s suggestion to increase the supply of transplant organs by changing the default rule to “presumed consent” instead of nondonation may leave you with similar thoughts. To say that an individual’s body is common property by default is to make a statement with cultural consequences that reach beyond the policy’s immediate effects. Anyway, why not just legalize markets in organs and tissue, a genuinely libertarian form of choice architecture likely to have even better results?

Despite the welcome weight Thaler and Sunstein throw behind a few genuinely liberty-enhancing policies, the thrust of the conceptual renovation behind the term libertarian paternalism is to empower, not limit, political elites. In the libertarian paternalist scheme, the rules will allow for an easy opt-out only so long as those writing the rules happen to care about preserving choice. The “libertarian” part of the equation is secured by good will alone. If that falters, we are left with paternalism, plain and simple. Old-fashioned libertarianism, it turns out, had a great deal to say about the choice architecture of politics. Given the intense attractions of power and the limits of benevolence, liberty (and therefore the commonweal) is best secured by setting in place a structure of rules that strictly limits the discretion of the powerful. That includes the discretion even of choice-loving technocrats.

All Nudge really offers is the idea that we should use what we know about how people act in order to design policy that will help them. And we shouldn’t take away their choices. Well, I’m for all of it. If some policy makers really are misled by unrealistic economic models of rationality, then they should cut it out and bone up on psychology. Behavioral economics provides valuable new information about what will and won’t work, and why. Thaler’s work on savings plans is a great example of what can be done by taking into account new findings in psychology.

But it’s no source of ideological realignment, no basis for what Thaler and Sunstein call a “Real Third Way.” This book offers some whiz-bang behavioral economics that can be used for any ideological end, and it gives us the agreeably banal doctrine of choice-preserving helpfulness. The whiff of paradox in “libertarian paternalism” may have set up hopes for a category-defying revolution, but Nudge is the book where those hopes, and that tiny monster of an idea, prove flightless.

Will Wilkinson is a research fellow at the Cato Institute.