Libertarian Paternalism

by David Gordon

Thaler and Sunstein have set themselves a seemingly impossible task. Paternalists maintain that it is sometimes justifiable to interfere with someone's freedom, if doing so will promote his own good. Smokers, putting aside the issue of secondary smoke, do not violate others' rights: they harm only themselves. Nevertheless, a paternalist about smoking would think it justifiable forcibly to prevent people from smoking. Libertarians deny that such interference is acceptable. Force may be used only in response to aggression.

Given these uncontroversial characterizations of the two positions, is it not obvious that they cannot be combined with each other? To devise a libertarian paternalism seems no more promising an endeavor than to construct a square circle. Our eminent authors, though, are not convinced: libertarian paternalism is exactly the position they wish to defend.

Their escape from apparent contradiction is ingenious. It is indeed unlibertarian, they say, to use force to compel someone to act for his own good. They do not favor doing so; but this leaves them free to support a less exigent variety of paternalism. It is all right to render it difficult for people to make certain choices, as long as doing so does not impose substantial costs on them. People, as their title suggests, may be subject to paternalistic "nudges," so long as these nudges do not coerce them.

Libertarian paternalism is a relatively weak, soft, and nonintrusive type of paternalism because choices are not blocked, fenced off, or significantly burdened. If people want to smoke cigarettes, to eat a lot of candy, to choose an unsuitable health care plan, or to fail to save for retirement, libertarian paternalists will not force them to do otherwise — or even make things hard for them… A nudge, as we will use the term, 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. (pp. 5–6)

There is a problem here that Thaler and Sunstein fail to note. Not all cases of using force on people impose substantial costs on them. Suppose that you are about to reach for a cigarette, and I hold your wrist to prevent you from doing so. You were aware that I might do this and could easily have avoided my company. Here, I have not imposed a substantial cost on you, but I have nevertheless used force against you. Given cases of this kind, their alternative ways of stating their proposal, i.e., not imposing substantial costs and nudging but not using force, are not always consistent.

They extend their view further. It is also justifiable to nudge people to act in a "good" way, even when doing so is not for their own good. In current American law, people are free to direct that their organs be made available for transplants after they die. Unless, though, they have filled out a donor card, doctors who wish to transplant organs must secure the consent of whoever has legal custody of the body. Thaler and Sunstein suggest that we change the default position. Why not allow transplants, unless someone has signed a declaration forbidding that his body be used in this way? They think it likely that many more transplants would be obtained under this system. But libertarians can support it, because it forces no one to donate. If you do not want to do so, all you need to do is sign a statement to that effect. (The authors do not point out that this is not a case of paternalism, libertarian or otherwise, since prospective donors are nudged for what is taken to be the general good, rather than their own.)

An obvious objection to their proposals arises, and their efforts to respond to this objection form the theoretical substance of the book. Let us return to those who, despite the health risks, continue to smoke. Is it not, in a free society, up to them to assess the costs of doing so, weighing them against what they gain from smoking? What right have other people to a say in the matter? Are not paternalists, even of the soft variety who confine their ministrations to nudges, substituting their preferences for the freely chosen decisions of others?

The authors respond that this objection rests on an unrealistic view of choice. Economics textbooks assume that people act on their preferences rationally, based on full information.

Whether or not they have ever studied economics, many people seem at least implicitly committed to the idea of homo economicus, or economic man — the notion that each of us thinks and chooses unfailingly well, and thus fits within the textbook picture of human beings offered by economists. (p. 6)

But a great deal of research, by both economists and psychologists, shows that the assumptions of the standard view cannot be retained in unmodified form. People often make mistakes in logic. Further, they are often divided against themselves: an impulsive self struggles against another self that takes greater account of the long run. (Thaler and Sunstein do not suggest that rationality always requires suppressing the impulsive self, but often it does.) As if this were not enough, people often lack the information required for an informed choice.

Thaler and Sunstein are right to think that the standard model is flawed, but they themselves remain too much its prisoner. What they in effect are saying is that unless someone meets the textbook criteria for rationality and information, he is not really choosing in the full sense. If, e.g., someone commits a "logical" error, in the extended sense they give to logical error, he is not giving expression to his freely chosen preferences. What he "really" wants is that his preferences be fulfilled in the way best fitted to do so. His faulty reasoning does not count as part of his free choice.

Further, people's ultimate goals are often not the concrete objects that they seek to obtain: these are only means to their ends. A smoker does not want a pack of cigarettes as an end in itself. Rather, he wants certain feelings, e.g., pleasant sensations or relief from unpleasant cravings, which he thinks the cigarettes will provide him. Thaler and Sunstein thus have yet another way to question people's choices: perhaps people have selected a poor way to achieve what they "really" want.

To reiterate, in their view, only actions that meet rigid requirements count as full choices. Smokers, research indicates, haven't fully taken into account the heath risks of smoking. Thus, they cannot be said "really" to choose to smoke. Further, people are often subject to so-called "framing" effects: they will "choose" differently when confronted with identical options, depending on how the options are presented. Choices in these circumstances, Thaler and Sunstein aver, are problematic: how can we say that people in the grip of conceptual illusions are freely choosing?

What is left? Given the authors' wide net, few actions count as rational choices. There is thus practically unlimited scope for the state to suppress liberty: in doing so, it is not interfering with what the self "really" wants. True enough, the authors preach a mild doctrine. Nudges, not force, are on their agenda. But they lack a rational basis for this limit. If people do not "really" choose their actions, why not forcibly restrict them? After all, doing so may enable them better to achieve what they "really" want — as experts, suitably instructed by Thaler and Sunstein, determine.

The authors consider a related objection, but they do not fully grasp the key point. They answer worries that libertarian paternalism will lead to more severe restrictions by treating this complaint as an ordinary "slippery-slope" argument. Why, they ask, need one progress down it? Their proposal endeavors to be both libertarian and paternalistic: why assume that the paternalistic aspects must supplant the libertarian?

[O]ur own libertarian condition, requiring low-cost opt-out rights, reduces the steepness of the ostensibly slippery slope… Slippery slope arguments are most convincing when it is not possible to distinguish the proposed course of action from abhorrent, unacceptable, or scary courses of action. (p. 237)

But the real issue is not the inevitable progression of the slippery slope but the rational basis for the restriction in the first place.

Those who wish to preserve liberty must take people's actions as they find them, not substitute for them "better" or more "rational" actions, based on an assessment of what people "really" want. To return to the transplant case, if the state says to people that their organs will be taken from them unless they explicitly direct otherwise, it is claiming to set forward the terms under which people can retain control of their own bodies. This is hardly libertarian. Instead, the state needs to step away entirely and allow people to dispose of their organs as they wish. Why not rely on a free market in organs, rather than concoct schemes to restrict liberty in the guise of preserving it?[1]

Thaler and Sunstein might respond in this way: Even if one grants (as of course they would not) that their scheme threatens liberty, it has not been shown that their view of choice is wrong. Is it not obviously true that people often act impulsively or illogically, in ways that they later come to regret? Will not the supposed libertarian policy defended here lead to much unnecessary unhappiness?

Libertarians need not deny obvious facts. People often do regret their choices. Those who find convincing the explanations of bad choices put forward by Thaler and Sunstein are free to make arrangements with others that will alleviate these problems. If you think that sudden impulses when confronted with tempting food will lead you to fall off your diet, you may contract with a friend to forfeit money should you fail to meet certain weight requirements. But, in a free society, doing so is up to you; the state may not nudge you into this sort of contract. The authors might answer that decisions on whether to restrict one's future choices are themselves less than fully rational and informed; but to say this is merely to reiterate their original argument, and the libertarian rejoinder to it is unchanged. Also, they do not present any evidence that choices of this kind are flawed by their criteria.

Thaler and Sunstein offer a further argument for the nudges they favor. They suggest that influencing choice is unavoidable: if we do not nudge people to make good choices, we will influence them to make bad ones. Sometimes their contention has merit. Suppose an employer has a voluntary plan that allows workers to save for retirement. He has a choice: either he allows his employees to sign for the plan or he selects the plan as the default option, allowing workers to remove themselves. He cannot avoid the choice altogether, so long as he wishes to offer the plan. But not all cases are like this, as the transplants example illustrates. Here, once more, the state does not have to choose a default option. It can avoid doing anything at all.

So far, I have not questioned the evidence Thaler and Sunstein offer that people act irrationally but have instead tried to show that, accepting their evidence, their case for libertarian paternalism has not been established. Sometimes, though, the evidence for irrationality, taken in their economic textbook sense, is weak.

They offer as a case of irrationality purchasing an extended warranty for household appliances. In many instances, the cost of the warranty, combined with the small chance that the appliance will break down, suggests that buying the warranty is a poor option. But what if the purchaser has a strong aversion to paying for repairs when an appliance has broken down? He strongly prefers that payments for breakdowns be handled in advance. There seems nothing "irrational" about this preference, but if someone has it, purchasing the extended warranty makes sense. Thaler and Sunstein may not share this preference, but their doing so is not a requirement of rationality in preferences.

$25 $22

The authors make their case in part by rhetoric, not argument: they cite an episode of the Simpsons that makes fun of these warranties. Richard Epstein has written a detailed response to many of the arguments for irrationality that Thaler has advanced on previous occasions: Skepticism and Freedom (University of Chicago Press, 2003). Though the authors cite two papers by Epstein, they do not respond to this book or even mention it in their bibliography.

Tocqueville long ago warned against the policies of which libertarian paternalism is an example:

Above this race of men stands an immense and tutelary power, which takes upon itself alone to secure their gratifications and to watch over their fate. That power is absolute, minute, regular, provident, and mild… The will of man is not shattered, but softened, bent, and guided; men are seldom forced by it to act, but they are constantly restrained from acting. Such a power does not destroy, but it prevents existence; it does not tyrannize, but it compresses, enervates, extinguishes, and stupefies a people, till each nation is reduced to nothing better than a flock of timid and industrious animals, of which the government is the shepherd. (Alexis de Tocqueville, Democracy in America, Volume II, Section 4, Chapter 6.)