Mandates and Bans vs. Nudges: When to Escalate Interventions

On Mandates and Bans: Beyond Nudging?

We have seen that libertarians are worried that we will start with a nudge and then move on to shoving. Some critics of a more progressive persuasion have the opposite worry, namely that we will stop at nudging when much stronger measures are required. Some people even think that if governments nudge, they won’t do anything more, even if nudges are palpably inadequate. Instead of getting serious about climate change, maybe regulators will rest content with energy efficiency labels!

If we (or anyone, for that matter) thought that most of the world’s problems could be fully addressed with light-touch interventions, then this would be a serious concern. Murder, rape, assault, and theft are criminal offenses, properly met with coercion. Some problems, such as pollution, arise because people are imposing harms on others. As we have emphasized, nudges are not an adequate approach to such problems. To be sure, they might help. You can impose taxes on gasoline and encourage people, through nudges, to buy fuel-efficient cars. There are plenty of externality-reducing nudges. But by themselves, nudges are not enough to control externalities.

Nudging might be thought of as a Swiss Army knife type of tool. These knives are designed to be versatile and can be highly effective in certain circumstances, such as when a can has to be opened or a screw tightened. In the right situation, nudges can also accomplish quite a lot at very low cost. But as we have repeatedly stressed, taxes, subsidies, mandates, and bans also have their place. Graphic warnings do not accomplish all that should be done to reduce cigarette smoking. Changing the default does not eliminate the shortage of organs for transplant. And when it comes to climate change, we have spent an entire chapter emphasizing why we will need jackhammers and bulldozers, with pocketknives helping where they can.

We very much doubt that the use of nudges is likely to discourage officials from taking stronger measures. (Sunstein worked in the U.S. government for four years, and he never saw that happen. Not once.) A nation might impose severe alcohol taxes, nudge people not to drink and drive, and top it off by giving offenders stiff fines if they get caught driving while intoxicated. Countries in Scandinavia do all three. A nation might nudge people to buy fuel-efficient cars, for example with fuel-economy labels—while also imposing high gasoline taxes or subsidizing electric cars. A nation might make it a crime to use certain drugs—while also nudging people not to use those drugs. It might seem clever to suggest that the availability of nudging will discourage the use of more aggressive instruments. In the history of the world, that might have happened—but realistically, we’re talking about a debater’s point. It’s not a serious concern.

Still, reasonable people can certainly disagree about when we should move from nudges to bans and mandates (or vice versa). To give one example, our friend and colleague David Laibson was involved with Thaler on the design of the retirement savings plan at a foundation. The plan has a well-designed inexpensive default, and a small number of other curated investment options. The question on which they (mildly) disagreed was whether eager participants would be allowed to access what is called the “mutual fund window,” in which a large number of other funds are available. They agreed that the employees who used this option were unlikely to make great investment decisions, and on average would probably do worse in the long run (as judged by themselves, of course) than if they stuck to the basic offerings. They also agreed (based on empirical evidence in other settings) that only a small percentage of employees would make use of this option, especially if there was a bit of sludge involved in accessing the window, such as a warning label. Laibson would have excluded the option, Thaler would have allowed it, but neither is sure he is right. (On this one, Sunstein agrees with Laibson. Majority rule?)

This is an illustration of a more general dilemma. Enthusiastic paternalists who are especially worried about the risk of poor choices might want to push for bans and mandates. If Humans really make errors, why not protect them by forbidding them from erring?

The truth, of course, is that there are no hard-and-fast stopping points. We have defined libertarian paternalism to include actions, rules, and other nudges that can be easily avoided by opting out. We do not have a clear definition of “easily avoided,” but we hold up “one-click” paternalism or its equivalent to be as close as we can get with existing technology. (We can hope for “one-thought” or “one-blink” technology in the future.) In many domains, it’s best to allow people to go their own way at the lowest possible cost. To be sure, some of the policies we have advocated impose higher costs than one click. To opt out of an automatic enrollment plan, an employee might have to fill out and return some form—not a big cost, but more than one click. It would be arbitrary and a bit ridiculous to offer an inflexible rule to specify when costs are high enough to disqualify a policy as libertarian, but the precise question of degree is not really important. Let us simply say that in general, we want those costs to be low. The real question is when we should be willing to impose some nontrivial costs, or even very high costs, in the interests of improving people’s welfare.

Consider a class of regulations requiring “cooling-off periods.” The rationale is that in the heat of the moment, consumers might make ill-considered or improvident decisions. Self-control problems are the underlying concern. A mandatory cooling-off period for door-to-door sales, of the sort imposed by the United States Federal Trade Commission in 1972, provides an illustration.18 Under the commission’s rule, any door-to-door sale must be accompanied by a written statement informing the buyer of his right to rescind the purchase within three days of the transaction. The law came about because of complaints about high-pressure sales techniques and contracts with fine print. Again, a cost-benefit test, looking at the benefits for those who are helped and the costs for those who are not, could be used to decide when such laws would be imposed. Using such a test, regulators would want to consider how big the imposition is on those who have to wait a few days to receive the product, and how often buyers would want to change their minds. When the costs are low (did anyone ever really need to complete the purchase of an encyclopedia right away, even before Wikipedia was online?) and there are frequent changes of heart, such a regulation makes sense to us.

For certain fundamental decisions, often made on impulse, a similar strategy might well be best. Some states impose a mandatory waiting period before a couple may get divorced.19 Asking people to pause and think before making a decision of that magnitude seems like a sensible idea, and outside of very extreme situations, we are hard-pressed to think of why anyone would need to change their marital status immediately. (True, spouses sometimes really don’t like each other, but is it so terrible to have to wait a short while before the deed is done?) We could easily imagine similar restrictions on the decision to marry, and some states have moved in this direction as well.20 Aware that people might act in a way that they will regret, regulators do not block their choices but do ensure a period for sober reflection. Note in this regard that mandatory cooling-off periods make best sense, and tend to be imposed, when two conditions are met: (a) people make the relevant decisions infrequently and therefore lack a great deal of experience and (b) emotions are likely to be running high. These are the circumstances in which people are especially prone to making choices that they will regret.*

Cooling-off periods are a relatively mild intervention. When should society take the next step to imposing bans and mandates? Again, reasonable people can differ on where they would draw that line. Our lodestar is human well-being, recognizing that reasonable people also differ about how to understand that concept. For example, fellow behavioral scientists Nick Chater and George Loewenstein have argued that the defined-contribution retirement plans we have discussed in this book allow too much opportunity for errors, and that they should be replaced by mandatory savings schemes at set saving rates.21 They praise the system in Australia, which is compulsory and does not permit people to borrow against their savings.

Now, it is true that a mandatory system will, by definition, achieve 100 percent compliance, at least for those employed in the formal economy. The comparable UK NEST system uses automatic enrollment and achieves over 90 percent participation. Is it obvious that this is worse? An evaluation will necessarily depend on how highly to value freedom of choice and how much harm will come to those who decide to opt out. It is at least possible that when people opt out, they do so for good and sufficient reasons, and some evidence supports that speculation.22 For example, people might opt out because they really do need the money now, or because they have their own, separate retirement plan—possibilities that help explain why it is often a good idea to let people go their own way.

Mandates and bans are certainly justified in some contexts. We have said that if people’s choices result in harm to others, we might want a prohibition (consider laws against assault and theft) or a corrective tax (consider greenhouse gas emissions). But even in such contexts, nudges can have an important place.

If people are making stupid, myopic, or self-defeating decisions, we wouldn’t put mandates off the table.23 We do not object to social security programs, to bans on trans fats, to energy efficiency mandates, to laws that require motorcyclists to wear helmets or drivers to wear seatbelts, or to requiring people to wear masks in public during a pandemic. If people’s choices impose serious harms on their future selves (consider smoking), we might want to go beyond nudges (we favor cigarette taxes and bans on smoking in restaurants). The only point is that taxes, mandates, and bans raise distinctive problems and concerns. So long as people are making informed decisions about how to live their own lives, we would favor an attitude of humility and respect—and hence a presumption in favor of freedom of choice. But that presumption certainly protects your right to disagree with us.