In the final section of the paper, we offer a few suggestions about how to resist the slippery-slope tendencies of new paternalism (p. 737-739):
How, then, might we protect ourselves against paternalist slopes? We have three recommendations, addressed both to the new paternalists themselves and to those who might be persuaded by them. These recommendations are intended to lower the probability of adopting new paternalist policies to begin with, but also to help resist more intrusive policies after initial policies have been adopted.
1. Have Reasonable Expectations of Decisionmakers
One lesson of behavioral economics is that we cannot reasonably expect decisionmakers to carefully consider the full ramifications of their choices in light of the best available evidence. Instead, they economize on information by using choice heuristics, and they sometimes myopically focus on present and concrete problems while ignoring more distant and abstract ones. This is no less true of public decisionmakers (including voters, politicians, judges, bureaucrats, experts, and rent-seekers) than it is of private citizens. Indeed, the problem is likely worse for public decisionmakers, because they lack the incentives to discover and control their own cognitive limitations. Private decisionmakers at least face the costs and benefits of their own mistakes, and thus have an incentive to correct them.
It is therefore insufficient to ask policymakers to carefully weigh the costs and benefits of each new paternalist proposal. The “careful, cautious, and disciplined approach” advocated by Camerer and coauthors is rather unlikely to guide real-world policy. We should not expect policymakers to weigh all the economic, scientific, and psychological evidence objectively, to stand on nuanced distinctions, and to adopt policies that carefully target just those people who need help most. We should expect policies to be blunt instruments.
2. Reject the Paternalism-generating Framework(As usual, full citations are available in the full paper. Cross-posted at ThinkMarkets.)
The new paternalists say that the framing of problems can affect decisions by emphasizing certain aspects of a situation and downplaying others. As we have argued earlier (see especially Section V.C), the new paternalists themselves have framed the public-policy debate in a manner that emphasizes opportunities for intervention while downplaying or ignoring private alternatives. Adopting that framework increases our vulnerability to slippery slopes.
In contrast to this paternalism-generating framework, we recommend a slope-resisting framework—one that emphasizes the limitations of public policy and the potential for private solutions. In this alternative framework, both private and public decisionmakers are understood as having essentially the same cognitive defects; they also have various tools for self-correction. For private decisionmakers, the tools include resolutions and commitments, conscious construction of their environment, and voluntary submission to social controls from third parties. For public decisionmakers, the tools include procedural, substantive, and attitudinal limitations on the scope and extent of government action.
3. Maintain Important Distinctions
Slippery slopes, including paternalist ones, can sometimes be resisted by standing on easily enforceable bright-line rules. One such bright-line rule is the distinction between public and private decision-making. Another is the distinction between coercive and non-coercive intervention. John Stuart Mill enunciated these distinctions in terms of the Harm Principle, which says that restriction of individual choice is justified only on grounds of harm to others. He argued:[The individual] cannot rightfully be compelled to do or forbear . . . because, in the opinions of others, to do so would be wise, or even right. These are good reasons for remonstrating with him, or reasoning with him, or persuading him, or entreating him, but not for compelling him, or visiting him with any evil in case he do otherwise.Of course, Mill understood, and we agree, that if a person harms himself and in so doing violates his legal responsibilities to others, he ought to face the relevant legal penalties. But the State has no legitimate interest that can be advanced through coercion strictly in the prevention of harm to oneself. To the extent that policy adheres to this principle, the paternalist slope will obviously never get started.
We do not contend that a single violation of Mill’s Harm Principle will send us hurtling toward heavy-handed paternalism in all areas of life; if that were true, we would already be doomed. We do contend that increasing numbers of such interventions, passed under the guise of helping people do better by their own preferences, and without any recognition of the lines being crossed, will tend to create momentum toward further interventions. Keeping the Harm Principle clearly in mind—and recognizing any given restriction on autonomy (however small) for what it is—will, we hope, retard movement down the slope.
To some extent—especially in Sunstein and Thaler’s book Nudge and Daniel Ariely’s book Predictably Irrational—the new paternalists have presented their position as self-help advice. That is, they offer behavioral economic insights into achieving better self-control and personal management. Under the Harm Principle, such efforts are perfectly unobjectionable. They fall in Mill’s category of remonstration, reasoning, persuasion, and entreaty.
The problem, as we have argued, is that the new paternalists do not clearly distinguish private, voluntary efforts from public, mandatory ones. Instead, they deliberately construct a continuum from soft to hard paternalism (see, especially, Section III.B). They define freedom of choice in terms of the cost of exercising a given option, without regard to whether the costs are imposed coercively or by the voluntary choice of resource owners. In this way, they effectively erase a reasonably bright-line rule—the distinction between private action and state coercion—and purposely replace it with a gradient. They also regularly present public and private, and coercive and non-coercive, paternalistic activities alongside each other, without recognizing any important distinction between them, and often simply ignoring the coercive aspects of their policies (e.g., the way in which allegedly pro-employee policies limit the freedom of the employer).
We suspect the new paternalists resist bright-line rules and encourage gradients because of an unavoidable feature of rules: they nearly always err by both over- and under-inclusion. A rule that allows private paternalism but not public paternalism would admit some varieties of paternalism that new paternalists might oppose, such as Walmart’s restrictions on what sort of movies it will stock; and it would disallow some varieties of paternalism they favor, such as mandatory terms in employment contracts. That is, however, the price of having rules. The compensating advantage of rules (or at least one advantage) is providing a bulwark against the problems of vagueness, including the threat of slippery slopes.
Bernard Williams’s distinction between logical and effective distinctions is frustrating, because it means we cannot always rely on the normative distinctions that make most sense to us. But it is also enabling, because it reveals that some distinctions may be useful—that is, effective—without being strictly logical. Thus, even if the new paternalists do not think the public–private and coercive–non-coercive distinctions track their ideal notions of right and wrong, such distinctions might nevertheless be practical as guides for law and policy.