Tuesday, January 23, 2007
Libertarian Paternalism and Vice Policy Robustness
Professor Becker, Judge Posner, and Professor Sunstein have had a blog exchange concerning “Libertarian Paternalism.” This phrase comes from work by Professors Sunstein and Thaler, and it concerns policies that push folks in a certain (presumably desirable, for typical people) direction, without limiting their freedom to choose otherwise. As a form of Libertarian Paternalism lies at the heart of my approach towards vice policy, I thought that I might say a little bit more about it.
The name I have adopted for my approach to vice control is the “robustness principle.” The robustness principle states that a vice regulatory regime should work well irrespective of the precise extent of rationality or addiction associated with vice.
The main rationale for the robustness principle lies in ignorance. We can’t easily judge when a habit becomes an addiction, or when rational consumption involves dynamic inconsistency or shades into compulsion. So we want to avoid a regulatory regime that only makes sense if there is no such thing as vice rationality, or an alternative regime that only works well if everyone makes considered, sober judgements about his or her vice participation. What we tend to end up with when we avoid these extremes is vice controls that offer some assistance to those who are misinformed or struggling with self-control issues, as long as those controls do not impinge significantly upon those who are rationally vicious. We entreat and induce but we do not compel -- a’ la libertarian paternalism.
Pure laissez faire towards adult vice if not so attractive, even if we depart from it through mandatory information provision. The difficulty with laissez-faire lies in the affinity of addiction to disease, and the problems with vice self-control that arise among non-addicts; in John Stuart Mill’s terms, vice consumers might often be in some state “incompatible with the full use of the reflecting faculty” -– and hence at least partially exempt from the deference that generally should be paid to adult self-regarding decisions. (And if laissez-faire is a first-best strategy, competing jurisdictions that impose differing robust regimes will eventually reveal that fact –- in the meantime, the departures from those first-best free market polices will not be very costly, given the criterion of robustness.)
The robustness-based vice exemption from our usual deference to adult self-regarding behavior is only a partial exemption. Pre-emptive controls on vice decision making hold the potential to be extremely oppressive, as Mill noted: “The preventive function of government...is far more liable to be abused, to the prejudice of liberty, than the punitory function; for there is hardly any part of the legitimate freedom of action of a human being which would not admit of being represented, and fairly too, as increasing the facilities for some form or other of delinquency.” Those policies that help guide vice decision making in the direction of rationality will become very expensive (in terms of the welfare of rational vice producers and consumers) if they establish substantial barriers to informed use. So unless we are absolutely certain that there is no such thing as rational, informed vice consumption, significant impediments are unwise.
A robust vice policy regime should stand up pretty well if our knowledge or situation changes. And our knowledge and our situation are constantly changing. Tomorrow we might learn that moderate alcohol consumption has more severe negative health effects than our current understanding indicates. The next day evidence might arise that moderate ecstacy consumption promotes mental health with little risk of addiction. A robust regime has already taken these possibilities – and their opposites, that alcohol has more benefits and ecstacy more costs than previously believed –- implicitly into account. We are quite unsure about the extent of rationality governing the use of these substances, so robustness instructs us to choose policies that operate effectively whether the case for rational use improves or deteriorates. (This property of hardiness in the face of altered circumstances is not exhibited by either broad vice prohibitions or laissez-faire.) There might be good reason to adjust even a robust regime at the margins if our understanding of costs and benefits changes – but not to radically revise that regime. Robust rules build-in substantial tolerances for errors in our understanding.
Robustness is a useful precept beyond vice policy, particularly where there exist significant departures from perfect information. A major virtue of democracy as a form of government, for instance, is that it is robust with respect to the personal qualities of politicians: democracy operates well when enlightened leaders are at hand, and it also works tolerably (though less well) when leaders are shortsighted or cruel or venal. Democracy represents a compromise: democratic institutions purposely make it harder for exceptional leaders to guide a country in desirable directions, to ensure that a bad person temporarily in charge will not be in a position to inflict enormous damage. A theoretically better system would be one with expansive executive powers when an enlightened leader is in charge, but much more limited powers when a mediocre or diabolical person holds the reins. But we cannot easily judge (or agree upon) who is enlightened and who is diabolical, so democracies institute a system of checks and balances that constrain leaders of any stamp. The theoretical benefits of basing the extent of power granted upon the character of the current executive are not available in practice. Similarly, the optimal vice controls that would not interfere with rational adult choices while guiding the decisions emanating from diseased or irrational minds are not viable in practice.