Disagreement Case Study: Robin Hanson and David Balan

The basic challenge posed by Robin is this.  To support the imposition of a paternalistic government policy on Peter, one must believe that: (i) the government is sincerely motivated by Peter’s welfare; and (ii) the government knows what’s best Peter better than Peter himself does, even taking into account the fact that in the absence of paternalistic policy Peter need not rely only on his own knowledge, but is free to seek the uncoercive advice of anyone willing to give it to him, including the government itself.

To my mind, point (ii) is the easy part.  There really are people who left to their own devices will ingest poisinous miracle cures and the like, and who really would be better off if they didn’t do so, and for whom actually existing paternalistic policies are the only hope of being saved.  Of course there are some cases where the government really just doesn’t know better and gets it wrong.  Robin and I would agree that these are the cases where policy should be restrained, and if it’s not restrained it’s more likely to be an abuse of power issue than an ignorance issue.  So in my view the real action is in point (i), the extent to which the government will shun restraint and abuse or misuse the coervice power that paternalistic policies give it.

This is a problem; there are good reasons to be afraid of government power, which has been and continues to be hideously abused in large parts of the world.  In my debate with Robin I repeatedly argued that things are different in "well-functioning" societies like the United States, by which I meant that such societies are likely to have governments that, humanly and imperfectly, can both gather the necessary expertise and exercise the necessary restraint so as to successfully carry off a modest paternalism mission.  Robin correctly responded that the mere fact that we are a more or less well-functioning society does not prove that paternalistic policies are good; maybe we’d be in even better shape without those policies.  And as Bryan Caplan pointed out after the debate, not even the fact that there have been big reductions in the kinds of bad things that paternalistic policies are supposed to prevent (people getting poisined by food and drugs, people getting fleeced in pyramid schemes, poverty among the elderly, etc.) proves the point; those improvements might simply be due to the fact that we have gotten richer and better educated over time.

So why do I think that well-functioning liberal societies can get this stuff more or less right when I concede that most societies across space and time get it wrong?  The basic answer is that non-liberal societies get everything wrong.  In my view, the ascendence of the Enlightenment principles on which modern liberal socities are based is the most important thing ever to have happened in the history of the world.  There is an almost complete overlap between the set of countries that are based (in practice, not just on paper) on Enlightenment principles and the set of countries that are fit to be inhabited by normal human beings.  The empirical evidence of the success of the Enlightenment project in general couldn’t be more overwhelming.  So the fact that that there are other paternalisms out there that I would disapprove of no less than Robin does (say a religiously-based  controlling of Peter’s sexual behavior to save him from hellfire) doesn’t strike me as a strong argument for doubt about paternalistic policies that I might consider favoring.

This doesn’t solve the problem of "how can I be so sure I’m right that I can justify imposing my opinion by force," but it does give it different contours.  We should think about paternalistic policies the same way we think about other policies whose effectiveness we can’t be completely sure of.  We don’t look for guidance to all the many failed societies.  We use some combination of reason and evidence, including reason and evidence about potential abuses of government power, and we do the best we can.  And we have made progress.  We know more than we did in the past about the kinds of things that lead people to self-destrictive behavior, and we are reasonably good at parcelling out government power so that it doesn’t get abused too much.  This is new, and it makes all the difference in the world.  The fact that we are not perfect at it is both an argument for restraint and an argument for trying to get better at it.  If the Enlightenment project progresses even more, I’d probably favor even more paternalism.

GD Star Rating
Tagged as: ,
Trackback URL:
  • http://profile.typekey.com/robinhanson/ Robin Hanson

    So what exactly is a “liberal” society, and what his this “overwhelming” “empirical evidence” “that non-liberal societies get everything wrong”? Is it more than the fact that we are rich?

    Also, I will keep repeating that the issue isn’t about who knows more; rational but ignorant people would not need paternalism. The issue must be some kind of irrationality that afflicts ordinary people more than governments.

  • Matthew C


    Cost / benefit analyses of the FDA point at hundreds of thousands of deaths because life-saving pharmaceuticals have been delayed by many years after European agency approval. The numbers lost dwarf any estimates of lives saved due to the additional safety protocols. For example,

    Gieringer (1985) used data on drug disasters in countries with less-stringent drug regulations than the United States to create a ballpark estimate of the number of lives saved by the extra scrutiny induced by FDA requirements. He then computed a similar ballpark figure for the number of lives lost owing to drug delay:

    “[T]he benefits of FDA regulation relative to that in foreign countries could reasonably be put at some 5,000 casualties per decade or 10,000 per decade for worst-case scenarios. In comparison, it has been argued above that the cost of FDA delay can be estimated at anywhere from 21,000 to 120,000 lives per decade. . . . .Given the uncertainties of the data, these results must be interpreted with caution, although it seems clear that the costs of regulation are substantial when compared to benefits.”

    Doesn’t this fact cast your faith in the benevolence of FDA paternalism into doubt?

  • Telnar

    One of the factors worth weighing in deciding the optimal level of paternalism is that paternalistic policies are always at an important disadvantage compared to alternatives like a store for banned products if you’re willing to assume (though no paternalist would) that Peter is as good at making or delegating decisions as the paternalist.

    That disadvantage is that Peter is an expert in his own situation. Peter knows minute details about his preferences, his experiences, his skills, his wealth, and many other things which might bear on his choice. Of necessity, the person setting paternalistic policies knows none of those things.

    Paternalistic ignorance of even large details which vary from person to person is true for both practical and political reasons. In practice, not enough studies have been done to a give a paternalist confidence in the conditional probability that people with a given (detailed) set of characteristics would make better decisions for themselves than those which would be made by a general rule. Let’s say that the data did exist, though. I still doubt that it would be politically feasible to say that many FDA regulations did not apply to anyone with an IQ over 130 (although some SEC regulations don’t apply to those with a net worth over $1 million, so perhaps I’m speaking too quickly).

    So, when something is harmful to some but not all people, regulating it will inevitably harm all of those who it benefits while only helping some of those who it harms (since most such people would have avoided it on a recommendation only).

    The trick here in doing a cost benefit analysis is that both the harm and the benefit are likely to happen to relatively small groups (compared to the much larger group who have no desire to do something once they know the government strongly discourages it — whether it’s banned or not). I don’t have confidence that voters who are mostly in the group unaffected by the regulation will do a good job of weighing the harm which might happen to those who would make foolish decisions absent a regulation against the harm done to those who would make sensible decisions absent a regulation.

    Making that decision well will hinge on understanding how much people vary (and therefore how much harm is done by policies which attempt to treat them as homogenous).

  • http://profile.typekey.com/nicholasshackel/ Nicholas Shackel

    Whilst I agree with the problems that Robin poses for government paternalism I think there is a more fundamental criticism available. For the sake of argument suppose that the government really did know what was best for everyone and would run society to achieve the best available approximation to that best, that still wouldn’t make government paternalism right.

    Non-aggression against persons is a fundamental duty that is grounded in the self ownership of persons, just because aggression infringes their self ownership. Coercion is an aggression. There is therefore no right to coerce someone except when that someone will otherwise aggress against another. Consequently, there is no right for a government or anyone else to coercively save anyone from their own folly. Government paternalism is saving people coercively from their own folly. Therefore government paternalism is morally wrong and all such laws are oppressions.

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    David, here is the question I should have asked earlier: Whatever the “overwhleming empirical evidence” you have in mind, does it convince a majority in other societies today, or would it convince such societies in the past, that we are far superior paternalists?

  • http://profile.typekey.com/sentience/ Eliezer Yudkowsky

    Balan, in my view by far the most convincing argument against regulation is that the market is a delicate mechanism which even economists don’t fully understand, and regulators are bulls in a china shop who lack even the incentive to understand what they’re tampering with because their career progress is based on media coverage of how much their proposed policies seem like good ideas to reporters.

    “Of course there are some cases where the government really just doesn’t know better and gets it wrong. Robin and I would agree that these are the cases where policy should be restrained, and if it’s not restrained it’s more likely to be an abuse of power issue than an ignorance issue.” That’s like saying: “Of course there are some stocks that go down, and both Robin and I would agree you shouldn’t buy those stocks.” The controversial part is the idea that you can tell the difference in advance, or even, in some cases, afterward – most people don’t even realize that the FDA is killing more people than it saves, because the noneffects of medications not approved don’t make the evening news, which is what regulator incentives are all about.

  • http://profile.typekey.com/halfinney/ Hal Finney

    I see two issues here (and note that this is not a meta-comment unfortunately, but goes to the meat of the debate). The first is whether, in general, government paternalism would help with regard to the specific issues where the paternalistic policies apply. Does it really make people better off on average for them to be forbidden to use certain recreational drugs, for example? Are most people benefited by the forced savings imposed by the Social Security system and similar programs in other countries? I think you can make a decent argument that most such policies do have positive effects.

    The second issue is larger, and it asks whether paternalism is wise even if we stipulate that specific paternalistic policies improve social welfare. This includes questions about the effects of losing personal autonomy and responsibility; and possible pernicious implications of having a larger and more intrusive government, which may go beyond paternalism into oppression and domination. That’s a much more difficult area to consider because of the size and scope of the questions raised.

  • Scott Wood

    Surely the rhetoric of (ii) is misleading. Agreeing that there are Peters in that circumstance is a very diffeent thing that asserting that the government knows who those Peters are.