Robin Hanson's ongoing discussion of positional goods, signaling, and consumption externalities at the new, exclusively Robin Hanson Overcoming Bias has been terrifically stimulating. I have a few thoughts about the relationship between externalities and “harm” that Robin's discussion has stirred up.
Suppose you're a Millian liberal devoted to the harm principle, which goes like this:
That principle is, that the sole end for which mankind are warranted, individually or collectively, in interfering with the liberty of action of any of their number, is self-protection. That the only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others.
Now suppose you, like many economists, are inclined to leap a bit hastily from “negative externality” to “harm.” And then suppose you, like Robert Frank, have a strong view about consumption externalities. When I buy a really expensive car, this story goes, it subtly shifts my community's frame of reference for signals of social status and for judging the the adequacy or socially acceptability of certain vehicles. By diminishing the status signal sent by older and/or less expensive models, my choice exerts a subtle pressure for others to increase spending simply to maintain the constancy of the signal sent by their cars. If we decide to count this kind of effect as a “harm,” then sumptuary taxes pass Mill's test and therefore do not count as paternalistic. We need not even torture the language and call regulation to reduce consumption externalities anything stupid like “libertarian paternalism,” since it's not paternalism at all! Sure, the person facing a steep tax on luxury may be helped by a fiscal inducement to stay off the conspicuous consumption gerbil wheel, but that's by the bye. The point is to prevent “harm” to others through individual consumption choices.
As Robin has been insistently pointing out, how good-looking we are, the quality of our mates, how smart and funny we are when we talk, and the impressiveness of our children's achievements signals at least as strongly as our cars. If our investments in appearance, mate selection, Bourdieuian cultural capital, and children are not equally harmful, then why not? If you think regulating luxury consumption passes Millian muster, then why wouldn't regulating extremely impressive feats of oratory or athleticism?
I think this line of thinking can be taken even further. Many so-called “culture wars” are largely about cultural externalities. Consider Linda Hirschman-like arguments to the effect that women who choose to stay at home raising children impose a significant cost on women who wish to pursue professional success by reinforcing traditional stereotypes of women's relative strengths and by creating rational expectations among employers that firm-specific investments in female employees will have a lower than average expected payoff due to the possibility of maternity leaves or long-term exit from the labor market. The argument that stay-at-home moms ought to be stigmatized, or at least be extended decreasing levels of social esteem, is basically an argument for the cultural version of a tax on choices that have negative spillover effects for others. If the state took a side on this and actually did tax stay-at-home moms, would that pass Millian muster, on the grounds that mothering choices have spillover effects that “harm” other women?
I think that at this point Mill would suggest that something has gone dreadfully wrong. It looks like we've defined “harm” so loosely that the harm principle, so understood, could be the basis for the state regulation of any action whatsoever that affects anyone else in a way they don't like.
Consider pecuniary externalities. If I open a hot dog stand across the street from your hot dog stand, I will take some of your business, or force you to cut your margins, and thereby make you poorer. Have I “harmed” you in some way that requires that you be made whole, or that suggests the wisdom of the state's preventing future instances of such harm? The law says no, and the law is right. You have no right to local monopoly profits from hot dog sales. Indeed, pecuniary externalities are so valuable that there is a whole body of antitrust law ostensibly intended to promote them.
Now, when a black family moves into a neighborhood of white racists, thereby causing great unhappiness, or when the recognition of the legitimacy of gay marriages causes traditionalists to feel that their traditional marriages have been “devalued,” that's the cultural analogue of a pecuniary externality. Somebody really is getting hurt in some real sense. But I don't much care, and Robert Frank probably doesn't either, if some racists are disgruntled by their neighbors' color, or if some religious folk feel aggrieved by a perfectly accurate sense that the social esteem afforded their particular type of marriage has fallen in relative value.
Coasean logic focuses us on the duality of externalities. In the land of the deaf, there is no noise pollution. In the land of cosmopolitan enlightenment, there is no “there goes the neighborhood.” Progressive social change occurs through a revaluation of where to locate “the problem.” Is it in the signal or in the receiver? To identify a “harm,” and to invoke the harm principle, the moment there is a complaint, is the essence of reactionary politics. It is to shut down the very possibility of relocating “the problem” from the source of a reaction to the reaction itself. This would be the very opposite of the intention of Mill in On Liberty, which is at bottom a call for the cultural version of dynamic, ideally competitive markets roiling constantly with the hurt of lost market share.