Smoking Bans & Externalities

I want to follow up on the discussion over at The Volokh Conspiracy that was started by Orin Kerr yesterday regarding smoking bans. I also posted on this discussion yesterday from the perspective of individual liberty. The issue of externality market failure seems to be a significant part of the discussion. For example:

There is a market failure because the major cost of a smoking bar - smelly clothes and hair - occurs after a short period of time. Since people typically go to several bars in one night, the value of any single bar being non-smoking is compromised. Thus, bar owners have an incentive to orient their bar towards the interests of smokers, who will always prefer a smoking bar instead of towards non-smokers who will only prefer a non-smoking bar if all of the bars they will be attending are nonsmoking. Aside from distorting incentives for any particular bar, this also creates an entry barrier for non-smoking bars as there must be several such bars before the value addedby being non-smoking can be realized. The result is that virtually no bars are non-smoking in cities where smoking is permitted.

Since the market cannot properly serve consumer preferences, the real question in terms of efficient allocation of resources to meet consumer preferences is whether the preferences of non-smoking bar goers to not be subjected to smokey clothes is greater than the preferences of smoking bar-goers to be able to smoke in bars. Unfortunately this issue cannot be fairly decided electorally becasue numerous non bar goers vote and infrequent bar goers get equal say as regulars. I am not quite sure how to discover this answer.

Also we find:

There are established, emperical economic &social costs to smoking that far, far outweigh anyone's so-called "right to smoke". When the day finally comes--and it will--that smoking is largely forgotten, noone anywhere will miss it.

I suspect that what we have here is a nice example of EXTERNALITY ABUSE.

Consider the idea that the "social costs" far outweigh the "right to smoke." To put this just a bit more in the language of economics, the idea would seem to be that the "social costs" far outweigh the smoker's benefits. Such a suggestion is either incomplete with respect to the idea of externality market failure or it is simply misguided. Actually this sort of statement abounds in policy discussions regarding externality market failure, and it seems to be based on the mistaken belief that any social cost implies an externality market failure. Of course, this is not the case because social cost represents the opportunity cost of using resources in a specific way. This opportunity cost, or social cost, is always positive. So, noting there are social costs has nothing to do with discovering whether there is an externality market failure. Instead, when an externality market failure is present, it will be the case that the marginal social cost exceeds the marginal private cost, and the difference will be the marginal external cost. With an externality market failure there will be a divergence between the marginal social cost and the marginal private cost, where the marginal private cost is the cost perceived within the market.

Consider now the possibility that we do have an externality market failure at issue, which is assumed by the first quotation above. My best definition of an externality market failure is that we have an (1) unintentional, (2) nonmarket (3) interdependence between economic decision makers. In the case of smoking in a restaurant, I suspect that these three elements are not all present. There is certainly an interdependence between a smoker and a nonsmoker in a restaurant that allows smoking. However, it seems to me that we cannot say this interdependence is unintentional. If the smoker and the nonsmoker both voluntarily choose to enter the restaurant, then both these economic decision makers must know there will be the interdependence in question because the smoker chooses to light up. This is really the point you can see made in the comments at Volokh Conspiracy that the nonsmoker has no costs imposed by others because the choice to enter the restaurant is made voluntarily. We might also want to say that the interdependence at issue does not meet the nonmarket condition. The restaurant itself seems to be a market setting. People choose restaurants based upon consideration of a broad range restaurant attributes, which probably include prices, menu items, entertainment, whether children are present, and whether smoking is allowed or not. These attributes essentially define the brand supplied by each specific restaurant, and as such, these attributes are part of the market. The interdependence at issue seems to me to be an intentional market interdependence between smokers and nonsmokers. Therefore, I conclude we do not have an externality market failure to be worried about with this policy issue. (I wonder. . . .I think many people like to eat at restaurants that do not have kids present because they can be noisy and disruptive. If we think that smoking in restaurants represent an externality market failure, then perhaps in the name of correcting market failure we will need public policies that prohibit kids in restaurants as well.)

Let's return briefly to the idea that with a externality market failure there would be a divergence between marginal social cost and marginal private cost. The idea is that there are people not party to market exchanges that are worse off because of the exchanges. This results in some part of the entire social cost of the market activity not being part of the market exchange values. This doesn't seem to be the case with our smoking policy concerns. The so-called external costs are costs "imposed" on nonsmokers in the restaurant. The nonsmokers are themselves part of the market exchange in question. The nonsmokers are well aware of the "costs" they bear when they enter a restaurant with smokers. The nonsmokers will see these costs as part of the price (subjective as it may be) they pay for their choice to eat at a restaurant with smokers. In other words, all of the social costs are perceived by the economic decision makers in these market exchanges, and there is no divergence between marginal social cost and marginal private cost.

Finally, while my analysis is that we do not have an externality market failure to be concerned about, let's suppose that smoking in restaurants could be characterized to involve a negative externality. Would the appropriate economic policy response to correct this market failure be to ban smoking? This is unlikely. If we did have a smoking externality, then we would have something very much like a pollution externality. Consult any microeconomic principles textbook on the question of the optimal (efficient) amount of pollution, and you will discover that in general the optimal amount of pollution is not zero. A smoking ban only makes economic sense if the optimal amount of smoking pollution is zero. For the optimal amount of smoking pollution to be zero it would have to be the case that the marginal damage caused by the first, very little puff of exhaled smoke was greater than the marginal benefit to the smoker of that first drag off the cigarette. This condition is very unlikely to be the case. Instead of a smoking ban, if we assume there is a negative externality involved, the typical policy response that economists look to would be to tax the choice that causes the external cost. By the means of taxation it is then possible to get the optimal (efficient) amount of smoking in the restaurant. If we consider the policy proposal of banning smoking in restaurants from the perspective of the economic idea of an externality market failure, then we are led to be concerned to get the optimal (efficient) amount of smoke in the restaurant. We are not led to argue in favor of no smoke in the restaurant.