Marginal Utility

by Ken Houghton

I mentioned this EnvEcon post earlier. It assumes several things:

1. That the status quo is efficient (the market is at equilibrium)
   
2. That the quotes from the article—all of which are from smokers and their suppliers—indicate what will happen (e.g., that a store where 70% of its revenues are from cigarettes "might just have to quit selling cigarettes.");
   
3. That those quotes and the localized result are representative of the overall state, and, most importantly,
   
4. That the Kentucky tax is also optimal for that state.

(In fairness, Dr. Whitehead does note that "an increase in Indiana's gas tax might encourage Indiana smokers to support their local smoke shop." But one suspects he does not mean this as a policy proposal so much as a snipe. He also provides some back-of-the-envelope analysis at the bottom of the post that indicates that he knows the quotes do not represent the expected behavior of the total population.)

Now don't get me wrong; this is a traditional Marginal Utility analysis—but it is applied to a smaller sample and tacitly assumes that all changes to the status quo should be not just optimal, but Pareto-optimal.

Let's now also look at this:

West Virginia and Kentucky, states known for high levels of obesity,diabetes and smoking, have the highest proportion of people with heart disease in the nation, health officials said Thursday.

From a Utility point of view, the costs of behavior should cover the incremental cost of participating in that behavior. (a variation is the Pigouvian Tax argument) Otherwise, we must conclude that there is a "free rider" issue (or, a subset of same, that "externalities" are not being covered).

Given that the prevalence of heart disease, per the CDC study mentioned, is 2.0% greater in KY than that in Indiana, perhaps the argument should be that the KY tax is too low (does not cover the full social cost of smoking) rather than that the IN is or will be too high?

After all, consider workers at a KY-based firm against those at an IN-based firm. If insurance costs are calculated optimally, it will be less expensive to do business in IN. However, non-smokers (the large majority of the population) in both states will pay a higher premium solely because their coworkers are more likely to require costly care. This is, of course, no less of a "tax," but the worker is provided with an even lower Utility.

(In fact, if we look at the county-by-county maps of deaths from heart disease in Indiana for the 1990s (from this CDC site), we can see that the area discussed in the article cited by Professor Whitehead is the highest in the state, matching the legendarily-dangerous and dirty steel-processing area around Gary. So IN workers treated as a whole are paying extra for the "free riders" on the KY border, whose behavior is clearly not socially optimal.)




Also, given the incremental costs, and absent charging full costs for behavior, is this not a reasonable alternative step?

After being thwarted for years, a bipartisan group of members of Congress reintroduced legislation yesterday that would allow the federal government to further regulate the tobacco industry by cracking down on marketing aimed at young people and requiring that reduced-risk tobacco products back up their claims with science.

The Altria Group, the company formerly known as Philip Morris, is among the bill’s biggest supporters.

Labels: externalities, free riders, marginal utility, Optimal Resources, Pareto-optimal, Pigouvian tax