Thursday, June 11, 2009

Drinking, rationality and taxes

In a post entitled The Economics of Cocktails and Taxes, Jodi Beggs at Economists Do It With Models gives an interesting set of predictions regarding the effects of a recent sales tax increase on the price of cocktails. Here are the three models with which Beggs does the question of "What happens to the price of a cocktail when taxes increase?"

1. Econ101 -- Basic supply and demand. Under this model, a tax drives a "wedge" between the price suppliers receive and the price consumers pay. This standard "wedge" theory implies that the producers pass some (but not all) of the tax on to consumers. In this way, the two groups split the check, perhaps unevenly, to be sent to the government (Beggs gives a really nice graph to this effect).

2. Irrationality -- People are drinking, after all. Under this model, drunkards who are consuming cocktails do not pay attention to the costs of their habit. In such a world, people will be too blazed to notice that they pay more at the end of the night. The bar owners can charge their customers the whole of the tax increase.

3. Menu Costs -- Changing the labels on the menu is costly. Under this model, bar owners will worry that the higher price drives away customers. They want to offer discounts to attract more customers, but marking the discounts costs more than it is worth. In Beggs' words, the bar owners "eat the whole cost of the tax" if the menu price includes tax, but the consumers will "eat it" if the menu price does not include tax.

The article is entertaining and well-written, but I can't shake the feeling that this argument is incomplete in its resorting to irrationality. I also think a more complete menu costs model does a better job of explaining what's going on. In this post, I make the argument against irrationality. On Monday, I give my version of the menu costs explanation.

On Irrationality: It might be true that some form of irrationality explains the problem, but as a rule, I make irrationality my last explanation, rather than my first (or second). I don't usually buy irrationality arguments because purported irrationality can explain any phenomenon. Don't believe it? Here are some examples:

Why did people get a mortgage they couldn't afford? They were irrational.
Why do people get married? They are irrational.
Why do people put money in a 401K and not look at it again? They are irrational.
Why does Tony attend University of Chicago, instead of some place that will push him less? He is irrational.

What's the problem with the irrationality explanation? Isn't it still useful to explain what's going on? When using models, we're interested in more than just an explanation -- we want to predict. The trouble with irrationality is that it cannot be predicted. People are irrational for idiosyncratic reasons that depend on context. Change the context and you change the prediction.

Even if we really expect people to be irrational, we'd like to have an explanation that we can apply to other settings. That's why we economists do it with models in the first place: we commit to a simplified version of the world to see if we can get by without resorting to irrationality. If we can get by without too much irrationality, we can trust the predictions more.
With the drunkards, let's try an alternative explanation out for size. People might not care that much about prices while they are at the bar, but when they get home, they realize what they have done. Or, when they contemplate going out on the town, they weigh the costs and benefits. People are foresighted and they also know their own habits better than anyone else.

A drunken night on the town might cost $100. In deciding to go out on the town, a person budgets (either explicitly or implicitly) that expense. We can expect that people will go out on the town less frequently if there is a sales tax, but they still get just as blazed when they do.

People might be fully irrational while drunk, but that need not imply that their drinking behavior be irrational. This is the same logic that led Nobel Laureate Gary Becker to develop a theory of rational addiction: people weigh the costs and benefits of acquiring addictions. Why shouldn't they weigh the cost of going out on the town?

No comments:

Post a Comment

Please feel free to share your ideas about this post in the open forum. Be mindful that comments in this blog are moderated. Please keep your comments respectful and on point.