Wednesday, February 14, 2007

Corcoran on Pigovian taxes

Terence Corcoran doesn’t believe in Pigovian taxes. In his latest anti-Pigovian tax article he reasons that decreased smoking rates don't coincide with increased price rates, therefore we shouldn’t expect gas consumption to decrease in response to taxes on gasoline. Here’s an excerpt from his article in yesterday's National Post:

According to [Prof Greg Mankiw], "economists ... have found that a 10% increase in the price [of cigarettes] causes a 4% reduction in the quantity demanded. Teenagers are found to be especially sensitive to the price of cigarettes: a 10% increase in the price causes a 12% drop in teenage smoking."

In practice, however, the link between price and smoking seems a little less direct and a lot more uncertain. Real cigarette prices in Canada (using Ontario as a measure) have rolled up and down by as much as 80% over the last 20 years, with impacts on smoking prevalence that are far from clear. The link between price and
smoking is certainly far from the mechanical one implied by Prof. Mankiw...

We appear to be witnessing a decline in smoking, including among teenagers, but the movement in smoking rates seems strangely disconnected from price. Teen smoking declined between 1993 and 2001, for example, at a time when prices first fell and then remained flat. We also know that smoking is now subject to the greatest onslaught of regulation short of an outright ban. What role has price really played?

Could it be, in the case of gasoline taxes, that people have a much better idea of the economics of prices and demand than economists? Using price to change behaviour and shift demand forces all individuals to make major compromising decisions. Large numbers would have to continue to consume gasoline at high prices, forcing them to curb their consumption of something else. That's the result planners want, but it is not obvious to consumers that it will work or that it is fair. Look what happened to tobacco.

Not knowing a lot about the tobacco industry, I can’t argue with Corcoran too strongly, but he does fail to take into account the impact of increased cigarette prices on substitutes, something regulations were not responsible for (unless you account for medications to curb cravings, etc). I can think of a few examples. Statistics Canada finds that fine cut tobacco has been used as a substitute when prices are on the rise. As another example, price increases have allowed cheap generic brands of cigarettes to enter the market (even as a non-smoker I noticed their quick market entrance). And finally, how many Canadians attempted to cut costs (say, by purchasing tobacco in the tin and rolling their own cigarettes) rather than immediately quitting? In sum, Corcoran fails to mention that increased prices may have encouraged substitutes to emerge on the market, which is the same affect that a Pigovian tax should have.

Dependence on anything, whether it’s cigarettes or gasoline, cannot be cut overnight, but (to return to the theory on gas taxes), increased costs of gasoline should prompt industries to invest and switch to lower-cost fuel options.
Still, I’m not entirely sold on the Pigovian gas tax. As a low tax it’s ineffective, as a high tax it can be harmful. The only way I can conceive it as having higher benefits than costs is if corporate taxes were slashed and replaced (in part) with a Pigovian tax charged to industries, those most responsive to changes in natural gas prices.
Addendum: Prof. Brian Ferguson has a good post on Corcoran's article.

1 comment:

Marc said...

Corcoran fails to distinguish among multiple causes that drive a result. He is looking at simple correlations and trying to see causation. But it would appear he needs more independent variables in his equation: dollars per capita spent on anti-smoking campaigns; time trend (to reflect the general shift in societal attitudes; cost of substitutes (marijuana?); cost of complements (alcohol?). Only be controlling for factors such as these could you get proper results, which is probably what Mankiw did in the first place.