In today’s New York Times, Thomas Friedman has a column lamenting the failure of politicians to enact a gasoline tax following 9/11. I am a strong supporter for a dramatically increased gasoline tax in the United States. The problem with Friedman’s proposed gasoline tax is that it is not ambitious enough.

Here are some data:


Sources:

U.S. gasoline prices
U.S. gasoline usage
U.S. GDP

On September 11, 2001 U.S. gasoline averaged $1.15 per gallon. By May 1, 2006 it was $2.90 per gallon. The difference of $1.75 per gallon is larger than the gasoline tax proposed by Friedman, and thus allows for a natural experiment on the effects on consumption.

US gasoline consumption in September, 2001 was 8.6 million barrels per day. In May, 2006 it was 9.3 million barrels per day. This does not suggest a strong relationship between price and consumption, although it is certainly possible to argue that consumption would have been higher with lower prices. Clearly, the $1.75 increase per gallon did not lead to reduced consumption.

One can look at the figures from the standpoint of the overall economy as well. In 2001 the US economy generated $3.22 of economic activity for every gallon of gasoline burned. In 2006 it generated $3.91 of economic activity for every gallon of gasoline burned. This does not suggest a strong relationship of gasoline use and economic growth. This is good news because it suggests that gasoline prices might be able to increase considerably without large negative economic effects.

One might argue that the $1.00 per gallon tax would be on top of the supply/demand fluctuations in price. But even gasoline at $4 or $5 dollars is unlikely to lead to dramatic changes in behavior or innovation.

Consider that a gasoline tax of $1/per gallon would have raised only about $3.4 billion in tax revenue in 2006, which is small in relation to overall U.S. incomes taxes, which in 2006 were more than $1 trillion (PDF). Thus it is very (!) misleading when Friedman quotes Philip Verleger in his column as saying, “We could have replaced the current payroll tax with a gasoline tax.” Well, I suppose that if the gasoline tax was about $300/gallon under present levels of consumption then that statement would be accurate!

What does the literature say?

There is a very nice review paper on the elasticity of gasoline demand based on a wide range of studies by Graham and Glaister (2002)(PDF). This paper concludes:

There are differences between the short- and long-run elasticities of fuel consumption with respect to price. . . Therefore, it may be right to say that “it won’t make much difference” or “people will use their cars just the same”, but only in the short run. The evidence is clear and remarkably consistent over a wide range of studies in many countries that in the long run there is a significant response, albeit a less than proportionate one. . .

So the effects of a gasoline tax are important and take place over the long term. However, the tax would have to be significant enough to generate significant responses, lest it be more symbolic than effective. I am not sure what that is in the United States, but I am sure that $1 per gallon is only a step in the right direction; it is not all that is needed by a longshot.

Both long- and short-term effects of gasoline prices on traffic levels tend to be less than their effects on the volume of fuel burned. . . Raising fuel prices will therefore be more effective in reducing the quantity of fuel used than in reducing the volume of traffic. . .

Anyone who has driven at rush hour in the UK where gasoline costs a lot more than the U.S. will be well aware of this reality. It is therefore misleading to suggest that a higher gasoline tax will reduce congestion, as some have suggested. It won’t. To reduce congestion would require other strategies.

The demand for owning cars in heavily dependent on income. . . The implication is that fuel prices must rise faster than the rate of income growth, even to stabilise consumption at existing levels.

Consider the difference between dollars of GDP per gallon in 2001 and 2006. US GDP increased over this period by 30% while gasoline prices increased by about 90%, and gasoline consumption still increased by 7%.

If goals of energy policy are to dramatically reduce the U.S. reliance on foreign sources of oil and to rapidly accelerate the decarbonization of the energy system, then a gasoline tax can certainly contribute to that end. However, it is misleading at best to suggest that $1 per gallon can do the job, or make a big step in that direction, when it can’t. Achieving a gasoline tax in the United States would be a monumental political achievement. It would be a shame to see such an achievement undercut by getting the policy wrong by reaching for too little. Of course, that might just be a good characterization of current debates on U.S. energy policy generally.