No Free Lunch, Part 1 million

It seems that Andrew Sullivan has joined Jonah Goldberg in calling for a gas tax.

Andrew puts forward an argument that I’ve heard from several conservatives, namely:

Higher gas prices were tough for a while for many people, but they reduced oil consumption, made alternative fuels more viable and helped the planet. Now, alas, the price has plummeted – and all those good green things are vulnerable to collapse. So wasn’t ten months ago the perfect moment to raise the gas tax? If done gradually, we could have essentially moved all that oil money from the pockets of people who hate us to the US Treasury – while avoiding a big price hike.

The problem with this argument is numerical, not conceptual.

Consider the limit case of no price elasticity – that is, no matter how high we raised the gas tax, people still used exactly as much of it as they would have with no tax. In this case, if the U.S. government added a $1 per gallon tax, the same number of gallons would be bought from the same suppliers who would make the same amount of money. The same amount of carbon would be emitted. A lot of tax would be collected from drivers.

But of course, this limit case is not reality. People do drive less when the price of gas goes up – just not that much less. Here is the long-term trend:

Even when faced with the huge price spike of the 1970s or the similar price spike over the past couple of years, demand just doesn’t respond dramatically to prices. The recent “mindset change” that pundits keep talking about is mostly rhetoric: U.S. gasoline consumption is estimated to have declined a total of about 3.4% in 2008 vs. 2007. The evidence of our modern history indicates that even spiking the price of gas by dollars per gallon doesn’t reduce demand by more than a few percent.

Of course, one might argue that while this is true for temporary price spikes, there would be much greater elasticity in the face of a long-term increase in price, as people buy different cars, move closer to work and so on. And he would be right. Academic analyses have generally estimated the long-term elasticity of gasoline demand to be 2 – 3 times the short-run elasticity.

Am I just therefore underestimating the creative forces that a big gas tax would unleash? Germany, as an example, consumes about half the total number of barrels of oil per person per day as does America. But of course they maintain a gas tax of about $7 per gallon. Not only this, but population density is more than 7 times that of the U.S. – this would be like the whole of the U.S. having the population density of Maryland or Connecticut. This naturally makes building out an extremely effective system of rail and mass transit feasible. Also, remember that the U.S. has GDP per capita that is more than one-third higher than that of Germany on a purchasing power parity basis.

In other words, if we had a $7 per gallon gas tax for decades, got people in the U.S. to live much nearer to one another, built a different kind of transportation infrastructure to exploit these new living arrangements, and accepted a lower material standard of living, it would be reasonable to assume that we could substantially reduce oil imports. That’s a lot more ambitious than keeping the price of gas at $4 per gallon. It’s also exactly the agenda of many progressive / liberal advocates.

A $2 per gallon a gas tax sustained for decades would surely reduce U.S. petroleum consumption somewhat, but logically by a small fraction of the difference between the U.S. and Germany. It might even be a good idea as compared to other ways to raise revenue, which is one thing it would certainly do effectively – mostly because burning a gallon of gas is so valuable to people that they’ll keep doing just about the same amount of it, even if you double the price. Of course, that’s also exactly why it’s not a very effective way to address projected global warming problems, or de-fund hostile regimes.