A new analysis of gas prices in U.S. cities finds that even minor jumps in fuel costs send people to buses and rails
Americans love to gripe about high gas prices, but they actually pay some of the lowest fuel costs in the world [PDF]. Part of the reason for this hidden discount is that lawmakers have refused to raise the federal gasoline tax since 1993. In fact the tax has lost value over time, since it's not even indexed to inflation; it sits at a flat 18.4 cents per gallon. That's to say nothing of the unaccounted social costs of traffic or the environmental costs of pollution. If gasoline were priced fairly in the United States, one has to wonder whether or not America's love for driving would remain so bold.
That question is at the heart of a recent analysis conducted by Bradley Lane of the University of Texas at El Paso. Lane examined fluctuations in gas prices in 33 U.S. cities during a period stretching from January 2002 to March 2009. He then compared these changes to transit ridership patterns in the same cities over the same time. In all cities he looked at bus ridership, while in 21 places, including Los Angeles and Chicago and Washington, he considered rail travel as well.
All told, Lane found a pretty strong link between changes in gas prices and shifts in transit ridership. Every 10 percent increase in fuel costs led to an increase in bus ridership of up to 4 percent, and a spike in rail travel of up to 8 percent. These results suggest a "significant untapped potential" for transit ridership, Lane reports in an upcoming issue of the Journal of Transport Geography. In other words, a significant part of America's love for the automobile may only be its desire for inexpensive transportation.
"Despite this being one of the most driving-oriented societies in the world, despite the fact that we have a lower national priority for transit than just about every developed society in the world, despite the fact driving is essentially free in our minds compared to any other mode, in some cities you still see some pretty large responses to gasoline prices," says Lane. "So despite the game being tilted totally in favor of auto use, gasoline price fluctuation in the past 7 or 8 years actually appears to have a pretty significant, consistent effect on limiting how much people drive."
Lane's analysis revealed two key relationships between gas prices and transit ridership. The first is what he calls an elasticity, which is essentially a behavioral response to an event. In this case the event is a change in gas prices, and the repsonse is a shift in transit ridership. The second is what he calls a "lagged effect." That means that some elascities — such as switching your commute from car to train — don't appear until several months after the initial change in fuel cost.
Take, for instance, the case of bus ridership in Atlanta. There Lane discovered three significant behavioral elasticities at three distinct temporal lags. The first, which occurred at 0 months (or roughly the same time as the fuel hike), saw a roughly 20 percent jump in bus ridership. The second, coming at 6 months, saw a 32 percent transit rise, and the third, at 11 months, a 12 percent spike. Over the course of about a year, then, one major rise in fuel cost in Atlanta led to about a 64 percent rise in bus ridership.
(Technical sidenote: Elasticity works both ways — so a drop in fuel costs would lead to a change the other direction, away from transit — but the important thing here is the direct connection between gas price and transit ridership. Lane did discover some negative elasicities, meaning an inverse relationship between fuel cost and ridership, but those were rare and possibly a limitation of the mathematical model.)
When Lane mapped the cumulative elasticities for each city in his study, he found some interesting patterns. Most notably, he found big behavioral responses to gas prices in places like Omaha, Des Moines, Kansas City, and Indianapolis — cities one typically thinks of as car-centric. "What that tells me is that there is actually a greater sensitivity to fluctuation of gasoline costs in cities that tend to be more auto-dependent," says Lane. "That to me is very interesting. People will go to transit even when there really isn't much transit to go to."
Here's the map for cumulative change in bus ridership in response to fluctuations in gas prices (the larger the circle, the greater the response to fuel costs):
And here's the map for shifts in rail transit:
The upshot of this analysis is a recognition that automobile use does not occur in isolation. It's strongly tied to both gasoline prices and the quality of the public transit system. Increase the first and improve the second, says Lane, and you may well find that America's love for the road is founded less on hard concrete than on an artificially soft market.
"We typically associate high automobile use in the U.S. with Americans' need to drive and love to drive. But ultimately there's a pricing and policy structure that enforces that," says Lane. "If we fully costed out some of the impacts on driving and had any inhibitions on car use — not to the level of inhibitions on public transit now; I'd never wish that on anybody — but simply had some way to make automobile travel more difficult and more expensive, and gave an alternative in the form of public transit or denser neighborhoods or shorter multimodal trips, then you could really see a pretty large change."
Photo credit: Reuters/Joshua Lott