We're now living with higher (real) gasoline prices than at any time in living memory. The media love to provide anecdotes about how specific people are coping, with the implication that surely these sky-high prices will reduce Americans' "love affair with their cars." Don't be too sure. For one thing, Europeans have been living with much higher gas prices (due to much higher taxes) than ours for decades, yet as those countries have become more affluent, their use of automobiles has become more and more like ours.
But more seriously, what do we know about the relationship between higher gas prices and Americans' driving behavior? And also about their choice of vehicle? The Congressional Budget Office released a pretty good study (.pdf) on these questions in January. Their findings are worth pondering.
The study used data from 2003 to 2006, during which time real U.S. gasoline prices doubled. And because Caltrans has detailed data on trips and speeds on Southern California freeways, CBO used their data to measure changes in driving behavior. What they found was a surprisingly small impact on that behavior. Over the period in which gas prices doubled, motorists on some freeways made slightly fewer trips and drove slightly less fast (at times when the freeway was uncongested). Specifically, on those freeways where there was a parallel commuter rail line (as is the case with several in the database), for every 50-cent increase in the price of gas, the number of freeway trips decreased by 0.7%. No such decrease was seen on the rest of the freeways. And for every 50-cent increase, the median speed on uncongested freeways dropped by 0.75 mph. This latter result is pr edictably small, since there is little change in gallons used per mile within the speed range of 30 mph to 60 mph, and the value of fuel savings is very small compared with the value of most people's time.
How generalizable are these results? First, most freeways in most metro areas don't have parallel rail lines, so the likely trip-reduction effect (based on these findings) is zero. And the speed reduction of less than 1 mph applies only on uncongested freeways, a declining portion of the total freeway system most places. So it's pretty unlikely that gasoline price increases in the range we've seen recently will have much impact on driving behavior.
However, the second part of the CBO research looks at what is likely to be a longer-term impact: choice of vehicle. And even though we have only three years of data, there are signs that price changes of the magnitude we've seen are affecting what people buy. First, the market share of cars relative to light trucks (pickups, SUVs, and minivans)-which had been declining steadily since the early 1980s-turned upward in 2005 and has remained above its 2004 low-point since then. The biggest single change is from vans and minivans to large passenger cars. Second, overall new-vehicle fuel economy is finally starting to rise again, after having been in a shallow decline since 1987. Third, the prices of used cars reflect the increased importance of fuel economy, with large SUVs and luxury cars declining in price, while full-size and mid-size cars show slight increases - presumably reflecting changes in demand.
These results reinforce the idea that automobility is highly valued in America, and hence that people will adjust to higher fuel costs primarily by seeking out more fuel-efficient vehicles rather than curtailing their driving.