What's the Driving Force Behind Less Driving?

Last time, I showed you evidence (courtesy of Robert Puentes and Adie Tomer of the Brookings Institution and Adam Millard-Ball and Lee Schipper of Stanford University) that driving per person seems to have peaked in the 2000s and now may even be dropping. This bucks every travel trend we’ve seen since Henry Ford got to work. What might be slowing down the acceleration in driving?
Are higher fuel prices to blame? Perhaps in part. There is some debate on the magnitude of the effect gas prices have on fuel consumption, but luckily you don’t have to throw out your basic economics textbook; people do drive less when the price of driving rises. A fairly recent study by economists Kenneth Small and Kurt van Dender found that a 10 percent increase in gas prices leads to a 0.2. to 0.3 percent reduction in driving in the short run, and an eventual reduction of 1.1 to 1.5 percent.
But does this explain the driving slowdown? Maybe partially, but not entirely. The growth of driving began to abate around 2000, and driving flattened out around 2004; the big gas price hikes didn’t come until late in the decade. Besides, though the graph I showed you last time has a couple of kinks in the 1970s, the relentless rise in driving basically shrugged off a comparable (in real terms) runup in oil prices during that decade.
Does the travel plateau have to do with government policy? For example, are our rising expenditures on public transportation getting people out from behind the wheel? The answer is no. Despite decades of rising spending on mass transit, transit travel has remained stubbornly flat. According to the American Public Transportation Association, in 2008 total transit ridership was actually below 2000 levels, despite growth in the population and economy.
Nor was government responsible for any big increases in America’s (disappointingly low, in my humble opinion) gas taxes during the period in question.
What about the land use policies favored by many urban planners, which are supposed to be reducing trip lengths by bringing origins (homes) and destinations (work, school, shopping) closer together? If they are having any effect, it is not showing up in the overall statistics. According to the National Household Travel Survey, the average vehicle trip rose from 9.0 to 10.1 miles between 1990 and 2009, and the average work trip rose from 11.0 miles to 13.7. If land use changes are at work, these numbers should be falling, not rising.
It is true that other countries Millard-Ball and Schipper studied – Australia, Japan, the U.K., Sweden, Australia and Canada – do have more anti-car policies, and considerably lower levels of driving, than we do. (The Japanese drive about 40 percent of the miles per person that we do, while at the other end of the spectrum Australians and Canadians drive about 67 percent as much.)
A lot (though probably not all) of the cross-national differences in vehicle miles traveled may indeed be attributable to government policies, including restrictions on parking, high gas taxes, stiffer registration fees, higher minimum driving ages, land use policies that promote density, more spending on transit, etc.
But fascinatingly, the slowdown in driving is evident in all of those countries, and it started at about the same time ours did. Since there was no coordinated push toward anti-car policies in the 2000s, it is hard to maintain that government is fundamentally responsible for what seems to be a trend across the developed world.
Is peak travel due to stagnant incomes? As my mentor, UCLA’s Brian Taylor, has pointed out, driving and economic health go hand in hand. To be sure, differences in national wealth probably explain a lot of the residual difference (not accounted for by policy) between driving in other developed nations and driving here. And right now we are, of course, finally waking up from a doozy of a recession, which is almost certainly having an effect on travel patterns.
However, the VMT plateau predates the financial crisis by years. As Millard-Ball and Schipper note, auto use flattened even as GDP grew in the late 1990s and early 2000s, both here and in other developed countries. Thus the link between rising wealth and rising VMT may have finally been broken.
Are there any more promising explanations? Next time I’ll outline some.