By Eric Sundquist
In the second half of the 20th century, as the nation rapidly suburbanized, it was pretty easy to predict how much driving Americans would do. United States vehicle-miles traveled (VMT) rose steadily year-by-year, with only temporary blips around fuel shortages or recessions.
In the 21st century, the pattern has been much harder to discern. The growth of VMT first slowed, then actually went into reverse during the recession. After the recovery took hold, VMT growth spiked to 20th century levels.
Causes for the recent short-term gyrations are many. Gas prices likely played a role, but over the long term the correlation between gas prices and per capita VMT is a fairly weak .31. VMT is more strongly related to the economy; the correlation between real per capita GDP and per capita VMT is .64.1 The big VMT drop in 2009 reflects this relationship. People can’t commute if they are out of work, and they won’t make trips for shopping, vacations, and other discretionary purposes if household budgets are tight. It may be that it took until 2015-17 for many people to regain their financial footing enough to make discretionary trips at pre-recession levels, so the spike could be seen as a kind of rebound.
More interesting than the year-to-year swings is the longer-term trend. Normalizing for population, VMT per capita has seen zero net growth since 2002. Leaving in the population-growth effects, the trend in overall VMT is heading toward zero growth as well (trend line, Figure 3). A variety of factors are at work.
Modeling, however, has had a hard time with the new slower-growth reality. FHWA’s “Conditions and Performance” series uses VMT estimates to predict congestion and other outcomes, and these estimates have been consistently—and often extremely—high in this century. The most recent C&P ratchets down the growth estimate to 1.04 percent annually. Yet the most recent year of VMT growth, through mid-2018, was about half that, and the trend line is pointing lower still.
The sense of where VMT is going matters, because overestimates of driving demand tend to exaggerate the need for highway-capacity spending, potentially leading to waste and diversion of funding for critical operations and maintenance needs, as well as for non-auto modal facilities.