This article is from the archive of our partner .

It is unquestionably true that Americans are driving less today than we did just a few years ago. Sometime around 2004, our addiction to driving – expressed on a graph in the decades-long steep expansion of “vehicle miles traveled” – took a turn in the opposite direction. Per capita, we began to drive fewer miles each year than we had the year before. As the U.S. population has continued to grow, our collective miles traveled by car has begun to stagnate.

"A New Direction: Our Changing Relationship with Driving and the Implications for America’s Future" by Dutzik, Baxandall

It’s not entirely clear, though, exactly why this has happened or whether the downturn will continue, two questions intimately tied to the behavior of Millennials as they age. Twenty-something Americans drive about 20 percent less today than their parents did in their 20s. But is that because of the recession? High gas prices? A lasting shift in consumer demand? What will happen to today’s 20-year-olds as they enter their 30s, raise families, and consider moving to the suburbs?

This question lurks behind every celebrated trend story about the Millennial generation: Will this group really push systemic change (as the baby boomers did) in how Americans live, work and relate to each other (sharing cars, for instance, as opposed to owning them)? Or is this moment – with its associated driving patterns – a hiccup in history?

"We've basically assumed in transportation planning for decades upon decades that the amount of vehicle travel and per capita VMT can go in only one direction, and that's up," says Tony Dutzik, a senior policy analyst for the Frontier Group, a public interest think tank. "And we have been planning our transportation system based on that assumption."

Data from the last few years clearly show that this axiom is no longer true. So what happens next? In an effort to at least sketch out some of the possible scenarios, the Frontier Group and the US PIRG Education Fund today released a report outlining three alternative futures for America's relationship to the car.

One assumes that Millennials will eventually revert to the driving patterns of their parents (the blue "Back to the Future" scenario on the below graph). The second assumes that America is in the midst of an enduring shift toward less driving, brought about in large part by the permanent new preferences of Millennials. And the last scenario assumes that the recent decline we've seen in driving will continue apace.

Notably, all three scenarios project car usage well below the status quo we might expect based on the last 50 years of driving trends (even with a growing population).

"Thus far in the transportation community, the question of declining vehicle miles traveled and declining gas tax revenues has been thought of as a revenue problem. There’s not enough money coming in to deal with the needs that we think we have," Dutzik says. "There's another important part of that question that we really haven’t addressed: What is it we think we're actually going to need for the future?"

What if we don’t need to find funding to for quite so many highway expansions?

"If the Millennial trend continues to play out," Dutzik says, "the amount of highway capacity we’re going to need in the future looks far different than it looked in government projections from just a few years ago."

In fact, here are some of those government projections:

Dutzik and report co-author Phineas Baxandall constructed these scenarios by trying to "layer on the unknowable over the knowable." Namely, we don’t how Millennials will behave in the future. But we do know that their parents, from the baby boom generation, will soon begin aging out of the workforce in massive numbers. And when you no longer have to drive to work, that cuts a sizable chunk out of your car consumption (the 2009 National Household Travel Survey estimated that commuting trips accounted for 27.7 percent of household VMT).

With these underlying demographic trends in mind, the "Back to the Future" scenario makes the assumption that Millennials will drive about as much at age 35 as their parents did at that same age, and so on as they get older.

The other two scenarios are built on something of a mystery. Researchers have not yet been able to disaggregate how much of our current decline in driving has been attributable to gas prices, or the economy, or changing attitudes toward car ownership or urban living. But it’s been driven by something. And in these two futures, Dutzik says, “whatever constellation of things it is that has caused the shift in per capita driving over the last decade – we think that’s a real thing.”

The way we work has also been changing, alongside the demographic shape of the workforce itself. And this trend, which the report did not address, could drive down VMT even further, as more people telecommute or join the freelance economy.

In all of these scenarios, Dutzik assumes that we're unlikely to ever surpass our 2004 peak in per capita driving. Today, we individually drive about 7 percent less than we did in 2004 (or at levels comparable to 1996). It’s plausible we could return to that peak, but many of the factors that drove us there no longer exist: cheap gas, the growing baby boom workforce, women entering the workforce for the first time.

Millennials will inevitably wind up driving more than they do today as they age. This is virtually always true of people in their 20s as they enter their 30s and beyond. Certain stages of life demand more use of a car than others. But the question is: by how much? And by how much compared to their parents?

This article is from the archive of our partner The Wire.

We want to hear what you think about this article. Submit a letter to the editor or write to