Ever since transportation analysts started to notice that the leveling off of vehicle miles travelled (VMT) by U.S. drivers that started around 2005 had actually turned into a steady decline, any number of theories have been put forward to explain why Americans are driving less. The most hype has been around the decline in driving by teens and young adults – most of whom, the argument goes, have given up on their parents’ cul de sac suburbs to move to transit-served urban cores. I’ll certainly accept some of that is going on (I was one of those myself, if a bit ahead of the trend in the mid-1990s). But I don’t for a minute buy the popular explanation in the tech world – that “Millennials” (god I hate that term) are shunning car purchases and cruising for smart phones and social media. They aren’t getting licenses partly because states have made it harder and raised eligibility ages, and they aren’t buying cars because they have no jobs to go to, and student debt of epic proportions. If auto manufacturers want someone to blame for the drop in teen driving, they should blame universities, not Facebook.

Some more recent plausible explanations point out that aging Boomers drive less as they get older, and that 2005 was around the time that the U.S. economy entered a new era of higher and volatile gas prices. (The rise actually starts in 2002, but it keeps rising up until 2012 with a brief dip in 2009). Gasoline is now about three times as expensive as it was in the 1990s.

I’ve been thinking about this for a while, and have come up with three other hypotheses I’ve not seen discussed yet:

First, fertility – more young people are growing up in areas served by transit. Foreign-born women (who tend to be younger) have significantly more children than native-born women (see the Pew Research data). And immigrants are far more likely to live in urban and suburban areas served by transit, and to use it to get to work. (At least in California according to UCTC, and probably everywhere else). Therefore, young Americans today are more likely to grow up in a community where transit is a realistic mode choice. I’ve yet to do a thorough analysis of this, but my bet is that this would be a significant contribution to the VMT trendline.

Second, logistics. Despite the exploding volumes of e-commerce, a decade of steady improvements in route planning in the logistics sector is leading to significant reductions in wasted travel. UPS claims that technology solutions for improved routing have saved millions of miles since 2004. This is a drop in the bucket considering that Americans drove nearly 3 TRILLION miles each year over the last decade. But it brings me to my main point – which is that the machine intelligence that has allowed UPS to route its trucks more directly and efficiently has diffused throughout the entire population.

And so my third hypothesis, is that the mass adoption of personal GPS navigation – first through bundled manufacturer installed systems, then aftermarket personal navigation devices like Tom Tom, and now through smart phone apps like Google Maps, could be contributing significantly to the decline in VMT. Instead of taking habitual routes, more and more drivers are letting their GPS send them on a beeline that might include some unfamiliar twists and turns. Eventually people learn these better routes and start to incorporate them even into their own unassisted wayfinding.

After this occurred to me, I took a look at historic sales data for various segments of the GPS market, using data from Statista, a statistics reference service available through the university library. The numbers are pretty stunning – reporting doesn’t even begin until 2005, because the market was quite small before then. But between 2005 and 2010, there were nearly 80.5 million “GPS equipment units sold in the automobile segment” in the United States, according to Statista (I’m still trying to verify exactly what this covers – but I’m pretty sure it is manufacturer-installed in-dash units only). The number of devices sold annually increased ten-fold from 2005 to 2010.

Is it a coincidence that the era of falling VMT and the mass diffusion of personal navigation services coincide almost precisely? It’s too early to tell.

But if this is true, its good news for both the US and the world. According to Garmin, only 25 percent of the world’s 700 million cars currently have GPS installed. As this technology continues to spread throughout the market, we could see additional gains in travel efficiency.

(This essay is part of the Rudin Center’s ongoing research project Re-Programming Mobility made possible by a grant from the Rockefeller Foundation.)