Tech. Science. Culture.
We may earn a commission from links on this page

How Self-Driving Cars Could Make Traffic Worse

We may earn a commission from links on this page.

Self-driving cars bring many promises like fewer deaths, smarter navigation, and no more ugly parking lots. But a new study highlights a critical disclaimer: Unless these vehicles are shared, we’ll probably see a dramatic increase of the number of cars on our streets.

The study published in Transportation Research Part A today by the University of Leeds, University of Washington, and Oak Ridge National Laboratory, looks at the impact of autonomous vehicles on energy demands in 2050. And it points to a much more complicated future.

When it comes to driverless cars, there’s plenty of potential for energy savings: improved traffic flow, no circling for parking spaces, the ability to drive very close together to reduce drag (platooning), and manufacturing lighter vehicles due to lower crash risks.


But there’s also a less obvious behavioral factor that might counteract all those energy savings. Without the act of driving (or being stuck in traffic) as part of the equation, using a car suddenly becomes a lot more attractive.

“There is no doubt that vehicle automation offers several efficiency benefits, but if you can work, relax, and even hold a meeting in your car, that changes how you use it,”said lead author Zia Wadud, associate professor at University of Leeds’ Faculty of Engineering and researcher for the Institute for Transport Studies.


Here are some examples. If a 45-minute commute is no longer stressful due to the ability to work or relax in the car, then people might live farther from their jobs, increasing their driving time and contributing to sprawl. Passengers who might have used high-speed rail for a longer trip would opt for their own self-driving vehicle for privacy and convenience. Certain demographic groups that would not have been able to drive human-piloted cars—the elderly, the disabled—might suddenly buy a ton of driverless ones. In the end, this wondrous innovation that spells the end of driving might end up encouraging a more car-centric lifestyle!


How to avoid this dystopian future? “A widely used, on-demand ride sourcing system with appropriately sized vehicles is key to realizing autonomy and freedom without everyone having a car in their garage,” says Don MacKenzie, assistant professor for the Department of Civil & Environmental Engineering at the University of Washington, and one of the study’s co-authors. “The size of the vehicle can be matched to the needs of each specific trip, so there’s no more energy wasted on moving around a five- to eight-passenger vehicle just to transport one person.”


Sounds great, but how to we possibly get from here to that robo-taxi future without everyone buying their very own autonomous car first? MacKenzie points out that Google’s working on cracking one part of the puzzle: The most important part of the robo-taxi reality is having fully autonomous vehicles, with no human drivers, ever.

The other factor is market penetration. “A robo-taxi system performs best when everyone is using it: more demand means more supply, which means shorter waiting times and faster service,” says MacKenzie. “If everyone is doing this, then there’s no need to own your own vehicle—at least if you live in urban or suburban areas with adequate density, as most of the US population does.”


In other words, that autonomous Lyft partnership with GM needs to arrive faster than the autonomous Tesla.

[University of Leeds Institute for Transport Studies]


Contact the author at or follow her at @awalkerinLA