Jeral Poskey is CEO of Swyft Cities, which specializes in elevated cable-rail vehicles for urban mobility. He previously was a transportation project executive at Google's real estate division.
A significant amount of time, effort, capital, brainpower and, yes, energy is being spent developing solutions intended to cut transportation’s greenhouse gas emissions. Many of these efforts — such as self-driving robotaxis and electric vertical takeoff and landing air taxis — are laudable, incorporating cutting-edge technologies. But in our rush to address the myriad contributors to the climate crisis, we often fail to see the bigger picture of how these forces interact. Tackling one problem can lead to unforeseen consequences that inadvertently push the goal line further away.
While self-driving cars and robotaxis were once hailed as a solution to climate-harming congestion, they may actually make the climate situation worse. I believe they will have major unintended negative consequences with second- and third-order effects that will seriously undermine climate goals.
The transportation climate problem
Transportation is now the U.S.’ leading source of greenhouse gas emissions, with cars, SUVs, pickup trucks and minivans representing the largest share of that category. Carbon dioxide emissions from transportation have grown from 2021 to 2023 after a brief pandemic pause and are on track to increase.
Many other sectors of the U.S. economy have dramatically reduced their carbon dioxide emissions over the past two decades and continue to trend lower. Electric power generation, which was the top GHG source less than a decade ago, has cut emissions by more than 30% over the last 20 years.
A major reason transportation emissions are growing is that total vehicle miles traveled, or VMT, continues to grow and hit its highest recorded level last year. Every year, the tendency is for people to take more frequent and longer average trips, in part due to growing urban-suburban sprawl.
Robotaxis = more miles traveled
Robotaxis will accelerate VMT growth even further. A 2022 Urban Institute report warned that despite some potential benefits, such as improving road safety and access to mobility, robotaxis may “produce several potential downsides for the transportation system,” one of them being “significantly higher VMT.”
The researchers attribute this likelihood of VMT growth to several factors, including the convenience of robotaxis, the ability to conduct other tasks while riding and the reduced financial burdens of car ownership, fuel, insurance and maintenance.
The result is likely to be a combination of more frequent short trips, longer daily commutes and more “empty trips” — robotaxis repositioning or “orbiting” for their next rides.
In 2021, more than half of our daily trips were less than 3 miles, according to a Bureau of Transportation Statistics report. Prior data from the League of American Bicyclists showed that a staggering 72% of those short trips were done by driving in 2010. The convenience of robotaxis means people will take more frequent trips, and an even higher percentage of those trips will be done by car. And at least one study suggests robotaxis will shift many of those trips away from transit, which is often already struggling to maintain both ridership and funding post-pandemic.
Longer commutes
Self-driving robotaxis will also encourage longer commutes, further increasing VMT. Although commuting to work makes up only a small percentage of daily trips, those trips tend to be longer than those for daily errands. In 2023, the average American commuter spent nearly an hour traveling to and from work every day, covering nearly 30 miles.
Once relieved of driving responsibilities, people who carpool or use mass transit tend to engage in significantly longer commutes, according to U.S. Census Bureau data. Self-driving cars will encourage people who currently drive themselves to work to move farther out of the city with the lure of an easier, although longer, commute.
A study by MIT’s Intelligent Transportation Systems Laboratory found that nearly a third of U.S. drivers say they would consider moving farther out from the city if autonomous cars were available. Or as one technology reporter noted, the result of self-driving cars will be that people will “stretch their daily [commutes], once again allowing them to live where they want.”
More empty trips
More frequent short trips and longer commutes will significantly add to one of the biggest problems that’s already arising from car-based ride-hailing services: empty trips. Ride-hailing services such as Uber and Lyft currently have passengers onboard only for about half of their trips. Empty trips represent a tremendous waste of resources — a situation that will worsen as robotaxis increase trip frequency and average length.
All told, with more frequent trips, longer commutes and more empty trips, a University of California-Berkeley study calculated robotaxis could increase total VMT by a whopping 83%, nearly double today’s levels. A study by the Institute for Road and Transport Science in Germany concluded that as “people are willing to spend more time in their [autonomous vehicle] and therefore travel farther” there could be “substantial negative impacts resulting from more road traffic.”
More sprawl, high total carbon footprints
The MIT study concluded that the longer robotaxi commutes will also encourage more suburban sprawl – the opposite of what is needed to improve sustainability.
More VMT will necessitate more infrastructure to accommodate cars, such as roads and parking, which are highly carbon-intensive. Across materials, construction, use and maintenance, each mile of highway releases 3,500 tons of greenhouse gases over its lifetime, according to a University of British Columbia study.
Even without increased travel, the lower density of suburbs, exurbs and rural living also produces higher carbon footprints because of higher energy use. Urban neighborhoods tend to have more compact buildings that are often adjoining, as well as more multi-unit homes, which can result in lower energy use and transportation costs.
In addition, sprawl becomes a driver of more sprawl. As destinations grow farther apart, more infrastructure, such as roads and parking, is needed. This, in turn, pushes destinations even farther apart, creating a spiral of sprawl that becomes self-perpetuating. The end result may be ever-growing levels of vehicle usage and infrastructure.
Building better cities, not better cars
So if self-driving cars could lead to more vehicle miles traveled, more energy consumption, and more sprawl, what are better approaches?
It starts by thinking about how to build better cities, not better cars.
Currently, our cities are designed around cars as the primary transportation mode, often making walking, bicycling or transit less convenient or, in some locations, even unfeasible. A 2020 survey from Data for Progress found that 80% of Americans feel they have no choice but to use cars as much as they do. By default, we live in a car-dominated urban landscape.
Whether cars drive themselves or not does not change that, except it threatens to make it even worse. And that means moving us further away from our climate goals.
