Does carsharing really reduce overall driving?

The research shows carsharing really is green.

At the Share conference that I covered for this week’s Guardian, there were wildly divergent claims for how many vehicles carsharing companies such as City Car Share, Zipcar, and Getaround take off the road. I was also a little skeptical of claims that carsharing dramatically reduces overall driving and greenhouse gas emissions, so I decided to take a deeper look at the issue.

“For every car that is shared, we take seven cars off the road,” Board of Supervisors President David Chiu said during his presentation. The next day Getaround founder and spokesperson Jessica Scorpio cited a study claiming that 32 cars get taken off the road for every shared vehicle.

Luckily, one of the country’s top researchers in this area is right in our backyard. UC Berkeley civil and environmental engineering professor Susan Shaheen heads the school’s Transportation Sustainability Research Center and has been doing peer-reviewed studies on car-sharing for almost 20 years.

Her research, which is consistent with the body of academic research on carsharing from around the world, has found that each shared car takes between nine and 13 other cars off the road, figures that she says are amazingly consistent around the world. That big reduction is because households that have cars tend to get rid of at least one of them when they sign up for carsharing, while car-free households that want access to a car will choose (as Shaheen says is the case for about 25 percent of the people in each group, which adds up to 90,000-130,000 fewer cars on the road nationwide).

But Shaheen told us that she’s highly skeptical of the study showing 32 cars being taken off the road, an industry-sponsored study that was far less rigorous than her work, and whose authors intially wouldn’t share their work with her when she requested it. “It appears to be some kind of idealized number,” Shaheen told me. “I take this stuff pretty seriously.

The part about up to 13 cars being taken off the road has always made sense to me, and it’s certainly a benefit to cities have fewer overall cars on the road, particularly as they use parking spaces on public roads. But I was more skeptical of the claim that overall vehicle miles traveled (VMT) is reduced by carsharing.

Shaheen’s biggest study (“Impact of Carsharing on Vehicle Holdings,” a 2009 study she did with Elliot Martin and Jeffrey Lidicker) found overall VMT was reduced by 27 percent, even though her study frankly admits that about 60 percent of carsharing customers come from car-free households.

How can that be? Doesn’t it seem like driving would increase when you give more people access to something as expensive as an automobile, particularly in a small, congested city like San Francisco that has such a variety of other good transportation options? Shaheen does admit that one study focused specifically on San Francisco did indeed find that overall VMT does increase among carsharing customers in their first year using the service.

But the studies by Shaheen and other researchers show that VMT among carsharing customers drops in subsequent years, often quite dramatically, as people figure out that it’s not really so hard to get around by public transit, bike, or on foot. “When people use carsharing, they use it less and less and less,” Shaheen told us.

And when it comes to air pollution and greenhouse gas emissions, Shaheen said her studies and those of other researchers — including the 2010 “Greenhouse Gas Emission Impacts of Carsharing in North America,” who she and Martin did for the Mineta Transportation Institute at San Jose State University — found dramatic reductions.

“This study contributes to mounting evidence that carsharing is lowering GHG emissions

by providing people with automotive access on an as-needed basis,” the study concludes, with the reduction in VMT compounded by the fact that carsharing encourages people to get rid of older, more polluting cars and instead drive new, more efficient vehicles.

Still, despite all of that evidence, Shaheen admits that it’s tough to say carsharing causes less overall driving in the city of San Francisco, although she thinks it’s likely that it does. Not only is her survey based on self-reported data, but it’s also based on national aggregations that can’t be unpacked to comment on specific locales.

And her study predated rapid and drastic changes on the streets of San Francisco over the last five years, where bike lanes and ridership have been expanded, bikesharing has been introduced (another area that Shaheen has studied in detail), Muni has had its ups and downs, the population has increased, parking costs have increased and morphed, and Lyft and other ridesharing companies have filled the streets with a new category of cars (that latter development is the subject of her latest research project that is now underway).

“Things are changing really rapidly,” she told us.

Each of those variables affect whether someone chooses to drive a car, and they can be constantly shifting. For example, driving might become a far less attractive option as automobile congestion increases, particularly if the city expands the capacity for cyclists and Muni riders.

Shaheen is also looking forward to technological advances in data collection, allowing her to use less self-reported data, something she is already able to do with Bay Area BikeShare: “With bike sharing, they are so transparent. They share all the data, it’s like a party for me.”

So as carsharing and ridesharing companies seeks to bolster their claims to being an unqualified benefit for society, those companies should be willing to share detailed usage data with Shaheen and other independent researchers, which will help us all move beyond the hype to make informed public policy decisions.

“This is important,” Shaheen told us. “I take this very seriously because I want these to be numbers that policymakers can use.”