Each shared Car2go vehicle removed between 2 and 11 personal cars from the road, depending on the location

LAURA BLISS@mslaurabliss

I sold my Prius when I moved to Washington, D.C.—I expected the city’s Metro system to be a reliable alternative. (It isn’t, but that’s another story.) Luckily, my feet, my bike, and several ride-hailing and car-sharing apps such as Zipcar and Turo have gotten me where I’ve needed to go.

These so-called “shared-mobility” alternatives have exploded over the past decade, and some of them are making a dent in urban car ownership. In a report published Tuesday, new research by UC Berkeley’s Transportation Sustainability Research Center quantifies the effects of car2go—a one-way car-sharing service powered by German automaker Daimler AG—in select North American cities. The report, co-sponsored by the U.S. Department of Transportation and car2go, among other funders, found that the service creates a net reduction on the number of vehicles on the road, which in turn seems to be reducing the number of vehicle-miles traveled and greenhouse-gas emissions.

Earlier work by the TSRC found that round-trip car-sharing services such as Zipcar also lower urban vehicle ownership.

Car2go boasts a somewhat different model: The company operates a fleet of diminutive, two-seater Smart Fortwo cars available for one-way rentals. Members are charged by the minute and can begin and end trips at different locations within a certain zone. These point-to-point services have shown significant growth in recent years and are “quickly gaining ground on ‘traditional’ car-sharing” models, according to the Shared Use Mobility Center. By focusing on one such company, the Berkeley researchers were able to include fine-grained statistics on how members use car2go vehicles, says Susan Shaheen, the co-director of the TSRC and the lead researcher of the new report. That allowed for more nuanced conclusions about what these services can do for cities.

How car-sharing changes car owners

Shaheen and co-author Elliot Martin surveyed close to 10,000 car2go members in Calgary, San Diego, Seattle, Vancouver, and Washington, D.C. between 2014 and 2015. (These five cities were chosen for their diversity in population size and density.) The key questions assessed the respondents’ overall driving habits, whether or not they had gotten rid of a vehicle as a result of car2go,whether they’d put off or decided against buying a vehicle, and how their driving habits might change if car2go suddenly disappeared from their city. Crucially, Shaheen and Martin were also able to compare the survey responses—imperfect as they were, given the fallibility of human memory—to car2go activity data provided by the company.

For the vast majority of respondents, car2go did not have any causal effect on their vehicle holdings. Just two to five percent of active members said that they’d sold a car as a result of car2go’s operations. Between 7 and 10 percent said they’d avoided buying a car. But those relatively small percentages translated to sizable impacts. For every car2go vehicle on the street, the researchers found, members sold somewhere between one and three personal vehicles and avoided buying between four and nine vehicles. Overall, each shared car2go vehicle removed as many as 11 personal cars from the road. (City-specific detail on this can be found in the table below.) Across the five study cities, that added up to 28,000 fewer cars.

(Transportation Sustainability Research Center)

By encouraging members to shed or avoid the purchase of personal cars, car2go is also having an effect on the number of vehicle miles traveled in these cities. Shaheen and Martin tried to estimate the number of miles driven by vehicles that were never acquired by using the number of miles that car2go members reported traveling in private vehicles they owned at the time of the survey. Analyzing these responses conservatively, they found car2go may be suppressing between 6 and 12 million miles driven in personal vehicles. Under a more liberal model, that mileage may be as much as 21 to 37 million miles. “We hypothesize that the number is somewhere in the middle,” says Shaheen. The savings added up to 5,300 to 10,000 metric tons of greenhouse gas emissions across the five cities—about 10 to 14 metric tons per year, per car2go vehicle. (How much is a ton of gas? Picture a round balloon with a 10-meter diameter.)

The future of car-sharing

On the other hand, the study also found that car2go poached some public transit ridership. A small minority of respondents reported that the service had changed their transit use, and most said that they used it less. But this really varied at the individual level: “Some members were increasing their use of buses, but not rail, while others were increasing both or decreasing both,” says Shaheen. For many people, she says, the choice to spring for car2go instead of the bus was often based on bad weather or on a one-off, time-sensitive need. And, based on her previous research, Shaheen believes that services like car2go could have a role to play in lower-density areas outside urban cores to act as a way to connect more people to transit.

Now it might be up to policy-makers to figure out where and how to position car-sharing services to make them complement buses and trains.

Car2go CEO Paul DeLong says that Shaheen’s research validates the company’s mission to “improve the quality of life in the communities we serve.” He adds that, from the perspective of car2go’s car-manufacturing parent company, Daimler, car2go’s services are also a way of exposing potential future car-buyers to Daimler’s automotive products. “Our business fits well in urban centers, and as people move out to suburbs and their family life changes, they might make a purchase or a lease on a car that they’re familiar with,” he says.

Given Smart cars’ somewhat dismal U.S. sales records, it seems unlikely that legions of former car2go members will one day clamor to put Fortwos of their own in their garages. (They might upgrade to one of Daimler’s other automotive brands, Mercedes-Benz.) Even so, DeLong’s point hits on somewhat of a tension in car2go’s business model, and that of a number of other car-sharing companies operated by auto manufacturers. If the ultimate goal remains to sell cars to members as soon as they’ve got a driveway, then there may be a limit to how much these services—and their benefits—can grow within a particular city.