About 560,000 Americans use some form of car-sharing service — mostly Zipcar — where they can rent vehicles from neighborhood lots by the hour. That sounds like a lot of people, but it’s only about 0.27 percent of all drivers in the United States.

Not living up to its potential.

Here are the numbers: If 7.5 million Americans signed up for car-sharing services, the RAND study estimates that greenhouse-gas emissions from all U.S. vehicles would decline just 0.6 percent. If we got outlandish and assumed a future in which 20.3 million Americans (or about 12.5 percent of all eligible drivers) used car-sharing, then emissions from light vehicles would still just drop 1.7 percent.

That’s not nothing, but the drop in pollution seems fairly meager. How come?

To see why, it’s worth looking at how car-sharing reduces emissions. For one, fewer cars need to be manufactured in the first place — according to one survey, one shared vehicle displace nine to 13 private vehicles. Second, people drive fewer overall miles when using a service like Zipcar, in part because they’re constrained by the hourly rate. Third, car-sharing vehicles tend to be newer and more fuel-efficient.

But here’s the twist. For the most part, the people who sign up for car-sharing services were barely driving anyway. On average, Americans who use these sharing services see their car ownership numbers drop from 0.47 cars per household down to 0.24 cars per household. In other words, they went from barely owning cars to… barely owning cars. In contrast, car ownership for the country as a whole is about 1.87 vehicles per household. That’s one reason the effect on climate pollution is so small.

Indeed, as StreetsblogDC’s Tanya Snyder points out, public transportation — taking the bus or train — is a much more effective way for city dwellers to cut emissions. “One person taking transit to work instead of driving can save more than two metric tons of carbon dioxide emissions a year, according to APTA,” Snyder notes, “as opposed to the above estimate of 0.89 tons per person car-sharing.”

Now, the RAND study authors — Keith Crane, Liisa Ecola, Scott Hassell, and Shanthi Nataraj — point out that increased car-sharing could have all sorts of other benefits that aren’t counted in the numbers above. If millions more people use services like Zipcar, that means there’s less need for parking spaces. That frees up valuable land in urban areas. Car-sharing can also cut down on traffic jams and congestion, which can reduce travel times and conventional air pollution. It can also reduce transportation costs for many people — renting a car periodically is typically much cheaper than owning one.

“Because many transportation problems can be improved by small changes in the number of vehicles using the roads,” the authors note, “these benefits may be particularly valuable in large metropolitan areas.”

So what would need to happen for car-sharing to take off — growing to 7.5 million drivers or even 20.3 million drivers? The RAND study argues that these services need to become a lot more convenient — by, for instance, becoming widespread enough in cities such that anyone who needs a car at any time can get one. And there are a few policies that states and cities could do to help these services along, from doling out more parking spaces to car-sharing to tweaking state insurance policies so that peer-to-peer vehicle sharing becomes more viable. California is already experimenting with the latter.

Is this realistic? That’s unclear. The report cautions that optimistic projections about car-sharing have proved wildly off-base in the past. One report from 1994, for instance, found that the “market potential” for car-sharing in Germany was 2.45 million people. A decade later, only 70,000 people were using the service.

By the way, the full RAND report isn’t even primarily about car-sharing. It mostly discusses a new and comprehensive way to analyze the energy savings from various technological advances — there’s also a section, for instance, on the impacts of switching from newspapers to e-readers (they find, again, modest savings in greenhouse-gas emissions).