How Uber and Lyft Are Trying to Solve America’s Carpooling Problem

First, the bad news: carpooling has been on the decline in America for nearly four decades. That practice could be helping the environment and America’s commuters, who are needlessly stuck for hours each day on packed highways. Multiple people sharing a single ride to a common destination is a simple act that has the potential to reduce CO2 emissions, ease traffic, lessen fossil fuel dependency, reduce stress on commuters, and even drive down rents in dense cities. Yet the practice fell out of favor after reaching a peak in the 1970s.

Now, the good news: popular tech companies Lyft and Uber are leading a wave of new services that have the potential to revive shared rides. “What fascinates me about these things is: can they move us closer toward a vision of an integrated public transit system?” asks Susan Shaheen, co-director of the Transportation Sustainability Research Center at the University of California, Berkeley. “And can it move us closer to filling empty seats in vehicles?”

Despite referring to themselves as “ride-sharing” companies, Lyft and Uber have largely been in the business of what transportation experts call “ride-sourcing,” because they essentially provide the same service as taxis through their own platforms. “I’ve studied ride-sharing for a long time, and the definition of ride-sharing is really carpooling,” Shaheen says. “And a carpool is an incidental trip.” That is, it’s a trip that a driver was going to take regardless of whether anyone else was with them in that car.

The distinction isn’t just academic. When the thousands of drivers working for Uber and Lyft in San Francisco are picking up a single fare and taking them from Point A to Point B, it’s probable that they’re adding to unnecessary congestion, pollution and fuel consumption. But last summer, within hours of each other, the companies announced that they were rolling out UberPOOL and Lyft Line in San Francisco, passenger-pooling options that would give riders cheaper fares if they’d be willing to share their vehicle with strangers traveling a similar route.

The companies say customer interest has been high so far. Each company has since expanded the service to Austin, Los Angeles, and New York City, and Uber has launched POOL in Paris. Lyft says that 50% of rides in San Francisco are Lyft Line rides, and a little more than 20% of all Lyft rides in the city start or end within a quarter mile of commuter rail stops. “That’s notable,” says Shaheen. “It means people are taking this short trip in one of these vehicles and connecting it to a longer line-haul transit trip. It’s basically enabling somebody to not take a single-occupant vehicle for this long commute trip and to rethink how they commute.”

Uber crunched the numbers on their “matched trips” for one month in San Francisco, comparing them to the number of miles that vehicles would have traveled if all those rides had been taken individually. They estimated that UberPOOL rides taken between February and March amounted to 674,000 miles of saved driving. That’s the equivalent of 240 people driving round trip from L.A. to New York. “UberPOOL is really about trying to reinvent cities from a transportation perspective,” says product manager Brian Tolkin. “Part of that means making Uber so affordable that it’s really available to anyone and a better alternative to, say, owning a car.”

Of course, before these companies start patting themselves on the back for saving the environment, they have to offset the number of cars they’ve brought onto the road. They aren’t releasing data about that, and there is other crucial information missing. The most important piece, Shaheen says, is knowing what the people using these services were doing beforehand. If someone is now using a combination of Lyft Line and public rail rather that driving alone in a car from San Francisco to Cupertino, that represents a greater environmental offset than if that person was previously taking public rail and a public bus.

Carpooling took off in America during World War II, when the government asked people to start sharing rides to work so they could conserve rubber for the war effort. The practice gained popularity through the 1970s, spurred by volatile energy prices, employer-sponsored programs and the advent of HOV lanes. But as gas prices dropped, cars got cheaper and more people and companies decamped for far-out suburbs and exurbs, more workers began taking their own cars to the office. Carpooling became associated with the inconveniences of neighbors’ inflexible schedules, awkward reimbursements and a lack of privacy. Nearly one in four people shared a ride to work in the 1970s. By the time census workers asked about that practice in 2010, the number had dropped to about 10%.

It’s too early to tell if Lyft and Uber’s early efforts will reverse that trend. But it is clear that they are benefiting from a changed landscape. Smartphone ownership has exploded, allowing people to connect and share useful information about where they are. Familiarity with social networks can encourage strangers to trust each other. The algorithms matching riders and drivers—while keeping routes convenient—are constantly improving. And though car sales have continued to climb in recent years, younger urban residents say they’re less interested in driving and owning their own vehicle.

Perhaps the most promising trend line for these services is that Uber and Lyft are finally solving the problem that has derailed past attempts to solve America’s carpooling problem with technology. “When you have a new system with a really small number of people in it—which any system will when it’s new—there’s a very, very low probability that you’ll have a match between all the potential origins and destinations of a driver and a passenger,” says Emily Castor, Lyft’s director of transportation policy. “So those systems that had tried to do that have been pretty uniformly unsuccessful, because they have a high failure rate.”

That’s what happened to Zimride, an early incarnation of Lyft. Among the key lessons for Zimride’s founders when they rebranded as Lyft: always have drivers available, lest you deter potential customers. “We’ve been able to build up a network that has enough density that it actually is getting to the point now where we do have a ton of people using it,” says Castor. “So we’ve kind of overcome that chicken and egg problem and we now can start doing really interesting things.”

Those experiments include “Driver Destination,” which allows a driver to specify where they’re headed and signal that they’re available to pick someone up. The app will only link the driver to a passenger going the same way. This type of trip can help eliminate wasted space in cars and potentially keep superfluous cars off the road–an efficiency that experts like Shaheen call the holy grail. “The next phase for Lyft is to look at how we can increase that commuter carpooling activity and to expand on our vision to make it so any time any driver is on the road, [they] can be using the empty seats in their cars to give rides to other people,” Castor says.

The key to long-term success may be money. For these services to truly take hold, drivers will need to see the upside of bringing a few strangers along for the ride. Old-fashioned carpooling was set up for passengers to reimburse a driver for just gas and wear and tear, a piddling bit of change per mile. “That’s just not enough to make people notice and think about doing that,” Castor says. “But if you could earn $15 on your way to work and your way home, that would probably raise your eyebrows.”