As ride-hailing services grow in popularity and transit use declines in many cities, a new study reports Uber and Lyft are luring bus and light rail riders from the public transportation system. The study collected data from 7 major U.S. cities, including Seattle.

Previous studies suggested shared mobility services were complementing public transit, but researchers at the U.C. Davis Institute of Transportation Studies found ride-hailing services have reduced the use of bus services by 6 percent and light rail by 3 percent. The data did show ride-hailing serving as a complementary mode of transportation with commuter rail services, with ridership increasing 3 percent. High-quality, long-haul transit actually benefits from improved access while short-haul, low-quality transit has a lot to lose.

2016 U.C. Berkeley study found that of Seattle’s 54,000 Car2Go members at the time, about 3 percent sold their car and a further 9 percent abandoned plans to purchase another. Although most Car2Go users increased their Vehicle Miles Traveled (VMT) by moving trips from transit to Car2Go, this gain in VMT was more than offset by people that sold their vehicles and therefore paid for car use by the mile.

However, there are important differences in ride-hailing and car-sharing; cost, ease of access, and likelihood of traveling in a group.

“Although we found that ride-hailing can be complementary to transit and reduce vehicle ownership for a small portion of individuals, we found that (overall) these services currently facilitate a shift away from more sustainable modes towards low occupancy vehicles in major cities,” wrote the lead author of the report, Regina Clewlow, in a statement.

“It’s a sign that those transit systems aren’t meeting the needs of the traveling public,” said Andrew Glass Hastings, transit and mobility director at SDOT.

However, with increasing transit ridership in Seattle, that conclusion doesn’t apply here, said Hastings.  Instead, he argued that “we can’t meet demand fast enough, and ride-hailing services are filling that need.”

With Seattle’s rising transit ridership, Hastings said other cities are beginning to follow the city’s lead reconfiguring their transit and street system, in part by prioritizing lanes for transit. And SDOT continues to make investments in HOV lanes, Hastings said, pointing to the seven RapidRide corridors planned — two-thirds of the Madison Street BRT line will be a dedicated transit lane.

The U.C. Davis study was based on a survey of 4,000 residents, suburban and urban, living in or near seven metro areas, including Seattle, New York, Chicago and Los Angeles.

Slow service, unreliability and lack of frequency topped respondents lists as reasons for using ride-hailing instead of transit services. And parking constraints and alcohol use were top reasons ride-hailing users gave for using the service rather than driving themselves.

When respondents were asked if Uber and Lyft were unavailable, thirty-nine percent said that their trip would have still been made with a car, either by driving themselves, using a carpool or taking a taxi. 61 percent of people surveyed said they wouldn’t have taken the trip or instead would have walked, biked or taken transit.

The authors said this suggests “that ride-hailing is likely adding vehicle miles traveled to transportation systems in major cities.”

“In addition, depending on the volume of deadheading miles associated with ride-hailing trips (miles traveled without a passenger, which have previously been estimated to be 20 to 50 percent), the vehicle miles traveled associated with a ride-hailing trip is potentially higher than a trip taken in a personal vehicle,” the authors wrote.

The researchers compared car ownership between users of ride-hailing, transit and riders who use a combination of both forms of transportation. They found vehicle ownership was higher among people who used both ride-hailing services and transit (52 percent) rather than riders who only used transit (46 percent). Among non-transit users, people who used ride-hailing were as likely to own a vehicle as people who used neither ride-hailing services nor transit.

Overall, the vast majority of respondents (91 percent) said the availability of ride-hailing services did not influence their car ownership, while nine percent had sold at least one vehicle.

The authors say congestion pricing and enforcing priority lanes for high-occupancy vehicles are tools cities can use to reduce the number of vehicle miles traveled that ride-hailing services are adding to roads.

One option in SDOT’s mobility playbook to ensure ride-hailing services continue to complement the existing transit landscape in Seattle, rather than siphoning off passengers back into cars, includes dynamically managing high-occupancy-vehicle-only lanes in dense areas throughout the city. Similar to express toll lanes, dynamically managing HOV lanes would allow usage to change based on demand or time of day.

The playbook also states the city needs the “right mix of incentives and disincentives to keep ride-hailing, even autonomous vehicles from cruising around empty while they wait for customers.”

“A lot of that can be done through pricing,” Hastings said. For instance, such a system could charge high prices for trips without a passenger or use congestion pricing to make travel during busy times more expensive, he added.

Riders can expect more city HOV lanes in the future, Hastings said, pointing to the seven rapidride corridors planned — two-thirds of the Madison Street BRT line will be a dedicated transit lane.

9 Replies to “Are Ride-Hailing Services Helping or Hurting Transit?”

  1. The U.C. Davis study was based on a survey of 4,000 residents

    I find it hard to put much value on a 3% number generated from a survey of only 4,000 smeared across 7 metro areas. Logically commuter rail could see an increase. Especially in the Seattle area where, assuming the ride sharing is available at the origin of the trip, station parking is a limiting factor. I’d put more credence in real numbers generated from before/after rideshare became available or trends in rideshare company vs transit agency data. With our transfer system I could see bus ridership (i.e. number of users) increase while number of trips decline. Like the commuter rail effect people might take a bus into the city if it didn’t take a god’s age to transfer and go the last 2-5 miles.

  2. While this study looks at changes in user behavior, there are other factors that can affect decisions about choosing to use transit or not.
    Is there data on the increase in congestion and traffic attributed to rideshare services, both in terms of numbers of additional vehicles on the road as rideshare vehicles (with and without passengers)and driver behavior (slow driving or trolling for passengers around popular areas, stopping traffic to either pick up or drop off passengers). The additional VMT of SOVs functioning as rideshare vehicles impacts overall congestion which has a direct impact on bus schedule reliability.

  3. It’s important to remember that the survey simply asks if you use public transport more or less, but doesn’t ask how much more or less.

    I took the survey and answered “less”, since many of my Uber/Lyft trips have indeed replaced trips I’ve made by bus in the past. But it’s still only a small percentage of my bus trips, and nearly always during off peak hours, so the actual impact on road congestion is minimal.

    Basically, Uber and Lyft are affordable enough to use for those oddball trips that happen once a month, where a 15 minute drive equals an hour+ bus ride. But still way too expensive for a daily commute.

    1. Another point, if you’re paying high rates by the minute, you may avoid using it at times and places where you might get into an hour-long traffic jam, such as on a freeway during rush hour.

      1. That’s Car2Go, where $0.50/minute sitting in traffic adds up really fast. Uber/Lyft, not so much, especially since they started quoting you fares in advance without knowing exactly how bad the traffic is going to be.

        In general, I decide between transit and ride hailing by estimating the money spent per time saved, and start to consider ride hailing or car sharing when the cost per minute saved drops below around $0.50 (including, of course, wait time for both transit and ride hailing in the calculations).

        Off-peak, I take trips that meet this standard at least a couple times a month, but during rush hour, almost never. Why? Well, multiple reasons. Rush hour is when transit is running at its maximum frequency, and transit priority, when it exists, has the maximum effect. It’s also when wait times for ride hailing are most unpredictable, since you have no idea what kind of traffic the driver is going to have to go through to get from wherever he answers the call to the pickup point.

        Even outside of rush hour, the time/money calculations can start to favor transit when the bus is coming very soon (which, of course, happens more often when the route is frequent!), the distance is relatively long (and ride-hailing relatively expensive), and transit priority (e.g. bus lanes, Link tunnels) exists to speed up the transit option. During bad traffic, even going somewhat out of the way to get to a Link station can be time-competitive with driving, provided that the surface part of the trip can be done on foot or bike.

        Even if a trip is just a lowly bus, without any priority, bad traffic can sometimes tip things towards the bus, particularly if it’s a frequent bus that OneBusAway says is arriving very soon. If the Uber driver has to spend 5-10 minutes trying to turn around and go 5 blocks to the pickup point, while the bus is right there, that head start can make the bus end up every bit as faster as the Uber.

        And, of course, with UberPool, all bets are completely off, as you have no idea what extra traffic or long stoplights are going to be involved getting to/from other passengers’ pick up points until you’ve committed. The bus, at least takes a fixed route, with a published schedule, so you can have some idea how long it’s actually going to take.

  4. There is a time-of-day factor that is not discussed here. Public transit is much less frequent after 8 or 9 pm, and choosing to ride transit and make a transfer can turn a 15 minute Uber ride into a 60- minute or longer ordeal!

    Finally, late night trips rarely cause systemic congestion as there are only a few place that have congestion after 8 pm (cough — SeaTac Airport), so that linkage to congestion pricing may be almost irrelevant.

    I don’t think the study authors have thought this issue through well.

    1. This is so true. I’ve pretty much stopped riding the bus in Seattle in the evenings. Lyft line is pretty cheap and just far too convenient for me to waste time on Seattle’s bus system.

  5. On a somewhat similar note, there’s an interesting story on the PI website today: “Horror stories of Seattle’s ride share, cab companies”

    Most customers slam Lyft, but this one goes after Uber:

    “I take Ubers to/from the Seattle airport many times per month. Your drivers picking up from SEATAC are becoming increasing unprofessional and unsafe. I routinely have drivers cancel my ride requests (presumably because I’m not going far enough) and when I do get a driver, they’re often rude and drive like maniacs. My driver tonight blasted crappy music, refused to turn on the AC, and swerved across lanes multiple times on two occasions. If I wanted that experience I’d hire a taxi or just hitchhike.

Comments are closed.