Uber and Lyft admit they’re making traffic worse

There have been a couple of really interesting articles recently on the impact of ride-hailing companies like Uber and Lyft have.

This CityLab article confirms some long-held suspicions that ‘rides-hailing’ companies like Uber and Lyft are ‘part of the problem’ and helping make congestion worse. The do this as they’re now making up a significant proportion of total vehicle travel in many of the cities they operate in. Even more worryingly, it seems that these trips make up a larger and larger proportion of traffic the closer to the city’s centre you get. Note: TNC stands for Transportation Network Company, shorthand for companies like Uber and Lyft.

…the industry is ‘fessing up. Today the ride-hailing giants released a joint analysis showing that their vehicles are responsible for significant portions of VMT in six major urban centers. Still, Uber and Lyft’s combined share is still vastly outstripped by personal vehicles. As Chris Pangilinan, Uber’s head of global policy for public transportation, wrote in a blog post accompanying the findings, “although TNCs are likely contributing to an increase in congestion, its scale is dwarfed by that of private cars and commercial traffic.”

Led by the respected transportation consultancy Fehr & Peers, the analysis provides a high-level view of the combined mileage contributions from Uber and Lyft, as a share of overall VMT, over a recent month in the Boston, Chicago, L.A., San Francisco, Seattle, and Washington, D.C. areas. Results are shown at the level of the larger metropolitan landscape, which includes both the central city and its surrounding suburbs, as well as the level of the core county that contains the city’s most concentrated homes and jobs.

Here are some of the key findings:

So across the wider urban areas TNCs account for 1-3% of total vehicle miles travelled but much more in the central core areas. What’s more, this is compared to total travel and that won’t be evenly spread across the day. There doesn’t seem to be any info on the time of the day but I suspect we can probably assume the numbers are higher at peak times than they are off-peak, meaning an even greater impact on cities.

This isn’t the first study to highlight this issue, but it does seem to be getting bigger over time:

These numbers suggest that ride-hailing is hitting traffic harder in many cities than previously understood. For example, independent research by the San Francisco County Transportation Authority in 2017 showed that, as of fall 2016, TNCs generated about 6.5 percent of the county’s total VMT on weekdays, and 10 percent of weekends. And the agency found that the grown in ride-hailing was already a major contributor to noticeable slow-downs on San Francisco streets.

Now, the Fehr and Peers memo indicates that TNCs accounted for nearly twice the VMT in San Francisco than the SFCTA had estimated, said Gregory Erhardt, a professor of civil engineering at the University of Kentucky who has researched Uber and Lyft’s effects on public transit ridership. That means the services are likely delaying commuters more, too. “This difference may be due to the continued increase in TNC use over the intervening two years,” Erhardt said. “With nearly double the TNC VMT, we would expect the effect of TNCs on congestion to be much higher in 2018 than was estimated for 2016 conditions.”

Now just because a trip is taken in the back seat of another vehicle instead of on your own, wouldn’t necessarily mean more overall vehicle travel if we saw a pure replacement. The problem though is that much of the travel by these “TNCs” actually doesn’t even have a passenger in the back, as drivers travel between trips. This is terrible for ‘real’ vehicle occupancy levels, as if someone needs to drive 2 kilometres to pick up a passenger for a 5 km trip and then needs to travel another 3 km to pick up their next trip it means that they’ve actually drive 10 km to shift someone 5km – a vehicle occupancy of 0.5!

A bit of data now seems to indicate that 38-46% of distance driven is for journeys without a passenger.

The findings also answer a question that many transportation researchers have been keen to find out: How many TNC miles account for an actual passenger, versus an empty backseat? It might be even fewer than other researchers have guessed. On average, between the six cities, just 54 to 62 percent of the vehicle miles traveled by Lyfts and Ubers were with a rider in tow. A third of these miles involve drivers slogging around in between passengers (“deadheading,” in taxi-driver argot); 9 to 10 percent are drivers on their way to a pickup.

It’s funny now to think that companies like Uber and Lyft used to promote themselves as helping to fix traffic congestion and support a more efficient overall transport system. These claims seem laughable now and the companies are backing away from some of their earlier statements.

Meanwhile, it’s become harder for the companies to stand by their original traffic-taming claims. In 2018, Lyft collaborated with the Rocky Mountain Institute, a clean-energy think tank, to produce a study that concluded ride-hailing vehicles were “more efficient” than private cars in several cities. But after vociferous criticism, the claim that was later retracted. “We represented certain conclusions as definitive, when in fact they are not,” the Institute wrote. Now, Fehr and Peers write that their analysis should help both Uber and Lyft “form appropriate narratives for both internal and external communication.”

Anecdotally it seems like every third car in the city centre these days is a Prius with a red sticker on its door – indicating that it’s probably ferrying someone around or (nearly just as likely) on its way empty to pick up someone. This adds huge traffic to our roads.

Meanwhile, an article last month in The Guardian highlights the issue of relying on these companies to replace public transport, as opponents of PT sometimes suggest we do. It is also relevant for discussions about Auckland Transports under-performing ridesharing trial in Devonport.

In 2017, the town in Ontario, Canada, embarked on an ambitious – and, to its critics, fraught – experiment. It handed responsibility for public transit to the ride-sharing app Uber.

Instead of buses or trains plying regular routes, it is Uber’s roving cars that function as the transit fleet. When a rider opens the app, Innisfil Transit pops up as the cheapest option to travel between a network of popular areas called “hubs”, such as libraries, the recreation centre or municipal buildings.

The costs per ride vary, but on average passengers pay an average of CAD$5 (£3), with the city subsiding the rest. Trips outside subsidised areas receive a flat $6 discount.

Two years later, the Innisfil authorities argue that the project has been a success. Ridership is high – in 2018 there were 85,943 trips – and many residents have embraced the service.“We just absolutely love Uber. I take it, my son takes it, my dad takes it. I take it,” said Shannon Kelly-Robb, who works at one of the libraries.

But beyond the excitement of essentially having subsidised taxi service, experts paint a more troubling picture of questionable economic and environmental sustainability. The city has now spent more on Uber than the traditional transit option it was considering, and has dramatically increased the number of cars on its roads, with worrying implications for air quality and the climate crisis.

The issue here is one of scaling and that the more successful these services are, the more they cost cities to provide. By comparison, in general with traditional PT, the more that people use it the less it costs.

“If you operate a regular bus system, you have a much better idea of what those costs will look like five or 10 years from now,” said Christof Spieler, an urban planner and author of Trains, Buses, People. “But if you have a system with too many people using it, and you can’t afford to provide [the service], how will you handle that?”

In fact, the town has taken the extraordinary step of deterring people from using Uber too much, capping the number of rides a resident can take per month. For mall worker Arrega, who has been “working like crazy”, that often means exceeding the limit midway through the month, although the town allows riders to apply for an exemption. It has also increased the cost of a ride by $1.

Capping trips will have a much larger impact on those on low incomes or who aren’t able to drive. This is even more notable as for much of North America, although much less so in Canada, public transport is mainly seen as something poor people use. So if those groups are priced and capped out of the market it could have wider social impacts.

Finally, there remains a few large elephants in the room with these companies

What’s more, with the precarious finances of larger ride-sharing companies now well known, municipalities have yet another reason to worry.

“We know that Lyft and Uber are both losing massive amounts of money. We know they are both underpaying their drivers,” said Spieler. “If you’re doing this to save money, what insurance do you have that five years from now, it will still save you money?

In the meantime, Innisfil officials have openly mused about fixing some of the problems of Uber – by turning to Uber. The latest suggestion to the cost overruns? UberBus: a bus that runs a fixed route, just like a normal city bus.

One of the things I find interesting about many new transport technologies and ideas is that despite the promise of something new and revolutionary, they almost all seem to eventually evolve to look similar to something that already exists.

Share this