My little stoush with Uber over the last few days — observing that its advertising seems to be presenting itself as a threat to all public transit, including rail — went modestly viral, coming close to a pageview record for this seven year old blog. Uber quickly pulled the offending ad.
Again, as I laid out here, if Uber and its ilk — collectively and misleadingly called Transportation Network Companies (TNCs) — successfully draw customers away from high-ridership transit services, rail or bus, the result would be several overlapping disasters for our cities, including massive increases in congestion, emissions, road space demanded by vehicles, and inequality of opportunity.
And of course, there will always be people saying that as long as customers want something they should have it, and the consequences don’t matter. This is how we got all of the environmental problems we deal with today, and many of our problems of inequality.
Meanwhile, the resistance is getting traction. A Boston Consulting Group report, flagged in Citylab, argues that autonomous taxis will destroy the passenger rail industry, including urban subways. Anyone with the slightest sense of geometry can see immediately what a disastrous increase in vehicle trips that would be.
Then, Greg Lindsay, a leading observer and commenter on urban tech, noted this:
When @FTAlphaville is quoting @humantransit approvingly, you know the tide is turning, and turning fast. https://t.co/IetlNGUCtv
— Greg Lindsay (@Greg_Lindsay) October 6, 2016
Alphaville is a blog at the Financial Times, usually a pro-business outfit. But they’re not just quoting me; they’re sounding a serious alarm:
At FT Alphaville we’ve questioned the rationality of glorifying a business model (Transportation Network Companies, ‘TNCs’ like Uber) that undermines decades worth of urban planning work focused on encouraging mass transit options like buses and trains, whilst marginalising petrol-guzzling space-consuming single-occupant cars. It seems backward, to say the least.
But the tech world seems oblivious to the limitations posed by urban geometry, seemingly convinced that app-controlled taxis can overcome the space constraints of London Bridge at 5pm more effectively than non-app controlled taxis. Why this should be the case, however, is never clearly explained.
They also cite this paper by Matthew Daus, outlining the numerous dangers of the current prevailing TNC business model, including environmental dangers as well as issues of equality of opportunity. (See his page 20 of his report for the formal litany.) Here’s the critical bit:
TNCs have grown at a near exponential rate, adding a significant amount of automobiles on the streets of already congested cities. For example, Uber grew from zero (0) drivers in 2012 to 160,000 actively partnered drivers (defined as drivers that have completed more than four trips per month) by the end of 2014 in the United States alone.
As demonstrated in the graph below, the rate of growth has risen rapidly since July 2012:
Uber and its fellow travellers can “grow” like this by continuing to flood dense inner cities with cars, but of course we will run out of space, so that’s not going to continue. The danger is that the attempt to do this will strangle all of our other options, notably public transit, so that once Uber has taken over our streets we’ll have no alternatives, even though they are all sitting in self-made gridlock, shoving cyclists and pedestrians out of the way. This is exactly how cars took over the world, 60-100 years ago.
In other words, this seems to be an issue where only strong government intervention, over the noisy objections of the techno-libertarians, is going to save us.
Of course, there is a free-market solution to this problem, which is to charge for the actual market value of limited urban road space. Once most cars on the road are run by some corporation instead of by individuals, it may be easier to impose the necessary free market forces so that traffic is limited to match the space available. That, in turn, would drive massive demand back to space-efficient modes such as public transit, but only if techno-libertarian public relations campaigns haven’t destroyed those options first, by encouraging apathy about them.
So it’s an interesting time. Maybe the tide really is turning.
Don’t agree that all environmental problems came about from “techno-libertarianism”.
Quite the opposite. Roads were built by government central planners all deciding that cars were the future and pushing through freeways like that. They were a product of their time. It was rational then. Even today, we have problems with gov’t officials pushing through roads that we might not actually need, or where another solution would work better.
There is nothing libertarian about driving for free on government roads. Or free parking. Or minimum parking mandates. Or single family zoning of large sections of our cities. Nor is there anything libertarian about dumping your rubbish/pollution in public or common spaces where it affects third parties- if it is your mess, you should clean it. That’s just personal responsibility. If you can’t clean it yourself, then you should pay to have someone clean up – that is fine.
I do not see Uber any more as a threat than people walking, driving a car or taking their bicycle as a “threat”. If public transport is good value for money, people will prefer to use that.
Even if Uber “takes over our streets” so what? Cars have already taken over our streets. It will just replace some cars with other cars. The beauty is that Uber has an app which calculates the distance and route travelled by GPS plus data on the time of travel. That’s perfect for calculating road user charges from that data.
yes, whoever uses Uber has already given up on his privacy, so the road to satellite congestion pricing is wide open.
The threat is the so-called “techno-libertarians” convincing people and politicians that geometry doesn’t matter and there is no need for further investment in bus lanes or transit systems.
If that happens, bus systems will be seriously damaged by the time the congestion traffic nightmare happens with driverless cars (it won’t happen before, too pricey), and we’ll have to start from scratch, just like in the SO private car era, fighting to reserve public space for more efficient ways of traveling.
The theoretical question here: what is the elasticity of transit usage with respect to TNC usage? There are several potentially offsetting effects, and the net magnitude of the change is an empirical question and may vary from place to place. Off the top of my head:
1. TNCs could make congestion worse by grabbing a transportation market segment which currently takes transit at peak times. A TNC taxi will always be faster than a bus, assuming they’re both using general traffic lanes. Some people who currently make a $2 transit commute will find a $15 TNC taxi sufficiently appealing. Others would be willing to switch to an $8 shared-ride style TNC service. Both of these possibilities move travelers into less space-efficient modes.
2. TNCs could make congestion better by pulling part of the transportation market segment that currently drives alone at peak times into shared ride service. For somebody who mostly drives alone, $8 shared ride service could cause a switch into the more space-efficient mode.
3. TNCs could make congestion better by making it conceivable for people to sell their cars. These people will switch from paying for transportation services mostly in bulk to paying for them mostly on the margin, and this will push people to travel less and to travel more cheaply. And high-ridership transit is the lowest-cost kind of transportation service. Studies using Zipcar data have shown exactly this sort of transit-supportive effect, where a small but quantitatively meaningful fraction of subscribers get rid of their car and ride transit more often– and of course Zipcar is a service which is much less convenient than a TNC. (Though I’m not confident that these studies are entirely believable.)
I don’t think there is any evidence that TNCs have made traffic congestion worse anywhere. In NYC there were some dueling claims about this, but the city ultimately released a report that concluded that Uber had not made traffic worse; instead, it mostly seemed to cannibalize riders from yellow taxis. On the other hand, “Lyft line” style services or robocars could change this, and I agree in principle that this is a possible outcome which we need to keep our eyes open for. But just because Uber wants to take riders from subway lines doesn’t mean that it can.
2 things here: yes, the external costs are born by society, not Uber and Lyft. But you forgot an equally troubling factor for transit – gold plated pensions and healthcare provided by transit providers. Hence you have situations like WMATA where despite very high fares, service is constantly reduced. Average employee compensation is 124k. Uber and Lyft don’t offer such generous (or any) pensions and healthcare.
And politicians have shown no guts to stand up to unions to address this.
Something has to give, and reducing frequencies of service has dramatically reduced ridership.
I’m not sure that the “transportation network company” model works at all, shout of receiving huge influxes of venture capital in a riverboat gamble to recover the huge current Uber, Lyft et al operating losses in the companies going public at some theoretical time in the future.
To really “work” the TNC financial model depends on self-driving cars “working.” But that is highly doubtful for many reasons, particularly the “robot-human handoff” problem. More in article about self-driving cars here: http://www.calrailnews.net/self-driving-cars-not-worth-the-trouble-wont-replace-transit/
Seen on my Chicago Transit subway car an advertisement that suggested that I “Bust a Commute” with Lyft. Next to it another that wanted me to “Do the Ride Thing” with a Lyft.
Of course they were not suggesting that I destroy my subway system and of course they were not suggesting that I was doing wrong by taking the subway….
We’ll see how it all shakes out, but if self-driving taxis actually ever happen, this seems like it would be a boon to passenger rail, since it would make it much more accessible to people who live in suburbs which lack the density to support quality public transportation. Of course this could create other effects like encouraging sprawl, but we’ll see.
For transit to compete with so-called car-sharing it must expand and protect exclusive dedicated lanes of travel. It then must maintain the fastest and most comfortable transport technology practical within those lanes.
Cost (time and money) followed by comfort are the driving factors in the decision between shared versus personal transit. Driverlessness pushes money costs down especially for less shared vehicles. But geometry pushes time costs up as vehicle size goes down.
Utopian Capitalism is an oxymoron because of inequality. Even on Startrek people mostly walked all day. The best always wants to be shared. The easiest way to share is to bring people together. Walking is the only transport mode that is both shared and personal at the same time. Large walkable only areas are quintessentially urban. Especially when they include all of life’s functions like home, work and leisure within the same area.
One thing I’ve noticed is that Uber and Lyft drivers often dropoff / pickup at bus stops or even bus lanes, causing undue delay. I wonder if the geolocation is sufficiently good to disallow that on the apps.
“Of course, there is a free-market solution to this problem, which is to charge for the actual market value of limited urban road space. Once most cars on the road are run by some corporation instead of by individuals, it may be easier to impose the necessary free market forces so that traffic is limited to match the space available. That, in turn, would drive massive demand back to space-efficient modes such as public transit, but only if techno-libertarian public relations campaigns haven’t destroyed those options first, by encouraging apathy about them.”
This wouldn’t drive everyone to public transit per se; it would drive them to transit. If pinpointed demand-responsive tolls could be achieved across an entire road network (which is of course a ways away), then the prices would encourage corporations to offer more space-efficient vehicles because then they can offer a lower price to their passengers. They’d do the math and see that a service splitting a $20 toll across 20 or 40 people in a vehicle would attract people who wouldn’t want to pay the full cost themselves. That dynamic doesn’t work on free roads because you can’t split the cost of a 20-minute traffic delay across 20 people; everyone has to wait whether they’re alone or on a bus. If affordability for low-income people is an issue, the government could provide vouchers. Those could end up being much cheaper than government trying to run transit itself.
There could be some hitches with this in practice, but if roads are priced right, we might end up with excellent private transit.