Against TNCs as transit 2

“Preferably, long distance trips should be transported by fixed route mass transit as much as possible so that system efficiency can be optimized.  On-demand service can focus on short trips that may require higher flexibility.”

Normally, that would be inefficient, as buses can typically hand the short-distance trips at low marginal cost—the bus is already going that way, and seats are plentiful. But with COVID...bus capacity has a whole new meaning.

I remain deeply skeptical that TNCs can handle even the short-distance trips. TNC service is costly: every vehicle requires a driver, every service mile requires an additional .67 non-service ‘dead-head’. Studies say drivers make less than minimum wage, after vehicle depreciation, so much lower wage-costs than CDL trained bus drivers. However, there is probably a fixed # of people who are willing to drive for TNCs, at least on any regular basis, as it’s effectively a minimum wage gig.

It might work, simply thanks to the network effect—more riders, more OD pairs, more ‘dynamic carpooling’, less dead-head miles,  but likely still inferior to regular buses in terms of cost per rider. Which implies less transit service per dollar, and with fixed dollars, less transit service over-all.

It also creates a future hazard—people get used to having subsidized TNC rides, funded by money not being used while buses are out of service. But when demand for buses pick up, the demand for TNC is not going to fall. Then transit agencies would have internal competition for resources between their bus/TNC ‘divisions’ for rider subsidy. Making the reasonable assumption that TNC’s are worse on a per-rider cost basis, more resources devoted to TNC means less devoted to more efficient buses.

Against TNCs

It seems reasonable to disaggregate trips by trip-type, and serve them with different vehicles, but that simply requires the duplication of services to serve the same route. Given that transit service costs are mostly labor, that means a dramatic increase in labor costs.

Further, TNC service is great, but it’s costly: every service mile requires an additional .67 non-service ‘dead-head’. Public-subsidy taxis are probably more efficient than most paratransit, simply thanks to the network effect—more riders, more OD pairs, more ‘dynamic carpooling’, but likely still inferior to regular buses in terms of cost per rider.

 

Undeniably, COVID has changed things. There is now a very real need to reduce crowding, and prevent people from sharing the same air. That can be achieved by limiting the # of people per vehicle (effectively turning 40’ buses into 15 passenger vans) but that lowers the efficiency per vehicle.

Until we reach the point were we have CAV transit vehicles (breaking the link between service and labor costs), ‘massification’ of persons into a single vehicle is going to be the most cost-effective strategy for transit.

I think often of what happened at UTA—COVID causes a ridership crash, so UTA cut service on most routes. But they maintained a few routes, but added buses/hour to reduce crowding and facilitate social distancing. That attracted more riders, so UTA had to add more buses, such that the routes now have more ridership during COVID than before it.

"Shared vehicle, doesn't park, doesn't occupy curb space"

"Shared vehicle, doesn't park, doesn't occupy curb space"

This was the claim I heard made about shared vehicles. First off, the vehicle is rarely 'shared'. If the driver is in the vehicle, that's not a 'shared' ride. You are being chauffeured. The driver is not a passenger. (If I drive my kid to school, it's not a 'shared ride'--I'm driving him to a place, being a 'taxi-mama'.) It's only a shared ride when there are 2+ passengers in the car. Which is rare. How rare? In my experience, perhaps 1% of all the rides I've ever taken, did I share a ride with someone. (And that was in a major metropolitan city).

"Shared vehicle, doesn't park, doesn't occupy curb space"
A TNC is just a dial-a-ride taxi with a better fare structure. Taxis occupy curb space. They aren't in motion all the time. (Or perhaps even most of the time). They _do_ require parking space. That said, taxis are probably in motion more than TNCs, because they spend much of their time 'cruising' for fares, rather than parked and waiting for fares. So Taxi's probably take up more road space  than a TNC would (a moving car taking up about 3x as much space as a parked car, due to safety-required separation between two moving cars.)

I have seen Lyft Vehicles, pulled over on the side of the road, checking their phones, as they wait for Google maps to locate their next fare in the queue. (Peak times, not so much, as trips get added to the queue faster). 

Do they take up space in the middle of the day? Ie, do they take up all-day parking space, the way that a car driven to work does? No, clearly not. But cars still need to be stored someplace. (Likely in whatever suburban wonderland TNC drivers live). Which brings up a VMT issue: If a TNC driver drives someone to work, then drives home. Then, in the afternoon, drives someone from work to the grocery store, and then someone else from the store to home. Then they drive themselves home. Point being that the driver made two trips home (one morning, one evening) to drive one person to and from work. So if we assume that everyone will take a TNC to work, VMT is going to explode. Rule of thumb (from a PhD student who drove UBER was he drove 1.69 miles for every passenger mile). So 69% more VMT, to achieve the same amount of mobility? The roadway network will never withstand that kind of increase.

 Of course, it might. If the demands on the roadway network were spread over a wider period of time. Presumably, fewer drivers want to drive during congested conditions, because it's slower. And while they get paid both per mile and per minute, I believe the incentive structure also pays them by trip. So there is a strong incentive to maximize the number of trips you make. And even if not, it's better to be earning for both miles and minutes than sitting in traffic. 

So you have fewer drivers during congested times, and people have to wait longer to take a TNC during congested times. (So the reliability declines, and more drivers are inclined to say "Screw it, I'll just drive myself"). At which point, you've basically enacted congestion pricing. Ie, when things are congested, it costs more to drive.  

Uber and Lyft are NOT transit

TNC are  not mass transit; it's a taxi. (There is no massification)

Shared ride _is_ transit. It's a jitney. Jitneys also follow routes. 

The idea that you can combine the 'on-demand' dial-a-ride of a taxi with a jitney is fallacious.

How much detour are you willing to tolerate? And when does it make sense to stop and add another person.

Massification relies not only on spatial convergence, but also temporal convergence: 

Same people in the same place at the same time. 

Which implies it will mostly take place during times of peak demand. 

Which becomes the classic transit agency problems: Sufficient peak capacity is excess off-peak capacity.

The real innovation of TNC's is not the 'dial-a-ride'. It's the ability to use 'surge' pricing to manage the match between supply and demand. It both reduces demand (through higher prices) and draws in additional drivers (through higher fares). 

This is core to the whole TNC model, as it draws in latent supply.

'Latent supply' is how the whole TNC thing works. Most cars are parked most of the time.

Commuting in the morning, the commute+shopping in the evening.