3 ms·

There are even investors and syndicates inserting themselves into the medallion markets now [1]. When you see even these market participants pulling out altoget

There are even investors and syndicates inserting themselves into the medallion markets now [1]. When you see even these market participants pulling out altogether, then I can easily believe the autonomous passenger vehicle is within 5-20 years of obliterating the taxi industry (though I doubt that will happen). I'm not aware of a way to short medallions, so I can't tell if that is going on. So far as I can tell, these people are still participating in the medallion market and various secondary markets, despite the recent downward tick of medallion prices. I'm guessing they and the industry in general have a lot of political pull, and are counting on that influence to wring another few decades out of conventional taxis before autonomous options mutate them into a different-looking industry. Unless there is a really broad-based public backlash that makes it political suicide to continue licensing however, even if autonomous vehicles are ubiquitous I can't see municipalities and various institutions and agencies in general giving up the affiliated revenue stream. Even if we end up with something like Johnny Cabs [2] and no humans drive taxis any longer, I expect some form of government-imposed licensing to still exist. [1] http://venturebeat.com/2014/06/21/add-taxi-medallion-investors-to-the-list-of-ubers-enemies/ http://venturebeat.com/2014/06/21/add-taxi-medallion-investo... [2] http://www.empireonline.com/features/helpful-movie-robots-that-will-ruin-your-life/p10 http://www.empireonline.com/features/helpful-movie-robots-th...

fab13n 11y ago
The thing is, autonomous vehicles don't need to take the taxi niche frontally. Here's how it goes, if legislation is too slow: 1. families buy one autonomous car instead of several driven ones: when dad's arrived at work, the car goes back to become mom's car for the day. With a nice twist: when the car goes and fetch kids from soccer, mom doesn't have to be in it. 2. People pool cars privately: one car for a couple of families if they're light users, X/Y cars for X families for more intensive users. It's shared property, neither a taxi activity nor even a company providing a service. 2bis. Some AirBnB of autonomous cars facilitates mutual lending of cars, in parallel with the development of privately organized pools 3. Cars are leased instead of bought upfront, from companies embedded with the AirBnB-of-autonomous-cars. What you have now is effectively an autonomous taxi company. The economic pressure to get rid of drivers will be enormous. Sure, car makers and drivers won't like it, but all of the money that won't go there will be up for grab by other sectors. Think of trucks: not only autonomous trucks will drive 23h a day, but everyone who sells anything moved by truck will have their margin augmented by skipping drivers' salaries.
- yourapostasy 11y ago
  Excellent observations. For trucks, there are other big downstream effects. The big diesel engines preferentially run continuously, and with autonomous driving we'll see run times between engine shutdowns increase pretty steeply. Manufacturers can tune engine development accordingly so continuous running is the norm. Diesel consumption per unit mass will enjoy a marginal improvement; diesel repair shops will see a marginal decline. If we see concurrent automation of refueling, truck stops will experience a dramatic decline. If medium and light truck bodies are aligned to a standardized set of chassis, then some of the fleets of delivery trucks that lay idle half the day or longer today can run around the clock, swapping bodies and switching duties when needed. If we can automate quiet garbage pickup, then a garbage truck that usually only runs during the day can run around the clock. If we can automate safe delivery of liquid fuels, then entire fleets of various fuel delivery trucks will lay off a lot of drivers. The pattern I'm seeing is the actual driving might get automated, but the endpoint activities might not be subject to automation just yet, but that might only be a matter of time and possibly standardization.

fab13n 11y ago

The thing is, autonomous vehicles don't need to take the taxi niche frontally. Here's how it goes, if legislation is too slow: 1. families buy one autonomous car instead of several driven ones: when dad's arrived at work, the car goes back to become mom's car for the day. With a nice twist: when the car goes and fetch kids from soccer, mom doesn't have to be in it. 2. People pool cars privately: one car for a couple of families if they're light users, X/Y cars for X families for more intensive users. It's shared property, neither a taxi activity nor even a company providing a service. 2bis. Some AirBnB of autonomous cars facilitates mutual lending of cars, in parallel with the development of privately organized pools 3. Cars are leased instead of bought upfront, from companies embedded with the AirBnB-of-autonomous-cars. What you have now is effectively an autonomous taxi company. The economic pressure to get rid of drivers will be enormous. Sure, car makers and drivers won't like it, but all of the money that won't go there will be up for grab by other sectors. Think of trucks: not only autonomous trucks will drive 23h a day, but everyone who sells anything moved by truck will have their margin augmented by skipping drivers' salaries.

yourapostasy 11y ago
Excellent observations. For trucks, there are other big downstream effects. The big diesel engines preferentially run continuously, and with autonomous driving we'll see run times between engine shutdowns increase pretty steeply. Manufacturers can tune engine development accordingly so continuous running is the norm. Diesel consumption per unit mass will enjoy a marginal improvement; diesel repair shops will see a marginal decline. If we see concurrent automation of refueling, truck stops will experience a dramatic decline. If medium and light truck bodies are aligned to a standardized set of chassis, then some of the fleets of delivery trucks that lay idle half the day or longer today can run around the clock, swapping bodies and switching duties when needed. If we can automate quiet garbage pickup, then a garbage truck that usually only runs during the day can run around the clock. If we can automate safe delivery of liquid fuels, then entire fleets of various fuel delivery trucks will lay off a lot of drivers. The pattern I'm seeing is the actual driving might get automated, but the endpoint activities might not be subject to automation just yet, but that might only be a matter of time and possibly standardization.