1. This is probably further out than a lot of people assume. Reliable door to door fully autonomous in dense cities is one of the hardest use cases to solve but it's more or less what you need for self-driving taxis.
2. The pricing probably also won't "plummet." The IRS mileage rate (about 53 cents/mile) is probably a reasonable floor to use. That's still about half current Uber rates.
https://www.irs.gov/newsroom/2017-standard-mileage-rates-for...
"The standard mileage rate for business is based on an annual study of the fixed and variable costs of operating an automobile. The rate for medical and moving purposes is based on the variable costs."
1. Maintenance. Electric vehicles require far less maintenance than ICE vehicles. (First thing I found on Google: https://insideevs.com/ev-vs-ice-maintenance-the-first-100000...)
2. Vehicle size. Once these systems are up and running, do you think Waymo & Uber, once their systems reach any level of maturity, are going to send you a 4 person vehicle to pick up 1 person? This reduces:
a- the capital cost of the vehicle
b- its cost of maintenance (less parts, etc)
c- the amount of energy required to get from point A to point B
Which brings me to...
3. Price of electricity. Waymo, Uber, et al will get their electricity at wholesale rates.
4. Low cost of capital. The cost to access the capital to buy these vehicles will be lower than anything a peon like you or me could ever access.
5. Bulk-buying 1,000,000 vehicles. No dealers, no dealer commissions, zero customizations, less parts, smaller vehicles, no car manufacturer marketing budget...
6. Maintenance scaling. The need to scale maintenance operations country-wide is going to lead to its own interesting effects. With a hard limit on the types of vehicles in a network, most cleaning and general maintenance will, in time, be doable by human-monitored robots.
7. Ride subsidization will effectively crush any margins any of these companies could ever hope to have.
8. Shared depreciation. Depreciation will be significant per vehicle but overall, cheaper, as each vehicle will have more utilization.
I'd hesitate to guess at the effects of all of the above, but it's not much of a stretch to anticipate an additional 50% reduction here, barring any unforeseen taxes, of course.
You might pay more for the network with great coverage and lux vehicles.
You might pay less for the crappy network with dirty cars and plastic seat buckets.
Etc.
And regarding (8), many taxis are rented by drivers for their shift, with multiple shifts per car per day. Uber could be doing the same right now, but they'd rather rely on drivers not realizing the full value of their car and giving the difference to Uber.
I totally envision the points mentioned. I put the $$ figures for on-demand hail service subscription a month ago, on a HN thread as follow .
https://news.ycombinator.com/item?id=15644680
2021 : Electric Self-driving on-demand FLEET Car 1000 miles/month SUBSCRIPTION from Google, DiDi, Uber, Renault/Nissan,Tesla, VW,Toyota,GM for $400/month
2024 :same 1000 miles/month SUBSCRIPTION $200/month
------
> Literally none of the old assumptions will apply. Let's go through some of them.
> 1. Maintenance. Electric vehicles require far less maintenance than ICE vehicles. (First thing I found on Google: https://insideevs.com/ev-vs-ice-maintenance-the-first-100000...)
> 2. Vehicle size. Once these systems are up and running, do you think Waymo & Uber, once their systems reach any level of maturity, are going to send you a 4 person vehicle to pick up 1 person? This reduces:
a- the capital cost of the vehicle b- its cost of maintenance (less parts, etc) c- the amount of energy required to get from point A to point B
Which brings me to...
> 3. Price of electricity. Waymo, Uber, et al will get their electricity at wholesale rates.
> 5. Bulk-buying 1,000,000 vehicles. No dealers, no dealer commissions, zero customizations, less parts, smaller vehicles, no car manufacturer marketing budget...
> 6. Maintenance scaling. The need to scale maintenance operations country-wide is going to lead to its own interesting effects. With a hard limit on the types of vehicles in a network, most cleaning and general maintenance will, in time, be doable by human-monitored robots.
> I'd hesitate to guess at the effects of all of the above, but it's not much of a stretch to anticipate an additional 50% reduction here, barring any unforeseen taxes, of course.
> You might pay more for the network with great coverage and lux vehicles.
> You might pay less for the crappy network with dirty cars and plastic seat buckets. Etc.
I also suspect that ride pricing and scale will not be the determining factors. Look at the variety that still exists among car manufaturers after a century of competition:
http://www.thetruthaboutcars.com/2017/08/usa-auto-sales-bran...
There's such a wide variety in terms of both cost and experience that I'd be very surprised to see that drop off to one or two players.
The car brands know it, which is why they're panicking.
(Why ALL car brands? Because you're going to have companies like Bosch and Foxconn competing. I'd like to see Ford beat Foxconn at a challenge like this.)
This would be especially true if people continue to own or lease private cars, which I'm sure will continue at some level for decades at least. Look at how much stuff families keep in their minivans, for example. Or all the tradespeople who keep working materials in their trunk.
Or look at commuters. Taxi-ish providers like Uber and Lyft will have a hard time dealing with commutes because they'll have to buy a lot of vehicles to cover the peak that just won't get used much during the off hours. If people treat it like a minibus, where they share, then the providers might do well. But I suspect a lot of people will be willing to pay up for a private commute, at which point they might as well just have a dedicated car. In which case, the market will look a lot more like it does now.
However, you are right to bring up autonomous cars. I always thought it amusing that Kalanick et al were so anxious to bet the house on autonomous cars, when in fact, they are the company's greatest existential threat long-term. Once those cars are readily available and street-legal, there's nothing stopping Enterprise or Avis or the automakers themselves from becoming overnight competitors.
And once drivers are taken out of the picture, ride prices will plummet. Margins will also thin because of the likely extreme competitiveness of this market. I would not want to be an Uber investor that's for sure. In theory, the "winner" will be whoever scales the most (i.e. has the most capital to put into their fleet) and has the most effective marketing. In practice, prices may be close enough that people will rent car X because they like brand/model Y better.
If I were on Uber's board, I would advise against trying to win that war which they will lose once/if the automakers get in on it, and instead get creative about how to conquer the market in a more niche way.
Someone is going to have to program self-driving ambulances :)