Rain is not unpredictable.
Define “depreciation”.
If you’re talking about that ‘your car loses 30% of its value the second you drive it off the lot’ silliness, then cars bought by your Uber owners suffer that loss too. That much is equal across the board, and if your Uber owners care about it any more than I do, it’s included in what they charge.
But neither Uber or I care about that kind of depreciation because neither of us plan to flip our car. We’re both planning to use our car, which is where the real definition of ‘depreciation’ kicks into play: the value of the vehicle as averaged across time until its effective usable life ends. Which is to say, how much it costs divided by how fast it wears out.
Uber cars will wear out pretty fast; they will put an order of magnitude more of mileage on them than I put on my cars. (Not a difficult trick since I drive around 5,000 miles a year.) My cars will last much, much longer than an Uber car.
But that’s beside the point: you argued that my car will depreciate while it sits in my driveway. But by any practical measure that’s not true - my car doesn’t wear out at all when it sits in the driveway. Sitting in the driveway extends its usable life. Sure, when I finally do trade it in I won’t get much for it due to its age, but when an Uber owner trades their car they won’t get much either because it’ll be a worn out husk. In no real sense does my driveway damage the value of my car.
You’re also not factoring in depreciation. Say (HYPOTHETICAL SCENARIO 1: ) you typically like to purchase a new $30k car, drive it 10,000 miles per year, then sell it at 100,000 miles. So 10 years later, you always sell your 10 year old vehicle with 100,000 miles on the odometer. You only use it to drive to work and back.
Instead, say (HYPOTHETICAL SCENARIO 2: ) you happen to have a group of 10 friends, all with the same cleanliness habits as you, and identical preferences in car purchasing, and your work schedules and home locations just happen to perfectly line up so that you drive to your work location in the morning, which is right next door to friend A, who then drives it to his work location, which just happens to be next door to friend B, etc. The process is reversed at the end of the day.
You all decide to pool your resources ($300k), and buy a new $30k car, that you share equally. The other $270k sits in a joint savings account. With each person driving 10,000 miles, at the end of the first year you sell the car with 100,000 miles on the odometer, and buy a second new car, withdrawing another $30k. This continues until the end of year 10.
So at the end of 10 years, with scenario 1, you’re selling a 10 year old car at 100,000 miles. Other than that, you have $0 in your savings account. You’ve spent the last 10,000 miles riding in a somewhat dumpy 9 year old car, which lacked many of the creature comforts, safety standards, and efficiencies of the year 2028.
With scenario 2, you’ve sold ten 1-year old vehicles at 100,000 miles (split 10 ways), plus all the interest gained from the sales over the years (deposited in the same joint account). You still have all the interest accumulated from the original $300k invested ten years ago. You’ve spent zero time riding in a car over a year old. You also have had complete use of an empty home garage, which you’ve converted to a workshop. Or maybe decided to sublet.
Based on my unscientific sample size of 1 looking at KBB, a 1-year old car with 100,000 miles is worth approximately 3X a 10-year old car with identical make, model, and mileage.
I don’t believe that driver downtime is that big of a factor in bus runs, but whatever. I’m willing to pretend that people who don’t drive cars to work don’t exist at all as far as this discussion is concerned.
YOU may be dismissing the idea that normal individuals will release their privately owned cars into the wild as Uber taxis, but as best I can tell you’re the only person in this thread that is doing so. A major point of discussion in here is how being an Uber owner will be so magically profitable that everbody’ll want to do it.
Bull, and stop moving the goalposts. You asserted that the idea of the Ubers spending their free time driving in circles was a viable solution, and that is the ‘issue’ that I had a problem with. It is most definitely not solved.
And yes, there exist workplaces that have ample parking. In those places the parking spaces will be filled with employees who aren’t stupid enough to think that paying through the nose for daily Ubers is a good idea. And if those folks can’t find parking spaces because Uber is squatting in in their lots, I predict litigation.
And of course there are lots of workplaces without dedicated parking lots, which of course are the workplaces under discussion in the discussion you’re responding to. Don’t move the goalposts.
Gobs of new infrastructure would be required, though, as you well know. Every single parking garage in the country would need their entry and exit gates replaced, at the very least.
You keep changing goalposts at the drop of a hat. And I don’t believe for one second that there’s anywhere near enough roadside parking to accommodate literally the entire roadside fleet. That’s absurd. And what roadside parking is available often has meters. Do Uber cars carry coinage? Yet more widespread infrastructure changes ahoy…
If there were enough empty roadsides to accommodate all the cars then parking lots wouldn’t exist.
And will Ubers be programmed to flee when the cops or tow trucks come? Is that how it’ll work?
And thus, parking lots will remain a thing.
Like hell it will - you think I want to pay twenty times my gas bill? Or electric bill, or whatever it is? My car is going to rest on its laurels while I shop, because anything else would be insane. If self-driving cars couldn’t be set not to go wandering off like an unattended child, burning gas and adding mileage for no benefit whatsoever, that would probably the one thing that could convince me not to buy one.
In the world you posit Ubers are a scourge on the face of the earth, wandering aimlessly, clogging roadsides, alighting on any free square inch of asphalt available like a swarm of insects. Almost makes me wish I tolerated guns so I could blow holes in any that try to squat in my driveway!
Tell that to someone trying to get a cab in New York City in the rain! 
It’s unpredictable in the sense that, unlike rush hour traffic, it doesn’t have a schedule, nor is it predictable in the timeframes that it would take to mobilize large numbers of additional vehicles that wouldn’t normally be out there anyway.
I am 99.99% confident that the anti ‘self-driving car/everyone will always own their own car’ side of the argument in this thread is going to look very, very silly in about 10 years.
I’m putting in a reminder in Google Calendar to revisit this thread in September 2028 so I can come back and point ‘n’ laugh.
I’m 100% confident that nobody in this thread has expressed opposition to the existence or use of self-driving cars, and that nobody in this thread has asserted that everybody will always own their own car.
Exactly. I think there’ll be an uptick in ridesharing, but I seriously doubt that it’ll become the default paradigm for personal transportation anywhere outside of very densely populated urban areas where taxis and other pay-as-you-go systems are already popular.
The whole argument seems to be predicated on some kind of vague notion that because there will be more cars than demand, and that prices will be correspondingly low, and that we’ll save money versus owning our own autonomous car.
Which doesn’t wash; the cost per mile will still be the same whether I own it or you own it. The difference is that if you’re pimping your car out, you have to charge MORE than what it costs to run it in order to make a profit. I don’t- I just have to pay what it costs me to get where I’m going. So by definition, it’s cheaper for me to use my own car. Another way to put it is whether or not it’s cheaper to ride-share for yourself or for someone else? You have to charge someone else enough to make a profit, but if you’re riding in your own car, you only have to pay what it costs you.
And the autonomous nature is not pertinent either- by that point ALL cars will be autonomous.
If the system is going to rely on what amount to taxi companies to provide cars instead of some kind of loose network of private vehicles, they’ll have an incentive to limit the number of cars, thereby allowing them to raise prices. Again, owning my own car comes out cheaper for me.
I’m not seeing the situation where it’s cheaper per mile/per trip for me to rideshare vs. use my own car. Yes, if I drive very little, it doesn’t make sense to own a car, but that’s true today. Basically the economic model doesn’t really change just because cars are autonomous- you STILL have to charge more than you pay in order to make a profit, and owning your own car will mean that you just pay the cost, not someone else’s profit.
Today’s parking lots are very inefficient. You have to have room for the doors to open and they ate designed so that any car can leave at any time. Imagine an AV parking garage that is just a queue of cars. Within a class of cars we don’t care which one picks us up. You could even have a 9-puzzle model where leaving just one empty space lets you get any vehicle out by shifting the rest. I’d guess that parking lots will fit at least twice as many cars. In all likelihood it will be 3x or more.
Cost is the lower bound of price, ongoing businesses can’t price below cost. Cost is a useful starting point because it isn’t beholden to market forces.
The concept here is whether or not someone will choose to own a car. For simplicity let’s state that the incremental cost of operating any car, personal or AVUber, is $1 per Standard Trip inclusive of maintenance type items. Let’s also state that buying and owning a car costs $300 per month, including financing and insurance. I would expect AVUber to be slightly more expensive in reality, but let’s not complicate the case just yet.
If you average 5 trips a day, you would spend $450 per month on vehicle expenses, or $3 per trip. 10 trips = $600/mo = $2/trip and 20 trips = $900/mo = $1.50/trip.
AVUbers would be used for many trips a day, lowering the fully burdened cost significantly, while personal cars are used much less frequently, increasing the fully burdened cost for the individual.
Each person offered AVUber as an alternative to ownership can look at their pricing, which is bounded on the low end by the total cost per lift, and decide if it is or isn’t cheaper than their total cost of ownership if they buy their own car. But you have to include the fixed monthly costs as well as the cost per trip. If I’m a light 5 trip/day user, my ownership cost is $3 per trip, it’s easy to see how AVUber could offer prices well below that, as their high usage puts their total cost approaching $1 per trip.
If you already use your car a lot, AVUber is going to be less useful, unless they come up with a convenient ride sharing idea that approximates the utility of car ownership, but spreads the cost even further.
I know that costs are the lower bound on price under most circumstances, but while you can amortize your fixed costs (insurance, financing) over the total number of rides, you can’t do that with your variable costs such as gas, tires, etc…
I can definitely see how if you didn’t use your car much, how some kind of ridesharing system might be cheaper. And I can also see how in today’s ridesharing systems, the bugaboo is the cost of the driver versus someone’s own time.
But I suspect that most people in the country use their cars such that it still won’t be a good deal for them to forego ownership and go with ridesharing. I mean, the flexibility alone, combined with the fact that the overall cost to own a car and price per trip for ridesharing is going to be similar. I’m not convinced that even if you utilize a car 95%, you’ll be able to outprice car ownership significantly on a per-trip basis, especially if most cars are involved in ride-sharing. That’ll mean that the number of trips any given car gets is small, and also that the price per trip will be close to cost. In other words, you won’t get too many trips (raising the per-trip fixed cost), and you won’t be able to charge much for them either.
It feels to me like this whole ridesharing things relies too much on some sort of community do-gooding mentality, and not enough on profit motive for it to work long-term.
This is an idea that’s going to start with the low hanging fruit, families with multiple cars who use one car to get to the train station 5 miles away, stay at home parents (without little little kids) who run local errands with the second car, teens who mostly go out on Friday to the movies, that sort of thing. Folks who have low utilization but are still cheaper than a human driven taxi. Hell, my town has a 10 year wait for a parking space at one local train station, you know there are a bunch of people getting in their cars every morning gearing up for a long walk from the only available parking spots to the station, and who would be THRILLED to get dropped off every day instead.
Some people will be up at the top of the tree. Drives 70 miles to work every day, has a 3 year old who loves to throw Goldfish crackers everywhere, has a big smelly dog who loves to ride in the car, has a trunk full of Amway products, that sort of thing. They are unlikely to share even if it’s dirt cheap to do so.
In the middle is everyone else, the market will sort through them, some will take the AVUber idea, some will buy and sign up to be providers, some will stick with ownership and not share at all.
BTW, I love this idea. AV Fleet storage, you just need to be sure it’s first in first out, and not last in first out.
Not remotely the same analogy. The analogy to the OP would be that you had to rent a phone from someone every time you wanted to make a call.
Another point in favor of costs of scenario 2 (Aka the idealized AVUber analogy) is warranty coverage. Typical auto manufactuerer’s drivetrain warranties are 5year, 60,000 miles. With scenario 1, only 50% of the miles driven are under the manufacturer’s warranty. Scenario 2 has 60% coverage. Some manufacturers have a 5 year, 100,000 mi drivetrain warranty, which would result in 100% drivetrain coverage over the entire 10 year period with scenario 2.
That too is a horrible analogy. A phone is something we use 150 times a day according to some studies. And it’s a device that’s designed to alert the user in real time when things happen (incoming text, alarm, etc).
A car is something you use maybe 2 or 3 times a day most days, even if you own one. And there’s never any need for your car to alert the user when the user isn’t in the car. (Or the car is being stolen, but that can’t happen if you don’t own the car.)
I don’t expect that many parking lots will change - most of them are already in place and sized appropriately for the store they’re next to, and also a slab of asphalt with painted stripes on it is really easy and cheap to maintain. They break down and accidentally trap the cars within their mechanisms at most once or twice a year.
Where you’re going to start seeing fancy stuff is in parking garages…but that won’t have anything to do with autonomous cars. It’ll just have to do with technology being cheaper and space being more expensive. You could already drive your car onto the loading platform, get out, and let the garage suck the car in and stow it where it likes, and retrieve your car the same way. (And I vaguely recall hearing of such things already existing.)
Whereas I see this as an excellent justification for a family that previously had two or more cars trading that for a single, personally owned self-driving car. You can use it as your own personal Uber for things like being dropped at the train station or dropping the kids off at school or appointments, sending it home afterwards if there’s somebody there that will need to use it in the meantime or just if it’s inconvenient to leave it parked at the train station.
Of course this won’t work for everyone; if two people in the house have exactly the same work schedule at opposite ends of town then it might not work so great, but there are a lot of multicar households that would be able to shed some or all of their secondary cars if they felt so inclined.
If you do this, you give up your main stated reason for wanting to buy your own car: 24 hour availability. E.g. you decide to head to work, and realize your car hasn’t come back yet because your son was late getting to school.
I suppose on those days, you’d call for a self-driving Uber. But if you’re going to do that, why own a car at all?
The assumption, obviously, would that you’d only do this if your family situation is such that you thought you could juggle things appropriately. If a family’s situation is actually that there’s a dude who gets on the train every morning and comes back every evening and in the time betwixt he had otherwise just left his car parked at the train station, it might behoove that family to get a self-driving car to drop him off rather than to have somebody else do it.
On reflection, I concede that there aren’t a lot of families in that situation. Separate cars for everyone!!
(And, having been part of a family at one time in my life, I can say with confidence that what will happen is that the father will buy a self-driving car in addition to all their existing cars, never let anybody else ‘drive’ it (that is, use it without him), and he’ll leave it parked in the train station lot all day while his family all has to share the '71 Pinto with no hubcaps and its trunk held shut by a bungee cord.)