I got into the details of their methodology, and I think they are making a fundamental mistake in how they treat depreciation. They claim that for IO depreciation should be based on residual value, but for TaaS it should be based on total life miles. Firstly, it is problematic to use two different and unrelated methodologies to make the comparison. This is apples and oranges. Second, the method used for TaaS is unrealistic in that it assumes that all miles will be billable and obtainable in the market for riders.
Let me spell this out for TaaS, they assume 500k lifetime miles on a $30k value vehicle. They apply straighline depreciation of 6c/mile. While this may be suitable for accounting purposes, it is not realistic from a value perspective. If it were, then the same depreciation would apply to IO usage. How? Suppose a car is used for a year and driven 20k miles. It then has residual value based on 480k miles left, which at 6c/mile is $28,800. Using this residual value the IO depreciation is just $1200 or 6c/mile. If this sort of depreciation had anything to do with the actual value of the vehicle, the TaaS operator should be willing to buy this lightly used vehicle for $28,800. So what I have shown here is that these analyst are using to incompatible methodologies to compare the economics of TaaS to IO. This is bad form intellectually.
But it gets worse. Let me propose a disruptive technology that would in fact disrupt the disruption they desire to see. In fact, Musk has hinted at this disruption already. All we need is a means to allow private owners of EVs to participate in offering ride sharing services. So I buy the car for $30k. I can drive it as much as I want and I can let the car drive folks around. Let's suppose I use this car for a year, putting 20k miles on it. I get sick of it and want to sell it. But I also know that I can net 7c/mile pimping it out for the next 480k miles. Let's say it can do this at 120k miles per year over the next four years. So if I keep the car I cannot $8400/yr for the next four years. Using a 10% discount, that income stream is $29,290. Or if 96k miles per year for 5 years, the present value is $28,021. So holding onto this car is a pretty nice investment with about a 10% ROI. I'm thinking that a uses car buyer would need to offer me at least $28k for this car, which is not bad considering that I just paid $30k for it a year ago. Does this sound too good to be true? Perhaps, but once we have a mechanism for monetizing a used car, then a proper way to value it is using a discounted cash flow model on that income stream. This, of course, is exactly what homeowners do when selling a home. The value of the home as a rental property sets a minimum price they are willing to sell the home for. If you can't find a buyer, you try to rent it out.
So renting your EV out on some TaaS provides an option to the owner that sets a minimum resale value. And that option disrupts the distinction these analysts make between IO EVs and TaaS EVs. In actuality the IO owner has the advantage, because they hold three options: rent, sell, or use personally. But the commercial TaaS operator really only has the options to rent or sell. This added option of personal use has real value. It explains for example why many vacation homes are owned privately. A family may fall in love with a cabin in the mountains, but renting it out all but 2 weeks of the year is just about enough to make a small profit. If a commercial vacation rental company had bought the same cabin, they would not have offered as much as the family that bought it for two weeks of personal use. The commercial investor would have needed to pay less for it so that it would payoff as a solid investment.
This is one basic problem that commercial TaaS operators may get into. Buyers for personal use will be willing to pay more for the same car than commercial operator would. But private owners will compete with them in offering cars for rent in TaaS platforms. This supports the resale value of cars for the owners, but undermines the profit opportunity for commercial TaaS operators. Imagine a platform where cars can bid on riders. A rider calls up a car to go from A to B. Rental cars near A or wanting to get close to B bid a fare amount and pick up time for the rider. And the rider selects the car they want to use. Such a competitive market is just fine for the private owner. They have a surplus of lifetime miles, so they don't care much whether the fare covers depreciation, they just want to net a small operating profit. So they are willing to bid fairly low. Meanwhile, the commercial operator needs to get enough per mile and enough miles not just to cover operating expenses, but to cover depreciation and financing plus profit. If the commercial operator has all in cost at 16c/mile, there are three things that can go wrong: not enough riders where the car currently is, may need to travel miles to next rider or for services, or private owners may bid less than 16c/mile. Indeed, if the TaaS operator finds that the car must travel 1 unbilled mile for every 2 miles with fare, then the minimum average fare needs to be 24c/mile. Through out the day, the vehicle must optimize between staying put and waiting for a rider or traveling to other locations to pick up riders. This is a tradeoff between low utilization or unbilled miles. So 40% utilization with 100% billable miles seems pretty steep to me. And worse yet competition from privately owned vehicles can maker the whole market unprofitable.
Curiously, this is the same sort of problem that central power generators face with competing with distributed power. If homeowners and business are generating solar power and storing in batteries, they first of all have the option of self-consumption. What surplus gets pushed out into the power market does not need to cover the fixed costs of these distributed resources. This can drive the market prices to such low levels that centralized power generators fail to get enough utilization and revenue to pay back the cost of the power plant. So, yes, coal and nuclear power plants are losing profitable utilization even to utility scale renewables. So while commercial TaaS may look like a great way to cut the cost of travel, a truly competitive market place may not allow these commercial concerns be profitable. Just like vacation homes and distributed energy, the option of self-consumption may be a critical advantage over commercial operators. And this is also the dirty secret behind Uber and Lyft: these platforms are largely exploiting private ownership of autos. And that may not change as autonomy and EVs are added to the mix.
