Quick Thread on Robotaxi Market Potential

As many of you know, I'm working on my sum-of-the-parts valuation of Tesla, where I value each separate business unit individually. As I have been thinking about the market potential for robotaxis, I wondered how far this could penetrate

Elon has indicated that the price consumers would pay would be over $1 per mile. Given Uber and Lyft pricing which is closer to $2, that seems quite reasonable to me. However, when we start thinking in the longer term, in a future where Tesla is selling 10M+ vehicles...

each year, I thought that price point would likely be too high to truly disrupt transportation in a such a way that personal car ownership becomes a thing of the past. After all, the IRS mileage reimbursement rate (which is based on average ownership costs) is only $0.58 per mile

To have truly disruptive potential, where nobody would opt to own a car anymore, the cost to the robotaxi rider would need to be lower than $0.58 per mile. To find out if true widespread adoption could occur, I tried setting the price per mile at $0.55, and working backwards.

Since the owner will be bearing the majority of the costs, Tesla's share would need to be quite small - only $0.10 per mile in my assumption. That leaves the owner with $0.45 from which to recoup all her costs, and keep in mind that not all miles driven will be billable.

That's not much to work with, so I started thinking "hmmm, how else could you minimize variable costs?" and then I realized that charging your robotaxi with solar is free on a variable cost basis. So I set out to find out how large of a solar array an owner would need.

Assuming modest improvements in efficiency (going from ~4.25 miles / kWh to 4.5 miles), and driving 90,000 miles per year, an operator would need 20,000 kWh annually to power their robotaxi. Using a 20% capacity factor for solar (would be higher in south, lower in north)...

This indicates a solar array size of 11.4 kW would be sufficient to power your robotaxi. You'd also want to buy a powerwall or two so that you can capture that energy during the day while your robotaxi is out making money. At Tesla's current pricing, this total cost is $17K.

I added these costs to the actual robotaxi purchase cost, which as you can see assumes modestly declining vehicle costs due to manufacturing improvements, and rapidly increasing FSD purchase costs as autonomy provides more value. We now have a total upfront cost of $79K.

However, no we have a $0 marginal cost to run our robotaxi. So the only costs left for the owner to incur are maintenance and overhead. While difficult to project accurately, we have good data from Tesloop about historical maintenance costs which have come in under 5 cents/mile.

If we nonetheless round up to $0.05 per mile, we get maintenance costs of $4,500 per year. I'm estimating an extra $7,500 per vehicle in overhead costs and taxes, although these would largely be a factor of how large the operator's fleet ends up being.

Putting it all together, the Operator ends up with a pretty nice and profitable business, even at the meager rate of $0.55 per mile, which would be highly attractive to the masses.

At a 6.5% discount rate, this equates to a $38K NPV per vehicle, indicating that it would also be highly attractive to robotaxi operators.

And perhaps most excitingly, for Tesla the math is very exciting. They not only get $0.10 per mile, but also the upfront FSD costs. Not to mention the solar and battery purchase revenues, which I excluded here for simplicity. This indicates a present value of $45K per vehicle.

If Tesla is selling 10-20M of these things annually, the cash flow and share price implications are simply staggering. It's hard to believe so much value can be created at such a low price, but that's what disruption looks like.

So we have here a proposition that makes sense for masses of riders, makes sense for operators, and that makes sense for Tesla. This is a good thing to keep in mind as the share price continues its correction. @threadreaderapp unroll.