Tesla estimates that $125 - $150B of remaining CapEx is required to get to 20M veh/yr and 1 TWh of scale. Why isn’t this getting more attention?
View attachment 913626
Zach said “Maybe this total investment looks large. I actually think it’s quite small relative to our ambitions”. That’s an extremely modest way of saying it. The ROIC on these investments will be crazy.
At 20M cars and 1 TWh per year, Tesla could easily earn this much
per year, and probably much more. That is without Solar ever providing significant earnings and without autonomous driving, Optimus or any other AI stuff. Conservatively, something like $7k gross profit car and $40/kWh gross profit for batteries would yield approximately $125-150B net income post tax. Tesla could probably do double or maybe even triple these numbers.
That’s means the entire investment will be recouped with the cash flow from a year or less of full-scale production. Actually, in reality it’ll get quickly recouped while the growth is happening because of the ridiculous IRRs on factory construction Tesla will be getting, so on a net basis most of the cumulative CapEx spend will have already paid for itself by the time full-blown 20M & TWh scale is reached. That’s why Zach was saying this will all comfortably be funded with free cash flow along the way.
I’ve
written about Tesla’s amazing return on investment on factory CapEx before but the Gen 3 platform might do even better than these numbers. I’m getting IRR estimates of like 100% per year. Not a typo, 100% internal rate of return.
For about 17 years, Tesla was a cash flow negative company because their investments in growth were exceeding their operating cash flow. Then in 2021 it started to flip sharply into the opposite situation, when the arrival of Giga Shanghai and the Model Y vaulted Tesla’s cash generation to a new level at the same time as their CapEx per unit of production capacity was being slashed aggressively by like 50% or more. Now Tesla’s doing it again with Gen 3 and GigaMex.
Let’s do some math. They have 2M veh capacity right now. Even if we’re as pessimistic as possible and assume all of that $150B goes into cars instead of batteries, then that’s $150B / (20-2)M = $8.3k / car per year of capacity. Now, Tesla is going to be earning roughly this much gross profit per car. Something like $30k avg price and $22k avg cost. Depending on how fast we assume the factory is built and ramped, this yields an IRR of around 50-80%. If the actual investment for car factories is maybe $100B of that overall $125-150B estimate, then that’s $5.3k per car per year of capacity and the IRR goes to approximately 100%.
Now guess what, economically it doesn’t make any sense to get 100% ROI and just stop there. What that means is you should keep investing in MORE growth until diminishing returns kick in and the ROI on each additional project drops to closer to the discount rate (10% or whatever you want to use, but nobody is using a 100% discount rate in the real world). It would make no sense with respect to Tesla’s finances nor mission to stop growth at the stated target levels, *if* the margins can be maintained and *if* these CapEx projections are even remotely accurate. What does make sense in the long run is to drop prices even lower, expand volume massively to more than compensate for the thinner margins, and basically take over a huge swath of the market. Tesla with Gen 3 will be able to offer great value for prices so low that customers just can’t resist and competitors are incapable of matching. The numbers presented by Tesla suggest that growth will continue far beyond the 20M and 1 TWh level.
