Though originally written in 2020, this article still rings true today.
Tesla’s market capitalization just crossed $300 billion. It’s the largest car maker in the world, even larger than Toyota, which produced almost nine million cars in 2019 and had a market capitalization of $200 billion. Tesla’s market cap implies that the market believes that its production will go up more than 20-fold from the 400,000 cars a year it produces today to ten million cars.
As the market valuation of Tesla raced to the moon, its debt rating remained as junk (Toyota is A+, GM is BBB). The largest automaker in the world is junk-rated.
In the past I have made the analogy that the transition from internal combustion engine (ICE) cars to electric motors is akin to the transition from dumbphones to smartphones. It’s a domain shift. So maybe another domain shift will bring higher margins to Tesla, as happened for Apple with the iPhone. Unlike other car makers, Tesla is vertically integrated: It manufactures most of the components that go into its cars (including seats); thus gains from the economies of scale that used to accrue to its suppliers will accrue to Tesla.
Also, software plays a bigger role in a Tesla than in a traditional car. There is self-driving, over-the-air updates, and an iPad-like interface that powers all the controls, for starters. So if advanced software helps Tesla get higher margins than traditional car companies, it may not have to make as many cars to get to Toyota’s profitability. Bulls would even argue that self-driving alone may send Tesla’s margins to the moon. I’d like to pour cold water on that argument: Full autonomous driving is a good decade away. (I discuss it in great detail here, in my 37-page Tesla writeup).
Most importantly, going from 400,000 cars to many millions a year is not an easy nor a cheap journey. The market confuses Tesla with other Silicon Valley tech companies. Yes, Tesla is much more a technology company than your typical ICE car company is. It creates its own software and even the microprocessor that powers self-driving, but it still cannot escape the reality that it has to bend a lot of metal to produce its electric cars.
Unlike Facebook, which a decade ago could increase its user base ten- or- twenty-fold by spending a few hundred million dollars on data centers, Tesla will require an incredible amount of capital to increase production many-fold . To produce fewer than half a million cars, as it does today, Tesla needed a $25 billion investment in property, plants, and equipment. This is where bits meet atoms and face financial gravity. Tesla is barely breaking even today and will need to raise and invest not just tens of billions but hundreds of billions of dollars to increase its production enough to grow into its current valuation.
And then there is an element of time. Tesla has been stuck at producing 90,000 cars for the last eight quarters. It can only blame the coronavirus for a quarter or two. Getting to an annual production of even a few million cars will require time – a lot of time. A lot of dirt has to be moved, permits issued, equipment installed, people hired.
In Star Trek there are convenient wormholes, which cut corners through space, getting you to that galaxy a billion light years away in hours. Low interest rates have messed with the temporal properties of the market and created a wormhole in time and in Tesla’s stock (as well as in many other stocks – I talked about them here). It will take years, maybe even a decade, for Tesla to produce enough cars to justify its valuation. Today’s market valuation assumes it has already happened – that the capital has been raised and spent and that it cost nothing.
A few additional thoughts on ICE makers.
What is interesting to me is that today the market is basically valuing ICE car makers as melting ice cubes (pardon the pun). It tacitly assumes that they won’t be able to transition to electric vehicles, and so it values only their ICE cash flows, giving them basically no terminal value for their businesses.
Transitioning from one domain to another is incredibly difficult – your assets turn into liabilities. Even your knowledge in the old domain is often a liability in the new one. Just imagine being the best horse carriage maker in 1910 and making cars at the same time. Your horse carriages still provide huge cash flows. You still have to come up with new, better horse carriages and market and advertise them. But you know that at some point in time that business will be worth zero. This is the feeling I got when Chevy introduced its new Corvette. This is yet another topic I explore in great depth in my 37-page writeup.
One route that might save the ICE industry is the standardization of the hardware that goes into cars. Think of smartphones. Most of the parts that go into them are standardized. The memory that goes into iPhones and Android phones is the same; so are the batteries, sensors, microprocessors. (Apple develops its own, but most Android phones are powered by microprocessors designed by Qualcomm or Samsung). Software, too: All Android phones, which are 80% of all smartphones globally, run software created by Google. This standardization of components brings costs down substantially. Imagine if every smartphone manufacturer had its own version of everything that goes into a smartphone.
Now think about ICE car companies today. The only parts that are standard among cars are the tires and batteries (thank god!). All other parts – engines, transmissions, running gear – are basically custom to each manufacturer and each brand. (Toyota parts may or may not work in a Lexus.)
The best way to drive down costs of EVs is to borrow the standardization approach from the smartphone industry. That’s the right answer for the industry – EV costs would decline tremendously – but I’m not sure car companies can bend their thinking that far without plunging into an identity crisis. If car companies don’t design the parts or even the software that go into their EVs, then what are they? Marketers? Assemblers? Designers?
Tesla doesn’t have this identity crisis to grapple with. Nor does it have to make horse carriages while it is working on new cars.
There are several significant differences between phones and cars, thus requiring a tweak to my dumbphone/smartphone analogy.
First, cars cost a lot more than phones. A few-hundred-dollar difference in dumbphone vs. smartphone does not impact consumer behavior that much. Thousands of dollars do. The functionality of EVs and ICE cars is not much different – they both get you from point A to point B. Thus EVs (and especially their batteries) have to come down in price to be a truly attractive alternative to ICE cars. Lower maintenance and fuel savings are good carrots, but they come in the future years of ownership, while paying a higher price is up-front.
Second, because functionality is so similar, most consumers driving ICE cars do not know what they are missing.
Which brings me to the final point: The replacement cycle for phones is a year or two, but for cars it is about twelve years. Thus the transition to EVs will be gradual, giving ICE carmakers time to adjust.
When I wrote my Tesla analysis I opened it with this quote from F. Scott Fitzgerald: “The test of a first-rate intelligence is the ability to hold two opposed ideas in mind at the same time and still retain the ability to function.” You can see that analysis of Tesla and the automotive industry today requires holding a lot of opposing ideas, including that Tesla’s stock price depends on Elon’s ability to maintain a temporal wormhole.
Disclosure: I own a Tesla Model 3 and love it; I have a small position in Tesla puts in case the temporal wormhole collapses.