Tesla Q2 2021 – Unreality gap.

Fundamentals and valuation remain divorced.  

• Tesla reported excellent results that begin to establish its long-term credentials as a player in this market but the notion that it is worth more than most of the rest of the industry put together remains as absurd today as it was 12 months ago.
• Q2 2021 revenues / EPS were $11.96bn / $1.02 both nicely ahead of consensus of $11.4bn / $0.98 as the company was able to deliver 151% more vehicles than Q2 2020 and it did so profitably.
• There was good news on margins also with gross margins at 28.4% against consensus of 26.1% and EBIT margins of 11% compared to consensus at 6.6%.
• Furthermore, it has done an excellent job of building its brand with virtually no marketing (outside of its retail stores) and users now aspire to own a Tesla over the EVs of other companies.
• This will stand it in good stead when the rest of the industry catches up with Tesla and it has to compete head-to-head with the likes of BMW, Daimler, and Audi who know a thing or two about making luxury vehicles.
• This is all good news because one of my biggest long-term concerns around Tesla has been its ability to ramp up volume, execute on manufacturing and make money at the same time.
• This concern has been pretty resoundingly put to bed and Tesla as an independent automaker is here to stay for the long term.
• While the fundamentals of this company are steadily improving, the outlandish nature of its valuation relative to its peers is not.
• Tesla’s market capitalisation is currently $633bn which is greater than a large portion of the rest of the automotive industry combined and yet it sells only a tiny fraction of the number of vehicles.
• To justify this valuation, Tesla needs to sell at least 40% of all vehicles globally which even with the strong aspirational nature of its brand is incredibly unlikely.
• This means that the rest of its strategies, of which robotaxis is by far the largest, have to plug the gap.
• This is how its proponents think that its share price is going to go to $3,000 or more.
• Unfortunately, Tesla’s robotaxi strategy suffers from two large problems which I think that the market continues to ignore.
  • First, AI: Tesla relies solely on cameras and deep learning to interpret the environment around the vehicle upon which the driving decisions are made.
  • Tesla’s philosophical approach to AI is that if you crunch enough data, eventually the right answer will pop out at the end.
  • This is the Open AI philosophy which I think of as a brute force approach that has demonstrably failed to improve AI’s ability to think outside of the box.
  • This means that deep learning remains constrained to tasks where the environment is both finite and stable and the road is none of these things.
  • This is why one sees a constant stream of videos on Twitter or YouTube showing Tesla vehicles making ludicrous mistakes that a human would never make.
  • RFM research indicates that this is not how one will be first to put autonomous driving on the road, meaning as using radar, lidar and other sensors greatly reduces the error rate as Mobileye has demonstrated time and again.
  • Hence, it is the hybrid sensor approach that is likely to be first to market which I think leaves Tesla coming to market after everyone else with robotaxis.
  • Second, business model: which in Tesla’s case is deeply flawed in my opinion.
  • Tesla thinks that the cost to deliver robotaxis will be about $0.30 per mile which is a figure that RFM research broadly corroborates.
  • However, it also thinks that it will be able to charge $1 per mile giving it gross margins of around 70%.
  • This is what many of the bulls of Tesla’s shares are relying on to support the current valuation.
  • The problem here is that by all indications, there will be many robotaxi offerings on the market (many of whom will be better than Tesla) leading to a cut-throat brutally competitive market.
  • Hence the price is unlikely to stay at $1 but instead, be quickly eroded to something like $0.4 to $0.5 per mile.
  • This means that scale, performance, and efficiency will drive margins making Tesla’s forecasts unrealistic.
• Both of these flaws add up to a robotaxi strategy and valuation that simply do not hold water in terms of supporting the valuation where it is today, let alone any higher.
• Tesla is a narrative-driven stock where no one cares about the fundamentals but eventually, the narrative will run out of road.
• I have no idea when this will happen and so I would not be taking any shorts against Tesla which are expensive to run and depend on an event occurring within a specified period of time.
• At the same time, I have no desire to get involved and would look instead to some of the companies that are involved in reducing the cost to deploy autonomy in vehicles whose business models in some cases are at least grounded in reality.