It’s all about margins. Gross margins are central to the Tesla investment case because the metric best gauges whether the company is on track to be valued as a tech company or just another automaker. Tech companies have higher valuations because they have higher margins. Tech margins are higher because of the ability to use hardware to sell higher-margin software and services.

While I had previewed a miss on gross margin in September, the magnitude of the miss was greater than I had expected. Even worse, the potential for margin improvement over the next year was essentially taken off the table, given the cost of ramping Cybertruck production along slowing demand for vehicles which reduces manufacturing efficiencies.

For the quarter, auto gross margins ex credits fell to 16.3%, lower than the Street consensus at 17.6% and down from 18.1% in June (and 29% from a year and half ago). While this drop is concerning, Tesla still holds a slight advantage over other carmakers with a 10-14% gross margin.

The bottom line when it comes to margins is that investors will likely have to wait a year before they start to improve. And the slope of improvement will be gradual. Once we get through the ramp of Cybertruck—which takes us into 2025—the investment phase for Robotaxi will begin. Putting it together, it may not be until 2030 that we see margins return to the 29% peak. The biggest unknown on the margin-expansion path is when contribution from high-margin FSD revenue steps up.

The sigh of relief when Tesla reaffirmed its 2023 production goal of 1.8M was followed by concern when Musk hinted that his previous comments about 50% compound growth through the end of the decade were too optimistic.

My read on the topic came from a question on the earnings call: “Current sell-side consensus assumes that Tesla will deliver 2.3 million vehicles in 2024, representing 28% growth versus 2023 guidance. Is this growth rate achievable without any mass market launches in 2024? And when does Tesla expect to return to its 50% long-term CAGR?”

Musk’s response: “Yes. I mean at the risk of stating the obvious, it’s not possible to have a compound growth rate of 50% forever or you will exceed the mass of the known universe. But I think we will grow very rapidly, much faster than any other car company on earth by far.”

Small adjustments to compound growth expectations have a significant impact on end-of-year deliveries. If Tesla were to grow at a compound rate of 50% from 2025-2030 it would exit the decade with 17M annual deliveries.  It’s worth noting the long-standing 20M deliveries target in 2030 was dependent on 50% compound growth in 2023-2030. It’s likely that 2024 will grow by about 25%, which essentially takes the 20M goal off the table. If Tesla were to grow at a compound rate of 30% from 2025-2030 it would exit the decade with 8.5M annual deliveries. I think a 30-40% compound growth through the decade is more realistic.

The first Cybertruck deliveries are scheduled for November 30th, and for the first time in three years the company updated preorders, which now are about 1M. Even after factoring in that thought that between 25-50% of preorders will likely not materialize, a backlog of 500-750k orders is impressive. As a point of context, the best-selling truck in the US is the Ford F Series at about 500k deliveries a year. Given Cybertruck will take orders away from the F Series, I believe Cybertruck will be the best-selling truck in the US in 2025 or 2026. That title may be short-lived given it will be driven by years of pent-up demand that, once filled, will create an air pocket of demand in 2027.

Elon believes the Cybertruck will be Tesla’s best product ever. I agree. It’s radically different than any other car or truck on the planet. This means production comes with new problems no other car manufacturer has seen before.  Elon stressed: “I do want to emphasize that there will be enormous challenges in reaching volume production with the Cybertruck and then in making Cybertruck cash flow positive.” With over 1M people on the waitlist, the ultimate goal is producing more efficiently and profitably. As touched on earlier, affordability is a major focus for Tesla. The Cybertruck will take 12-18 months to get to significant positive cash flow, with a goal of reaching a 250k/year run rate by the end 2025 (they will sell less than 250k in 2025 and more in 2026).

In September, Tesla lowered the price of FSD in the US by $3k, from $15k to $12k. Sales of the software have slowed as consumers want to see more substance in the product before buying. That likely impacts the cumulative FSD miles driven in the quarter, ending at about 520M, up from 300M in June and 150M in March. I was expecting the number to be above 600M in September. The slight miss is noteworthy, given every mile driven is key to model training to unlock FSD’s full potential.

Setting aside miles driven and product milestones, Musk was steadfast in his belief that a polished FSD would see the light of day in part because of his assessment that Tesla has “the best real-world AI team in the world”. I agree. We will see a ChatGPT moment when it comes to FSD, when the product quickly moves from a science product to a must-have mainstream feature. The question is will that happen in three years or a decade?

Despite all of the negativity around margins and unit growth outlook, Tesla is still in the pole position to capitalize on the shift to EVs which is a large addressable market. That auto market is massive, somewhere between 2.1-2.6T annually, assuming between 70-85M cars and light trucks are sold globally with an average selling price of $30k (the US is about $45k average today). As a point of perspective, the car market is about 4x bigger than the smartphone market.

Tesla has the upper hand when it comes to building a car company with an electric focus from the ground up. Today, traditional car companies are facing multiple headwinds when it comes to making the shift to electric vehicles. Ford and GM recently reduced expectations for EV deliveries over the next year, and GM delayed the opening of their light truck EV manufacturing plant, Orion, by more than a year (now expected to produce trucks late in 2025). Other carmakers must also navigate large CapEx spending to retool for electric, growing labor expenses, difficulty in integrating software and hardware, and a legacy dealer network that reduces profitability.

Ultimately, this leaves traditional automakers in a catch-22. We believe traditional carmakers have two options going forward:

• Option 1: Release a car with features and range at parity with a Tesla, and sell it at cost. This car will be priced 10-25% higher than a comparable Tesla, thereby softening demand and leading to further market share loss.
• Option 2: Subsidize vehicles in order to gain market share from Tesla. With a limited margin cushion, this will increase losses. The more they sell, the more money they lose.

Taking it to the logical end, we believe car companies that have been around for 50+ years will eventually (10 years from now) be forced to restructure or go out of business.

Disclaimer