Two new quarters have gone by since my last analysis of Tesla’s (TSLA) Selling, General and Administrative (SG&A) expenses. The release of the Q2 numbers is now about a month past. In the meantime, not much has happened besides some lawsuits, pedophile accusations, negative reports, positive reports, 86% doubts about Model 3 quality, and the occasional floating and subsequent sinking of the idea to take Tesla private. So really no excuse for this delay, my apologies, I know you have all been awaiting eagerly for an update on the SG&A situation. The first four graphs will simply be updates of the ones I used in my previous analysis. After that I will show two scenarios varying the gross margin for Model 3.

My view was and is that SG&A is one the most important items determining Tesla’s success or failure, while at the same time being the most neglected. (UncleBrian Research is one of the few that did pay attention to it in a recent article.) Up to at least 2017 Q4, SG&A always went up nicely with revenue, thereby assuring no profit would ever be made by Tesla. That is, unless Tesla manages to break that trend and, on a per-car-basis, bring down those costs. Now that the Model 3 is being delivered in sufficient numbers to make an impact, let’s see how things are developing.

I start with my traditional first graph, an overview of SG&A per total revenue:

Nothing dramatic has happened, but there is some improvement on the overall SG&A front. Let’s see how this compares to overall gross margin:

We can see that after deducting SG&A/revenue from gross margin we still end up below zero. In other words, Tesla is still making a loss, even before deducting Research and Development (R&D) and interest expense. On the other hand, glass half full, while SG&A per revenue goes down, gross margin goes up, leading to a nice up-tick in Q2 for the difference between the two.

I continue with SG&A per delivered car. Here we can also see an improvement:

Note that the above three graphs were based on the company as a whole. Non-automotive revenue is about 16% of total revenue, so this may influence results, especially when extrapolating going forward into the future and beyond.

To estimate “automotive SG&A”, i.e., SG&A that has only to do with the automotive part, as opposed to, e.g., solar cells, I have (as before) used the following steps:

- Start with total SG&A.
- Subtract the SolarCity part. For 2016 Q4 and 2017 Q1, this was explicitly mentioned in the financial reports. For the rest of the 2017 and 2018 quarters, I have made estimates per quarter, using the revenue of SolarCity and assigning a similar percentage to SG&A as in Q1. Note that the further out from 2017 Q1, the more uncertain this becomes.
- For all other non-automobile revenue, I have assigned a percentage to SG&A and subtracted that from the total SG&A. I don’t know what the right percentage is, so I have used varying assumptions, ranging from 20% to 70% for SG&A per non-automotive revenue.

This has led to the following estimate for Tesla’s automotive SG&A per car, assuming non-automotive SG&A to be 20% (similar to overall SG&A vs. revenue) of non-automotive revenue:

The change is less pronounced than when looking at overall SG&A, but there is still an improvement, especially in Q2 of 2018.

### Profitability based on 2018 Q2 SG&A numbers

I now do two profitability estimates using the range of 20% to 70% as per point 3 above. Now that delivering more cars has actually led to lower SG&A on a per car basis, I figure an extrapolation to a total of 350,000 cars annually based on regression makes sense, using the numbers from 2016 Q3 (first time over 20K cars) to 2018 Q2. The two estimates differ only in the assumed gross margin on the Model 3: 25% (as per Musk) vs. 15% (more conservative).

My assumptions are:

- Model S and X: 100,000 delivered per year (capped by Musk at that level) at an average price of $100,000, with a gross margin of 25%. Result: 100,000 x $100,000 x 0.25 = gross profit of $2.5 billion per year.
- R&D: In the first half of 2018, this was about $750 million, so for a full year that is $1.5 billion.
- Interest: The net amount was about $160 million in 2018 Q2. Times 4, round it down: $600 million for a whole year.
- Other income and restructuring expense will be ignored.
- Other lines of business will be ignored.

Add this all up, we get $2.5 billion – ($1.5 billion + $600 million) = $400 million.

- Model 3: I assume a production rate of 5,000 per week, totaling about 250,000 per year. I will vary the average selling price from $35,000 to $60,000. A gross margin of 25% gives a range of gross profit for the Model 3 of about $2.2 billion to $3.8 billion per year, whereas a gross margin of only 15% gives a range of gross profit for the Model 3 of about $1.7 billion to $2.7 billion per year.
- SG&A: I use the range of 20% to 70% as per point 3 further above. Now that delivering more cars has actually led to lower SG&A on a per car basis, I figure an extrapolation to a total of 350,000 cars annually based on regression makes sense, using the numbers from 2016 Q3 (first time over 20K cars) to 2018 Q2. This gives a range of ca. $2.0 billion to $3.8 billion per year, where the lower number corresponds to assuming a higher (70% of revenue) SG&A for non-automotive segments, and the higher number corresponds to 20%. Note that range is for all models combined, e.g., S, X and 3 together.

Taken all together gives me a rough estimate of the profitability of Tesla’s automotive business in a year in which the above-mentioned assumptions would be reality.

Assuming a gross margin of 25% leads to:

We can see that if at some point Musk’s expectations regarding gross margins and number of delivered cars come true, and if Tesla manages to bring down SG&A costs as it has towards 2018 Q2, then there is a chance at a modest profit even at lower average selling prices. At the higher prices, profit is “secured”, but the question remains whether there is enough demand at such prices. Another question is whether the profits would be enough to justify the current stock price. The above analysis suggests they wouldn’t be.

Assuming a more conservative gross margin of 15% leads to:

We can see that under this assumption, it is very unlikely to see any profitability, because again, an ASP of $50,000 or more seems doubtful for the number of cars assumed to be sold.

### Conclusion:

Tesla has managed to improve SG&A on a per car basis in the first half of 2018. Whether this is sufficient for profitability in the future depends on whether Tesla manages to achieve its aimed for gross margin on Model 3 of 25% and on the exact internal cost structure of Tesla’s SG&A. Even with profitability, the share price seems higher than warranted. Without a gross margin on the Model 3 of 25%, future profitability seems very difficult.

**Disclosure:** I am/we are short TSLA.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

**Additional disclosure:** I am short TSLA via long-dated, out-of-the-money puts.