Published on October 4th, 2018 | by Vijay Govindan0
Tesla Q3 Financials Estimates & My 99.6% Accurate Delivery Estimate
October 4th, 2018 by Vijay Govindan
In the last article I wrote on this topic (“Tesla Q3 Revenue, EPS, & Delivery Numbers: Estimates vs Reality“), I mentioned my overall delivery estimates along with my revenue and EPS for Q3 of 2018.
On October 2nd, Tesla released its actual delivery numbers. The company delivered 83,500 total vehicles: 14,470 Model S, 13,190 Model X, and 55,840 Model 3s in Q3 2018. This is a record number of deliveries for the company. These are great numbers — whether you are for Tesla or against Tesla, you have to admit that. Only the most biased media will spin this negatively. The numbers are so good it is causing challenges for major media sites.
To put this in perspective, in just Q3, we delivered more than 80% of the vehicles that we delivered in all of 2017, and we delivered about twice as many Model 3s as we did in all previous quarters combined.
— Tesla Q3 2018 Vehicle Production and Deliveries
I estimated Tesla would deliver 83,181 deliveries. I was decently close, 99.6% accurate, so I’ll give myself a goal on that. For Model S and X, my delivery numbers were 27,360, out of a total of 27,660. That is 98.9% accurate or a little more than 1% wrong. I guessed 55,821 Model 3s delivered out of 55,840 actually delivered. That’s 99.97% accurate. Mild-mannered economists have killed other economists for this kind of accuracy.
I mentioned Estimize is a great site that allows people to enter their Tesla total delivery numbers, revenue, and EPS every quarter. Let’s see how my delivery estimates compared with Estimize.
Due to feedback on my last article, Appendix A has more detail on how to read the chart.
The green line is the amount Tesla actually delivered. We want to look at the top of the chart, find FQ3 ’18, and then look down. We see Tesla delivered 83,500 Model S / X / 3 in the quarter. My purple estimate would have been 83,181 (not shown but imagine it’s there right above the yellow point). The blue Estimize Consensus is 77,856, and the yellow point is the prediction from an analyst I follow called minor_threat — 79,000.
According to Estimize, the Estimize Consensus is “an intelligently weighted consensus consisting of all estimates excluding flagged estimates that takes into account estimate recency, analyst confidence, and more.”
The Estimize Mean was 67,654 vehicles delivered. 112 people had ranges for vehicles delivered from 26,147 (completely laughable) to 93,831 (extremely aggressive). The Estimize Mean is adding up everyone’s estimates, and then dividing by the number of total estimates.
What does it all mean? Three people hit the bullseye, 83,500 deliveries. Four more people estimated above 83,500 deliveries. 104 people were below my guess of 83,181 deliveries. About 92.9% of guesses were wrong and underestimated Tesla. Let me repeat: 92.9% of people underestimated how many vehicles Tesla would deliver in the third quarter.
Based on the new delivery information, I made some slight revisions to my version of “Vijay’s SimTesla Game” for Q3. I am estimating 17% gross margin for the automotive division (2% above Tesla’s goal), 15% gross margin for the Energy storage business, and 13% for the Energy generation business. Services slightly improves to -41.6% margin.
This leads to a profit of 12 cents per share.
Here are updated charts showing my EPS and Revenue changes:
What Did Wall Street Estimate?
I have read conflicting reports. The best analysis I have comes from InsideEVs. Comments in italics are my own.
RBC, Joseph Spak
Model 3 production estimate: 52,000; deliveries: 55,400 (below on production and deliveries)
Goldman Sachs, David Tamberrino
Model 3 production estimate: 50,000; deliveries: 52,000 (below on production and deliveries)
Cowen, Jeffrey Osborne
Model 3 deliveries: 45,000 (below on deliveries — this is two penalties, the estimate is so poor)
Vertical Group, Gordon Johnson
Model 3 production: 52,582 (below on production)
Consumer Edge, James Albertine
Model 3 production: 50,000; deliveries: 55,000 (below on production and deliveries)
Model 3 production: 47,500; deliveries: 52,000 (below on production and deliveries — this is is either one penalty shot or two, depending on whether you give one each for the missed production and delivery estimates)
I can take another side gig if any of the above companies are interested.
This was reported in a MarketWatch piece:
“Analysts polled by FactSet had expected Tesla to deliver about 80,000 vehicles in the quarter, including 55,600 Model 3 sedans. Model S and Model X deliveries were expected to reach 12,900 and 12,500, according to FactSet.”
Adding up the polled numbers gives me 81,000.
TheStreet was more negative but revised its piece a bit later in the day:
“Tesla produced 53,239 Model 3s in the third quarter, a substantial gain from 28,578 Model 3 output in the previous quarter and within company forecasts of 50,000 to 55,000. Model 3 production hit 5,300 in the last week of the quarter. Deliveries of 55,840 Model 3s were slightly below expectations of 56,000.”
I am not sure how they came to expectations of 56,000. That doesn’t jive with the analyst numbers I quoted above.
Almost every way you look, Tesla beat delivery estimates. This is remarkable, because Tesla kept a big promise. Media sites are quick to slam Tesla for missing any goal, but it produced between 50,000 and 55,000 Model 3s and delivered above 55,000, just as it said in the Q2 Shareholder Letter it would do. This gives us confidence Tesla will meet its Q4 numbers after announcing them next month.
Quarter 4 Numbers
I think it is too early to have an idea of Q4 numbers. I am positive Tesla will produce more vehicles in Q4 than it did in Q3, but it’s hard to predict more than that. Panasonic is increasing battery output. Tesla is getting battery machines from Germany that will increase speed by 3x while reducing costs by 3x. I don’t think it’s unreasonable for Tesla to produce 5,000 to 5,500 Model 3 a week. Along with higher Model 3 gross margins, Tesla will become sustainably profitable, no question.
- Across the top of the chart is the fiscal quarter and year.
- On the right hand side of the chart is the scale, in this case it’s millions in Revenue.
- If we look down the bar for the FQ3’18 we see some points and some boxes.
- The purple point is my estimate. (If you add your own estimate to Estimize, it will show up as purple.)
- The light blue Estimize Range box shows the range of estimates for all people submitting an estimate.
- The blue diamond is the Estimate Consensus. This is weighted based on how trustworthy Estimize finds your estimate.
- The gray Wall St. Consensus box shows the range of estimates for professional Wall St analysts.
Putting it all together, you can see how you compare with everyone else. In this case, I am looking for Revenue at the very top of the Estimize Range.
Appendix B – How to Calculate Average Selling Prices
(Ahoy, math ahead!)
Average Selling Prices are quite easy to calculate.
Average Selling Price (ASP) is simply:
Total Number of Units Sold
Let’s use the following information:
14,470 Model S, 13,190 Model X, and 55,840 Model 3s were sold in Quarter 3.
The price for each Model S is $99,322, for each Model X $106,632, and for each Model 3 $60,332.
Total Sales = 14,470 * $99,322 + 13,190 * $106,632 + 55,840 * $60,332
Total Sales = $6,212,604,300 or $6.2 billion.
Total Units = 14,470 + 13,190 + 55,840
Total Units = 83,500
Average Selling Price = $6,212,604,300/83,500
Average Selling Price = $74,402 for each 1 unit sold
The beauty of this method is you can go as fine or as broad as you want. Usually, it makes sense to group related items together to find their Average Selling Price. It wouldn’t make sense to combine Tesla’s Auto division and Tesla’s Energy division.
Appendix C – Profits vs. Operating Cashflow
Profits and Operating Cashflow are items analysts throw around a lot. For the normal person who hasn’t suffered through numerous finance classes in college, or in my case an almost 3 year long MBA that seemed twice as long, finance is intimidating.
Profits = Revenue (Sales) – Costs – Interest – Taxes
Not difficult at all. In Tesla’s case, we are almost at the point where Revenue (Sales) equals Cost + Interest + Taxes. In Q4, Tesla’s Revenue (money from sales) will be greater than Cost + Interest + Taxes.
What about Operating Cashflow (Cash flows from Operating Activities at the bottom of the Shareholder Letter)?
Operating Cashflow is more involved. Here it is at a really, really high level:
Operating Cashflow = Profits (from above) + Depreciation + Stock-based compensation + Change in items needed to run the business
Depreciation. If you have ever owned real estate or owned a car, you know about depreciation. Depreciation is a fancy way of saying something is worth less after you have used it for some period of time. Tesla has depreciation. This depreciation is related to its factories, plants, and other equipment it purchases. For the purposes of Operating Cashflow, Depreciation will add to Profits.
Stock-based compensation. If you get stock options, that’s an expense to the company. The company doesn’t have to pay you the value right away. That’s cash saved. If you issue stock options, you can add the value back to your Operating Cashflow.
Change in items needed to run the business. This relates to changes in Inventory and other items needed to keep the business running. Let’s say you have to hold onto a lot of finished items and raw materials. That’s cash you can use for other things that is tied up. If your Inventory increased, that will reduce your Operating Cashflow.
The above references an opinion and is for information purposes only. It is not intended to be investment advice. Seek a duly licensed professional for investment advice.
I am currently long Tesla shares. I do not plan to make any investment related to Tesla in the next 48 hours. If I did, they may be shares or options.