Near-future quarterly financial projections

[EVNow]

I know we've discussed this before but I still disagree.

The earlier discussion was around Q4. It turned out the production of 3 in Fremont was less than 7k/wk.

Now let’s talk about Q4 ‘20.

- Fremont is at full capacity
- GF3 is at 3k/wk (or 4K) for Model 3
- Total production of 3+Y is 160k/quarter.
- But the demand is still higher.

What should Tesla do now anticipating the above.
A. Accelerate Y production in GF3
B. Increase capacity in Fremont, may be temporarily, by expanding pant ship and stamping

Option A is wildly superior to B.
- It is anyway already planned to be done. No extra capex. Easy and cheap funding.
- It is likely faster than complicated changes to Fremont where space is tight

 

[mongo]

The earlier discussion was around Q4. It turned out the production of 3 in Fremont was less than 7k/wk.

Q4 2019 Fremont 3 output was only less than 7k (on average) if you assume more than 86 days of production out of the 92 possible (12.3 weeks out of 13). Or less than 6 days of shutdown. End rate very likely was over 7k per week regardless.

*Knocking off 1k for GF3

 

[dc_h]

Here is my latest model.

View attachment 495983

Tsla value stock, I can hardly imagine tslaq head spinning and hand waving.

 

[EVNow]

Q4 2019 Fremont 3 output was only less than 7k (on average) if you assume more than 86 days of production out of the 92 possible (12.3 weeks out of 13). Or less than 6 days of shutdown. End rate very likely was over 7k per week regardless.

*Knocking off 1k for GF3

Yes it was around 7k/wk. Earlier discussion was whether Tesla would increase production to significantly more than 7k/wk.

ps : See the discussion starting with this post - Near-future quarterly financial projections

 

[dc_h]

I've written about this before - my contention is Tesla is not interested in expanding Fremont capacity beyond 10k/wk for 3 + Y. I think that's the demand for NA. They will similarly have about 10k/wk for Europe and 10k/wk in GF3. This will be a total of 1.5M/yr capacity in 2022/23 compared to 350k/yr capacity now in Fremont for 3.

If they can get over 10k per week in Fremont they will. Marginal investments have been minor to increase capacity, so resulting margins over 7000 cars per week are likely over 30%. The mix would likely be higher Y vs Model 3 if they hit 12,0000 a week in Fremont.
key capacity limit is paint. Stamping should have capacity and assuming assembly continues to improve, they should increase capacity at least another 20% this year. A redesign of S and X would likely streamline production as well, using Model Y wiring concepts and other assembly component reductions.

 

[mongo]

Yes it was around 7k/wk. Earlier discussion was whether Tesla would increase production to significantly more than 7k/wk.

ps : See the discussion starting with this post - Near-future quarterly financial projections

Gotcha, thanks!
Based on @Artful Dodger 's analysis, Tesla has been increasing 3 production at a rate of 90 cars per day per quarter all year (or one car per day per day aka the Dodger Unit). Hard to say if that trend will continue or stop at some target level.
Tesla, TSLA & the Investment World: the 2019-2020 Investors' Roundtable
Tesla, TSLA & the Investment World: the 2019-2020 Investors' Roundtable

 

[Artful Dodger]

Gotcha, thanks!
Based on @Artful Dodger 's analysis, Tesla has been increasing 3 production at a rate of 90 cars per day per quarter all year (or one car per day per day aka the Dodger Unit). Hard to say if that trend will continue or stop at some target level.
Tesla, TSLA & the Investment World: the 2019-2020 Investors' Roundtable
Tesla, TSLA & the Investment World: the 2019-2020 Investors' Roundtable

Don't forget that 2020 is a Leap Year, so add another 1,040 Models 3 from Fremont and ~500 from Shanghai to the 2020Q1 total. There'd also be ~204 extra Models S/X produced if their production rate remains unchanged from 2019Q4.

Cheers!

 

[kengchang]

Wasn't there a report of electricity outage at the Factory for one day and workers were sent home? That's 1K units of Model 3 not being built

 

[Artful Dodger]

Here is my latest model.

View attachment 495983

Does you Model account for the extra day of production in 2020Q1 due to the Leap Year?

Cheers!

 

[dc_h]

Regarding demand, yes BMW sells about 2.5 million. But

• Tesla sells better cars at a lower TCO
• In some places like urban centers in China, it's not possible to get an ICE car
• BMW has equivalent competition in every segment they sell, unlike Tesla who has no competition at the moment
• Governments are pushing EVs with incentives, which is only going to increase as climate change awareness increases, though this may over time shift to penalties for ICE (e.g. Europe)
• BMW doesn't make trucks
• The Tesla Stretch

The biggest roadblocks are awareness, production capacity, and urban charging infrastructure for apartment dwellers. Of course Tesla needs to make more models: a larger SUV / minivan thingy with seating for 8, which can also double up as a work van.

Remember, the most valuable consumer product the iPhone, has only a few models. Between these products, Tesla can easily get to 6-7 million worldwide. A model 2 and a smaller Y can easily get it to double digits if priced under 30k.

Pivoting to valuation for a bit, Toyota makes 20B dollars in profit for 10 Million cars sold. Tesla will make more even at 5 million in sales. Let's say 12% at 40k ASP. Remember a lot of this value is software, where marginal cost is zero unlike traditional automaker. Also Tesla is more integrated both upstream and downstream. So 12% is really an underestimate. At 24 Billion in profit (12%*40k*5 Million), slap a 30 PE, that's a 720B in valuation. Assume 240 million shares at that point, which is 3K per share.

This assumes no energy and storage, no FSD, no Tesla insurance, no Tesla credit (which becomes a reality when the ratings and cost of funds improve), no premium service subscriptions. The biggest one I am excluding is the Tesla Semi, where the purchase decisions are solely economic in nature and a flip in economics will mean a sudden shift in buying decisions as long as infrastructure is there.

Bottom line, if Tesla is approaching 5 Million in sales, it should be trading at close to 3k, and that would be undervalued. I expect this to be met from 6 giga factories, each doing 15-20k per week. Expect to see a US Midwest GF, one near Beijing or in Japan, and another in Europe. I don't see a reason why any of this cannot happen before end of 2024.

I realize this is a tangent, but the BMW comparison forced me to respond.

Would love to see a Midwest plant, especially Illinois. Indiana is most pro-business, Illinois might not support without some UAW support. Wisconsin has a huge undone Foxxcon site not being used. Michigan, like Illinois needs the support but won’t support evil Elon. Indiana seems like the best bet and would be great for the cyberteuck.

 

[mongo]

Would love to see a Midwest plant, especially Illinois. Indiana is most pro-business, Illinois might not support without some UAW support. Wisconsin has a huge undone Foxxcon site not being used. Michigan, like Illinois needs the support but won’t support evil Elon. Indiana seems like the best bet and would be great for the cyberteuck.

Don't be so quick to write off Mordor. The Tesla v Michigan lawsuit has been delayed by the request of both parties.

 

[EVNow]

If they can get over 10k per week in Fremont they will. Marginal investments have been minor to increase capacity, so resulting margins over 7000 cars per week are likely over 30%. The mix would likely be higher Y vs Model 3 if they hit 12,0000 a week in Fremont.
key capacity limit is paint. Stamping should have capacity and assuming assembly continues to improve, they should increase capacity at least another 20% this year. A redesign of S and X would likely streamline production as well, using Model Y wiring concepts and other assembly component reductions.

The basic point is it is not possible to increase capacity with marginal investments after 10k/wk.

See this below post and subsequent discussions.
​

There's no question it was a 2 stage plan with incremental capex. But the 2nd stage was never a full duplicate line. Analysts specifically asked early on about a 2nd line/duplicate line and while the replies were vague it was clear that was not the plan.

The paint shop was explicitly announced as a 10k/week installation. You can search the original Eisenmann press release. The gigantic Schuler press was also spec'd for 10k/week, or at least to be easily expanded to 10k. The body shop was floor-planned for 10k, though welding lines are modular so they obviously didn't buy all the robots on day one. The seat factory was laid out for 10k, though once again they didn't buy all the equipment in stage one. General assembly, well..., who knows what the plan was there.

They're installing a new final assembly line (GA5) and new welding robots. Even that involves moving a bunch of stuff to other locations. I've heard zero about new stamping presses or paint shop. They specifically said building Y in Fremont would save a lot of capex, which wouldn't be true if they were putting in new stamping, paint, etc. Which they don't have room for, anyway.

That's why I say 3+Y will be 10k/week. We'll find out in January, I suppose.

 

[dc_h]

The basic point is it is not possible to increase capacity with marginal investments after 10k/wk.

See this below post and subsequent discussions.
​

Understood. The cost to increase production in Fremont is probably more than Shanghai, but still the return on investment will still be too significant to pass on. It’s probably less then 100 million to increase production 10,000 cars a quarter, which would pay off in one quarter. They have a lot of competing investments with quick payback, so you may be right, but it seems hard to pass up increasing Fremont as much as possible.

 

[PhaseWhite]

Thanks all for your contributions in this thread. Looking through some of your models I didn't see a one time item for the 2k Acceleration Boost in Q4 2019.

If 20k people purchased it that would contribute $40M in 100% margin revenue. The big questions are: How many eligible AWD owners are there and how many would pay up?

 

[EVNow]

It’s probably less then 100 million to increase production 10,000 cars a quarter

Do you have something to backup this claim and why it would easier/faster/more beneficial to do this in Fremont than in GF3 ?

 

[ammulder]

Wasn't there a report of electricity outage at the Factory for one day and workers were sent home? That's 1K units of Model 3 not being built

Sheesh, someone tell this company that people market large battery systems to provide power in case of outage...

 

[dc_h]

Do you have something to backup this claim and why it would easier/faster/more beneficial to do this in Fremont than in GF3 ?

I don’t have costs, but Elon has said incremental capacity increases were mostly in the 10’s of millions, vs starting the 3, which was billions.
Regarding incremental increases at Fremont vs capacity additions in Shanghai, I don’t see any mutual exclusion. If Fremont can increase capacity another 50% for less then a billion, it shouldn’t reduce any investment in Shanghai or anywhere else. If they were sitting on 2 billion in the bank, I’d agree, but getting combined 3&Y production over 10,000 will be immediately profitable. Shanghai once running will increase capacity quickly as well. More then adding another line in Shanghai I want to know if they have enough local battery and extended supply chain, as well as logistics and charging to sell 5000 or 10,000 cars a week in China? I think they can get there faster then Fremont, but they have a lot more work. Totally agree it’s a bigger growth opportunity in Shanghai, no disagreement at all, you are 100% correct.

 

[mmd]

That sounds like a weird assumption. How would you explain over 5k registrations in November at a much higher asp vs MIC model 3s. That's more demand than 1k a week.

The previous 4 months? July to Nov is 2700 monthly average. ASP was only 3% higher if I remember.

Regarding your request for China Tesla registrations for the earlier 10 months, I found this graph below. The absolute numbers are not posted; you'll have to estimate the numbers. November may be high because October was very low...just a guess.

 

[kbM3]

Thanks all for your contributions in this thread. Looking through some of your models I didn't see a one time item for the 2k Acceleration Boost in Q4 2019.

If 20k people purchased it that would contribute $40M in 100% margin revenue. The big questions are: How many eligible AWD owners are there and how many would pay up?

I estimated $80M in this post. Don’t know if anyone modeled an amount.

Tesla, TSLA & the Investment World: the 2019-2020 Investors' Roundtable

 

[EVNow]

I estimated $80M in this post. Don’t know if anyone modeled an amount.

Tesla, TSLA & the Investment World: the 2019-2020 Investors' Roundtable

I'd be wary of drawing any real world conclusions based on self-selected polls in enthusiast sites.

Electrek’ s poll of 2000 AWD owners shows:
22% = yes
28% = not sure (including moi)

So that would be 36% if half of the not sures purchase.

However let’s assume 25% before Jan.1 = 160000 *0.25 * $2k = $80M xMas bonus
​

The poll tells me of the most enthusiastic 2k of the 160k owners, some 500 said they will upgrade. Rest of the 158k will likely have a much lower take rate. BTW, AWD wasn't even available for a few quarters in the beginning (when I bought my 3) - so that 160k AWD owners may be a bit of an overestimate.

So, the question is - what is the real take rate of the upgrade, if not 25% ? Is it 15%, 10%, 5% … no way to model. We I'd just take that as an upside potential of $25M or so. Given the likely error of +/- 100M in the model, probably not worth worrying about.

ps : My non-lease auto margin assumption is 22% (Q3 21.8%). A 1 p.p. change here causes $60M change in profit.