TSLA Market Action: 2018 Investor Roundtable

[Singer3000]

Any colour you want so long as it's black?

For the record, here's the colour selection on some other EVs:

Model 3: 1 base, 3 premium (was 5), 2 super-premium
Bolt: 5 base, 3 premium
Leaf: 3 base, 2 premium, 2 super-premium
i3: 1 base, 8 premium
Ioniq: 6 base
Kona: 2 base, 5 premium, optional 2-tone
Zoe: 1 base, 1 semi-premium, 4 premium, 1 super-premium
e-Golf: 12 base

More to the point, the real competition in the BMW 3 Series has 10 colour options. Episodes like this are a timely reminder that Tesla is a looong way away from the public goal of being “the world’s best manufacturer”.

On the plus side, for there to be a significant improvement in volumes and margin (particularly as the product suite broadens), Tesla doesn’t need to be the best but just become mediocre-to-good at the things that the industry at large have already figured out.

 

[avoigt]

It's from Elecktrek.co commenter "carsonight" who lives in NW Nevada and knows people that work at GF1. He says Panasonic's battery cell capacity is enough for 5,500 LR packs/wk on 10 lines each running flat out. Panasonic is labor constrained in Sparks, NV and has occassionally had to shut down some production on the graveyard shift due to insufficient staffing.

Here's his latest from today:

"GF1 can produce at most enough cells for 5500 Model 3s per week"

Cheers!

Thats indeed an interesting piece of information. Sounds credible. Posting all together for reference.



"The bottleneck is battery cells from Panasonic. Three more lines are supposed to be in full production by Christmas, which will bring them up to cells enough for 7000 long range Model 3s per week.

If you're talking about whether the new Grohmann machines will be for the SR Model 3s, I don't know.

I have been told that the short range Model 3 was ready to go this September but was delayed because of the unexpected popularity of the long range and all wheel drive units. I'm told it takes the same type of battery modules that are in the long range Model 3 but fewer of them, and I was told that the short range battery packs are much easier to handle than the long range because they're so much lighter. That is really all I have heard about it.

If the machines have the cells they can produce the battery packs. Cells for 8K Model 3s per week are not coming from GF1 until next year at the earliest.

That said, it would not surprise me at all if Elon Musk pulled a rabbit out of his hat. SDI can produce the 2170 cell, for example. That was a surprise when Tesla built the Australia battery on time using SDI cells, because GF1 was running at capacity at the time and there were those who were betting it couldn't be done. I would be the very last person to bet against Elon Musk.

The rate limiter is the 2170 cells produced by Panasonic. Those can be produced by others including Samsung, but I have not heard anything about that happening. In the past, Samsung has produced cells for storage when GF1 was overloaded.

Last I heard everything is for the Model 3, but I read that on line. To go above 5500 Model 3s per week will require more production from Panasonic. Those new Grohmann machines mean fewer people needed to assemble the battery packs."

 

[Johann Koeber]

Interesting that cell production seems to be the constraint for EV production not only for the competition, but also for Tesla ATM. And if cell production is truely limited by the workforce in Nevada, they better do something about it.

I trust Tesla will have the highest leverage to obliviate this bottleneck by building out the GF as planned.

 

[Tslynk67]

Utterly off topic, but so nice to see a positive Tesla article on BBC:

This woman drove her Tesla 20,396km around Australia

“The concern about the environment was front and centre. These cars are bloody expensive. But once you’ve paid for the car, your running costs are bugger all.” You tell ‘em Sylvia.

 

[Anstandswauwau]

I kind of wish Tesla would reduce its color selections further.

Me too.

I firmly believe three to four base colors are enough for such a sought after car. Sprinkle in some exotic color every now and then and market those accordingly as Limited Editions. Let people pre-oder the next n Limited Editions in advance and determine the appropriate schedule by popularity.

Easy money, increased desire, reduced complexity.

Hell, they could even give it the hypercar treatment and sell primed-only cars as a standard tier.

 

[MaChiMiB]

Thats indeed an interesting piece of information. Sounds credible. Posting all together for reference.



"The bottleneck is battery cells from Panasonic. Three more lines are supposed to be in full production by Christmas, which will bring them up to cells enough for 7000 long range Model 3s per week.

If you're talking about whether the new Grohmann machines will be for the SR Model 3s, I don't know.

I have been told that the short range Model 3 was ready to go this September but was delayed because of the unexpected popularity of the long range and all wheel drive units. I'm told it takes the same type of battery modules that are in the long range Model 3 but fewer of them, and I was told that the short range battery packs are much easier to handle than the long range because they're so much lighter. That is really all I have heard about it.

If the machines have the cells they can produce the battery packs. Cells for 8K Model 3s per week are not coming from GF1 until next year at the earliest.

That said, it would not surprise me at all if Elon Musk pulled a rabbit out of his hat. SDI can produce the 2170 cell, for example. That was a surprise when Tesla built the Australia battery on time using SDI cells, because GF1 was running at capacity at the time and there were those who were betting it couldn't be done. I would be the very last person to bet against Elon Musk.

The rate limiter is the 2170 cells produced by Panasonic. Those can be produced by others including Samsung, but I have not heard anything about that happening. In the past, Samsung has produced cells for storage when GF1 was overloaded.

Last I heard everything is for the Model 3, but I read that on line. To go above 5500 Model 3s per week will require more production from Panasonic. Those new Grohmann machines mean fewer people needed to assemble the battery packs."

The standard range model will likely be introduced when the new battery pack production is up and running.

Just having that very rapid iteration between design and production is incredibly helpful and we understand for example, what are the rate limiters, what makes it hard to produce battery modules. We came up with a new design that achieves the same outcome, that's actually lighter, better, cheaper and will be introducing that around the end of this year, probably reach volume production on that in Q1 or something. That will make the car lighter, better, and cheaper and achieve a higher range. That line is under construction, will be active in about six months.

-Elon, Q2 18 call

 

[Tslynk67]

I believe we can thank Fred Lambert for the pop in the last few moments.

Tesla production picks up in the past few days as end of record quarter approaches

He got his Model 3 now...

 

[Tslynk67]

This kind of price movement scares me. What on earth is causing this? The line is practically vertical...

It's vertically down that scares me, vertically up makes me feel all warm and funny inside...

 

[Tslynk67]

For the same reason that there’s no such thing as a Tesla Killer, the Taycan will sell as many as they make. Even if it isn’t as good as a Tesla—and I suspect it won’t be— it’ll be much better than ICE cars in that price range. It won’t conquer Tesla, it’ll conquer Tesla’s competitors.

It will conquer Porsche's profit margins...

 

[Tslynk67]

Thats pretty crazy:

The chard looks really healthy, the market still underestimates what the Q3 numbers will/can bring, all FUD did not have any effect, and every day passing people may reconsider. Still shorts are investing as if there is no tomorrow...

I abandoned the view that all shorts intend to make money. Sounds strange but I expect some of the larger manipulators have funds they received approval to lose if they manage to keep the SP under control. They figured thats financially a good deal. One day that will be discovered but it may take decades.

Chard, chart, chard-chart...

 

[Tslynk67]

I wonder how much of this is a bear trap?

Especially the smoking part, EM (not an idiot).

Obviously Elon would never knowingly manipulate $TSLA, but it does look like a perfect storm to me...

 

[Fact Checking]

Thats indeed an interesting piece of information. Sounds credible. Posting all together for reference.



"The bottleneck is battery cells from Panasonic. Three more lines are supposed to be in full production by Christmas, which will bring them up to cells enough for 7000 long range Model 3s per week.

If you're talking about whether the new Grohmann machines will be for the SR Model 3s, I don't know.

I have been told that the short range Model 3 was ready to go this September but was delayed because of the unexpected popularity of the long range and all wheel drive units. I'm told it takes the same type of battery modules that are in the long range Model 3 but fewer of them, and I was told that the short range battery packs are much easier to handle than the long range because they're so much lighter. That is really all I have heard about it.

If the machines have the cells they can produce the battery packs. Cells for 8K Model 3s per week are not coming from GF1 until next year at the earliest.

That said, it would not surprise me at all if Elon Musk pulled a rabbit out of his hat. SDI can produce the 2170 cell, for example. That was a surprise when Tesla built the Australia battery on time using SDI cells, because GF1 was running at capacity at the time and there were those who were betting it couldn't be done. I would be the very last person to bet against Elon Musk.

The rate limiter is the 2170 cells produced by Panasonic. Those can be produced by others including Samsung, but I have not heard anything about that happening. In the past, Samsung has produced cells for storage when GF1 was overloaded.

Last I heard everything is for the Model 3, but I read that on line. To go above 5500 Model 3s per week will require more production from Panasonic. Those new Grohmann machines mean fewer people needed to assemble the battery packs."

Very interesting and seems to rhyme with a number of other snippets we've learned about in the past couple of months:

• July 31th: "Panasonic to increase Gigafactory cell production by more than 30%". Includes information that lines increase from 10 to 13.
• This Electrek article about the Panasonic expansion mentions "35 GWh/year by end of 2018" figure. This figure is the same that Tesla's website currently claims as the target capacity of the Gigafactory.
• If 13 lines produce 35 GWh/year, then the current maximum output with 10 lines would be 27 GWh.
• Aug 1: according to Q2 shareholder by Tesla: "At the end of July, Gigafactory 1 battery production reached an annualized run rate of roughly 20 GWh, making it the highest-volume battery plant in the world by a significant margin. Consequently, Tesla currently produces more batteries in terms of kWh than all other carmakers combined."
• Model 3 LR has a battery pack of 80 kWh with a range of 320 miles, the 220 miles range of the Standard Range battery pack extrapolates to about 55 kWh.
• 20 GWh/year (end of July output) is 685 LR packs per day, or 4,794 Model 3 LR's per week max
• 27 GWh/year (max with 10 lines) is 924 LR packs per day, or 6,472 Model 3 LR's per week max
• If capacity utilization is 90% due to third shift labor shortage and inevitable line down-times, then the current output is about 5,760 LR battery packs per week, which is very close to the 5,500/week source you quoted above.
  
• 35 GWh/year (max with 13 lines by end of 2018) is 1,199 LR packs/day, or 8,390 Model 3 LR's per week max
• If they only made Standard Range Model 3's, then the max battery pack output would be 12,204 Model 3 SR's per week rate, by the end of 2018
• I believe that is where the "three Grohmann machines" are needed: if assembly of LR packs is currently limited at around 6,000/week, then they'll need to raise the rate by the end of year to be able to manufacture the SR
• The Grohmann machines are also going to free up assembly capacity for Storage/Energy products: right now there's a PowerWall 2 line re-purposed to make Model 3 LR packs - which capped Storage/Energy revenue growth in Q1 and Q2.

The exact sustained utilization will likely fall somewhat below the maximum rate, but it's pretty clear that they delayed the SR because they were able to sell so many LR versions in the U.S. alone.

As a next step I'd expect Tesla to make one of these moves (speculation):

• Once they are certain that Fremont can comfortably assemble 8k cars/week, introduce the SR but with Premium Interior and Dual Motor (AWD) only. This would price it at around $46k. If they drop the AWD price back to $5k again then it would be an entry price of $45k - still not $35k but getting closer.
• This would free up some cell output for much needed Storage growth.
• If by end of 2018 Tesla releases the off-ramp-on-ramp FSD feature then the EAP and FSD take rates will both increase, which have higher margin than the Long Range option ($9k).
• I.e. it will make sense for Tesla to introduce the SR version to improve margins: if a $9k LR option is converted into a SR+EAP+FSD plus color option sale ($8k+$1.5k), then margin of those options in a similar $49.5k sale will improve margins on those options from about 50% to above 90% (!). This improves (gross) margins from a $55k LR sale from 31% according to the Munro tear-down, to about 39.3% (depreciation/amortization cost-of-goods and other temporary inefficiencies not included). I.e. the introduction of the first variants of SR could, somewhat against expectations, increase Model 3 margins, at least temporarily.
• Of course there will also be a lot of sales at $45k - which would have an approximate gross margin of 25.1%. (Which is higher than the Munro tear-down of 23.6%, due to the recent increase of AWD.) If AWD stays at $6k then at $46k the margin would increase to 25.6%. But note that by introducing SR+PUP+AWD initially they lower the base price from ~$50k to ~$45k, which would keep the lowest-price-option demand at manageable levels.
• I.e. this is the first sign of Tesla potentially making money from 'software as a service' sales, where the Model 3 is a 'vehicle platform', and were lowering the price of the vehicle platform drives sales of near 100% margin software options.

 

[Tslynk67]

That’s a lot of S3X!

I think all of us here are in agreement that you can never get too much of that...

 

[Tslynk67]

There are “irrational” actors on both the bull and bear side — irrational in the sense that they are not strictly looking at Tesla/TSLA in isolation. Both sides are using their money to try to shape the future. Rather than offer an advice, I’d merely suggest that you consider the consequences of this supposition.

BTW, I’m long — very and firmly.

That's what lots of S3X does for you...

Sorry, I'll stop now

 

[McMountain]

Austria´s Automobile Club ÖAMTC has tested a Model 3 over here in Europe and are quite enthusiastic about it (especially notable since they are usually quite critical of anything electrical):

Tesla Model 3: Spannung pur beim ersten Test | ÖAMTC auto touring

Somewhat freely translated:

"Both reservation holders and critics of Elon Musk´s daily business now have to be very, very strong:

because for the first group the impatience will grow exponentially until Day X after reading these lines while for the others their favourite arguments against the electrification of individual mobility (low range and high price) have suddenly evaporated."

 

[Artful Dodger]

Weekend rap: "Space got a Tesla before I did"

 

[Artful Dodger]

The Grohmann machines are also going to free up assembly capacity for Storage/Energy products: right now there's a PowerWall 2 line re-purposed to make Model 3 LR packs - which capped Storage/Energy revenue growth in Q1 and Q2.

That line was converted from producing cells for the Powerwall to be a 10th line for producing cells for Model 3. It is a Panasonic line, not a Tesla/Grohmann battery pack assembly line.

Staffing those lines is an Panasonic HR issue, somewhat out of Tesla's control other than placing production orders.

There will be 13 lines in operation by the end of Q4 2018. There is room for 20 battrey cell lines in the present footprint of GF1, according to carsonight. That's equivalent to 11,000 LR packs per week, so more packs total when the SR packs starts to be produced. Model Y will need all that, and likely more.

We should expect Panasonic to punch out battery lines like movie tickets, and Grohmann to fill the seats with robot-like automatons.

Cheers!

 

[Fact Checking]

I'm surprised Panasonic would only be building enough lines to get Tesla to 8k. I would think that they would build enough to get to the full 10k and use any overflow to provide capacity to powerwalls/packs as the most recent information I have heard is that they converted a line used for PW/P into a M3 cell line to meet demand.

I'd put it this way: Panasonic is going to extend Gigafactory cell manufacturing capacity up to an about 12,200 packs/week rate by the end of 2018, when measured in Standard Range battery packs.

The LR/SR mix is unknown at this point, and clearly Tesla did not expect such high demand for the LR version - and is, as we all are (including Panasonic), pleasantly surprised about it.

Eventually they'll reach steady state:

• If the long-term take rate between SR and LR is going to be a 50%/50% then the Gigafactory output of 35 GWh at the end of 2018 is going to be enough for about 9,950/week Model 3 battery packs,
• a split of 60%/40% of SR/LR will shift output to about 10,325/week,
• a split of 70%/30% makes it 10,740/week, etc.

I believe those calculations are roughly what Tesla made back in 2016 when they planned for 35 GWh output - but neither they, nor we can know what exact options mix consumers are going to opt for.

As for the Powerwall/Powerpack assembly line that was re-purposed to make Model 3 battery modules: once the new Grohmann machines are installed in the next 1-2 months this line can make Storage/Energy products again, to utilize any cell output that isn't used to make Model 3's. The introduction of the Standard Range version will reduce the average battery usage of every Model 3, so there should be more cell output by that time at the latest.

Until then, higher than expected demand for the Long Range version of the Model 3 is going to tie up Gigafactory output.

BTW., I'd not be surprised that once Tesla hits 6k-8k packs/week output that Panasonic would add another three lines in the first half of 2019, to extend from 35 GWh to 43 GWh, and then another 3 lines to get to ~50 GWh. There's clearly enough growth potential on the Storage/Energy side, plus there's the 2170 use of the Model S/X refresh that could happen in late 2019 ...

We learned these Gigafactory capacity and expansion details shortly before The Tweet, and I'm not sure whether the (positive) implications are fully priced in yet.

 

[KarenRei]

Very interesting and seems to rhyme with a number of other snippets we've learned about in the past couple of months:

• July 31th: "Panasonic to increase Gigafactory cell production by more than 30%". Includes information that lines increase from 10 to 13.
• This Electrek article about the Panasonic expansion mentions "35 GWh/year by end of 2018" figure. This figure is the same that Tesla's website currently claims as the target capacity of the Gigafactory.
• If 13 lines produce 35 GWh/year, then the current maximum output with 10 lines would be 27 GWh.
• Aug 1: according to Q2 shareholder by Tesla: "At the end of July, Gigafactory 1 battery production reached an annualized run rate of roughly 20 GWh, making it the highest-volume battery plant in the world by a significant margin. Consequently, Tesla currently produces more batteries in terms of kWh than all other carmakers combined."
• Model 3 LR has a battery pack of 80 kWh with a range of 320 miles, the 220 miles range of the Standard Range battery pack extrapolates to about 55 kWh.
• 20 GWh/year (end of July output) is 685 LR packs per day, or 4,794 Model 3 LR's per week max
• 27 GWh/year (max with 10 lines) is 924 LR packs per day, or 6,472 Model 3 LR's per week max
• If capacity utilization is 90% due to third shift labor shortage and inevitable line down-times, then the current output is about 5,760 LR battery packs per week, which is very close to the 5,500/week source you quoted above.
  
• 35 GWh/year (max with 13 lines by end of 2018) is 1,199 LR packs/day, or 8,390 Model 3 LR's per week max
• If they only made Standard Range Model 3's, then the max battery pack output would be 12,204 Model 3 SR's per week rate, by the end of 2018
• I believe that is where the "three Grohmann machines" are needed: if assembly of LR packs is currently limited at around 6,000/week, then they'll need to raise the rate by the end of year to be able to manufacture the SR
• The Grohmann machines are also going to free up assembly capacity for Storage/Energy products: right now there's a PowerWall 2 line re-purposed to make Model 3 LR packs - which capped Storage/Energy revenue growth in Q1 and Q2.

The exact sustained utilization will likely fall somewhat below the maximum rate, but it's pretty clear that they delayed the SR because they were able to sell so many LR versions in the U.S. alone.

As a next step I'd expect Tesla to make one of these moves (speculation):

• Once they are certain that Fremont can comfortably assemble 8k cars/week, introduce the SR but with Premium Interior and Dual Motor (AWD) only. This would price it at around $46k. If they drop the AWD price back to $5k again then it would be an entry price of $45k - still not $35k but getting closer.
• This would free up some cell output for much needed Storage growth.
• If by end of 2018 Tesla releases the off-ramp-on-ramp FSD feature then the EAP and FSD take rates will both increase, which have higher margin than the Long Range option ($9k).
• I.e. it will make sense for Tesla to introduce the SR version to improve margins: if a $9k LR option is converted into a SR+EAP+FSD plus color option sale ($8k+$1.5k), then margin of those options in a similar $49.5k sale will improve margins on those options from about 50% to above 90% (!). This improves (gross) margins from a $55k LR sale from 31% according to the Munro tear-down, to about 39.3% (depreciation/amortization cost-of-goods and other temporary inefficiencies not included). I.e. the introduction of the first variants of SR could, somewhat against expectations, increase Model 3 margins, at least temporarily.
• Of course there will also be a lot of sales at $45k - which would have an approximate gross margin of 25.1%. (Which is higher than the Munro tear-down of 23.6%, due to the recent increase of AWD.) If AWD stays at $6k then at $46k the margin would increase to 25.6%. But note that by introducing SR+PUP+AWD initially they lower the base price from ~$50k to ~$45k, which would keep the lowest-price-option demand at manageable levels.
• I.e. this is the first sign of Tesla potentially making money from 'software as a service' sales, where the Model 3 is a 'vehicle platform', and were lowering the price of the vehicle platform drives sales of near 100% margin software options.

Agreed. The basic logic is that if cells are the limit and Fremont has excess capacity, they should start making SRs loaded up to the point where profit margins are 2/3rds or higher than that on LRs, thus making up the margin difference based on the number of cells. Increase the required margin to account for somewhat of an increase in SG&A. Decrease it for the PR value of selling SRs.

Shouldn't actually be hard to do. The current average sale price is $59,3k and they expect a 15% margin this quarter, meaning $8895 per vehicle. So SR would need a per-vehicle margin of $5930 to maintain the same per-cell margin. The zero-margin price for the LR this quarter (100% PUP, most with AWD or P, most with AP, some with FSD, most with paint options, half with wheel options, a few with Performance Upgrades, some with white interior, etc...) would be $50405. Reducing cell values at a cost of $100/kWh puts the SR at $47905, meaning they'd have to sell them for $53835, meaning that their average option value that they'd need to sell them with is $9835 (minus the cost savings on hardware of the options chosen at a lower rate than LR, plus the expense on hardware options chosen at a higher rate than LR). LR by contrast currently has an average spending on options of $15,3k (of which $5k - PUP - is mandatory).

In short, so long as they keep the options rate up, selling SRs looks quite plausible if cell supply is the limit.

 

[mongo]

As for the Powerwall/Powerpack assembly line that was re-purposed to make Model 3 battery modules: once the new Grohmann machines are installed in the next 1-2 months this line can make Storage/Energy products again, to utilize any cell output that isn't used to make Model 3's.

Minor correction:
Panasonic's PW/PP cell line was converted to making cells for the 3. The module assembly equipment is not compatible between the two types of products.

Nikkan reports that in order to meet the burst rate of 5,000 Model 3s per week, Panasonic had to pull some creativity out of its manufacturing hat. “Panasonic was forced to respond, for example, to convert cells for household batteries etc. to Model 3.”

Per BREAKING: Panasonic To Increase Gigafactory Cell Production More Than 30% By End Of 2018 | CleanTechnica