General Discussion: 2018 Investor Roundtable

[Ugliest1]

Just been told by relative in NW who ordered M3 at the end of June, their car is ready for pick up tomorrow.

Wow, this was a fast turnaround, since he was told 3-5 months upon order. In reality, it was only about 2 weeks.

Does anyone have similar stories? Is this happening on a wider scale?

(Add: they had a reservation for two years)

I posted the full story in the Canadian Orders thread. Short version: my son got concerned with waiting for the AWD to open up, and ordered RWD the night before the AWD announcement. The next day he was notified his order was accepted and within 24 hours of that he was notified there was a car with his colour/specs ready and waiting in Vancouver.

What’s happening in many cases is, Tesla is building a bunch on spec and shipping them to various areas hoping someone will order that configuration. Makes sense when there’s such a low set of differences right now.

Epilogue: he’s had it for two weeks and the little bugger won’t let me borrow it. After all I’ve done for him, too!

 

[dc_h]

Slowing down now may actually be key to building millions of cars faster. If 3 quarters of profits & being added to the S&P 500 cause a significant bump for their market cap (say, $80-100B), it becomes much easier to buy one of the legacy automakers.

Purchasing a legacy automaker would get Tesla massive production capacity & a substantial amount of cash (more than $10 billion in most cases). ICE production lines could be converted to EV production lines as fast as Tesla can build out battery capacity.

And getting their ship together and knowing exactly how to design and build the car and plant to minimize capital expenditures and car costs. By waiting until their building at 10k per week they can fund internally and the production processes will be optimized.
Don’t forget they have investments planned in the Roadster and semi also.

 

[MikeC]

Just been told by relative in NW who ordered M3 at the end of June, their car is ready for pick up tomorrow.

Wow, this was a fast turnaround, since he was told 3-5 months upon order. In reality, it was only about 2 weeks.

Does anyone have similar stories? Is this happening on a wider scale?

(Add: they had a reservation for two years)

Yes, a friend ordered June 27 and is picking it up on Tuesday.

 

[dc_h]

When Tesla bought the NUMMI factory there was a metal stamping press already there. Though you are correct NUMMI was mostly just final assembly and most of the parts came from somewhere else.

Tesla has been moving sub-assembly work out of the main factory. Battery packs used to be made at Fremont, but are now done at the GF, even the S/X packs. I'm not sure where motor assembly is done, but I believe the seat shop was moved somewhere else in the Bay Area.

I was thinking Tesla could build some Model Ys in Fremont. By the time the Model Y is ready for production, it's likely their Chinese and possibly their European factories will be up and running. If they build all car models in each factory, that reduces the demand on the Fremont factory and they can scale back some Fremont production of existing cars to make room for North American Model Y manufacturing.

On the other hand if some foreign manufacturers close factories in North America due to the trade war, Tesla might be able to pick up those recently abandoned factories on the cheap with a local workforce ready to go.

As far as capital raise goes, Tesla could possibly raise money in 2019 by selling more stock. Especially if it goes up later this year.

Hopefully the 3 and Y can share most of the assembly line. That would save capex and allow for more regional production. Probably can’t sell 10,000 3’s a week in the America’s alone. Model Y maybe, but would be happy to see them production constrained in Fremont while they invest in other plants. Preparing now for recession in 2020 may be wasting time, but I think it’s a good time to start getting away from betting the farm. 2020 would be 12 years between recessions. With tech change acceleration we could go longer, but human nature is likely to intervene.

 

[Duffer]

Highly, highly doubtful Tesla would engage a Chinese battery manufacturer for Shanghai GF. Nuts actually. They will either partner with Panasonic again or do the battery manufacture themselves in the GF.

Are you asserting that 20 years of China stealing western IP is just "western propaganda" ? This is understood and documented by every tech company and security agencies in western countries.

Tesla contracting to have a Chinese battery company make it's cells would just save Chinese security agencies from the effort to try and pry it out piece by piece. Having better cells than other EV companies is a key element of Tesla's competitive moat.

I was being a smart*** and I forgot the smiley when I wrote "western propaganda". They have been stealing... I mean "acquiring" IP for years. In fact, I wouldn't be surprised if at some point it's found out that gunpowder was actually developed by someone else.

I agree that having better batteries than everyone else is why Tesla is the leader. However, Tesla has released patents previously all in the name of furthering the mission. The question is do you license the battery patents to technically keep the IP in-house?

In this case, I'm sure that at least a few MS/MX that were delivered to China were carefully disassembled and analyzed by the leaders in their respective fields, including battery production. Yes, there probably a few "tricks" that would need to be figured out or downloaded by a GF worker but not having a Chinese battery manufacturer is only delaying the inevitable.

 

[22522]

Hopefully the 3 and Y can share most of the assembly line. That would save capex and allow for more regional production. Probably can’t sell 10,000 3’s a week in the America’s alone. Model Y maybe, but would be happy to see them production constrained in Fremont while they invest in other plants. Preparing now for recession in 2020 may be wasting time, but I think it’s a good time to start getting away from betting the farm. 2020 would be 12 years between recessions. With tech change acceleration we could go longer, but human nature is likely to intervene.

The Y is supposed to be made elsewhere.

 

[Todd Burch]

I’ll be grateful if it causes a burn of some sort.

 

[JRP3]

Maybe the cat dragged in a flamethrower.

 

[AlMc]

Maybe the cat dragged in a flamethrower.

I would settle for a video of the factory with EM and a reporter.

 

[Pale Blue Dot]

And you have zero evidence the 5k is not a sustained or sustainable rate.

I remain thankful for your added bear fodder.

Every quarters average production rate is lower than the burst rate announced at the end of the previous quarter. This clearly shows that the burst rate is not immediately maintained right after it's announcement but takes several weeks to become sustainable. I've just written a new article showing more evidence that this lag is real and predictable using VIN registrations as a predictor.

Indications That Tesla's Burst Rate Is Not A Fluke - Tesla, Inc. (NASDAQ:TSLA) | Seeking Alpha

 

[mongo]

Every quarters average production rate is lower than the burst rate announced at the end of the previous quarter.

So, assuming 3 quarters of data in a ramp is even meaningful for determining a trend, are you speaking of the average over all weeks in a quarter, or only the weeks the plant is running?
Q1 last seven days: 2,020 vehicles. Q2: 28,578 vehicles. 2,598 over 11 weeks, 2,381 over 12 weeks, or 2,198 over 13....
Wait, even over 13 weeks the average is higher than the last week of the previous quarter...
What are you talking about???

 

[sundaymorning]

Yes, a friend ordered June 27 and is picking it up on Tuesday.

Do you know what the specs of the car he ordered was? AWD, Performance, etc?

 

[wdolson]

When the factory went into production, there was a line of four stamping presses. I'm not sure how many were already there, but at least the biggest one was bought second-hand from a factory in Michigan and moved.

I saw something contemporaneous from when Tesla got the NUMMI factory and there was at least one large metal stamper already there. I believe it was a couple of stories tall and Toyota abandoned it because the building was essentially built around it. I'm not surprised they got some more presses.

 

[Esme Es Mejor]

Every quarters average production rate is lower than the burst rate announced

No. It. Isn’t.

Using 12 weeks of production, which is what the steady-state S/X does:

The exit rate for Q4 was 797. Tesla averaged over 800/week in Q1.

The exit rate in Q1 was 2020– Tesla averaged over 2300/week in Q2.

The exit rate in Q2 was 5031. If they average more than that in Q3, it will be... completely normal and predictable.

 

[dmckinstry]

Every quarters average production rate is lower than the burst rate announced at the end of the previous quarter. This clearly shows that the burst rate is not immediately maintained right after it's announcement but takes several weeks to become sustainable. I've just written a new article showing more evidence that this lag is real and predictable using VIN registrations as a predictor.

Indications That Tesla's Burst Rate Is Not A Fluke - Tesla, Inc. (NASDAQ:TSLA) | Seeking Alpha

I really think you meant to say "Every quarter's average production rate is higher than the burst rate announced at the end of the previous Quarter."

 

[Pale Blue Dot]

So, assuming 3 quarters of data in a ramp is even meaningful for determining a trend, are you speaking of the average over all weeks in a quarter, or only the weeks the plant is running?
Q1 last seven days: 2,020 vehicles. Q2: 28,578 vehicles. 2,598 over 11 weeks, 2,381 over 12 weeks, or 2,198 over 13....
Wait, even over 13 weeks the average is higher than the last week of the previous quarter...
What are you talking about???

At the end of Q4 17 they announced a burst rate of 1000 but averaged only 800 for Q1, including the burst rate of 2100 in the final week. If you subtract the 2100 they averaged 689 in the 1st 11 weeks of Q1 below the 1000 burst rate.

Likewise they averaged 2100 in Q2 for the first 11 weeks while making 5k in the final week. Mid June, Musk wrote in an internal email that they achieved 3500/wk. That means they did 12000 in the last 3 weeks alone (3.5K+3.5K+5K) which gives us an average of only 1842 for the first 9 weeks of the quarter, lower than the 2100 burst rate.

Also if you read my current article i discovered VIN registrations predict total production by 45 days which generates an approximate average of 5K/wk no sooner than the month of august based on the recent VIN registration velocity.

The point being is that the sustained rate is not immediately achieved after the burst rate. Even in Musks latest interview with Bloomberg he says these are proof of concept burst rates achieved under extreme conditions which eventually become sustainable in the following weeks. There is also a direct quote in my last article , from Musk himself, that says a target burst rate of 6k/wk by end of quarter will move them closer to a sustained rate of 6K/wk a few months later.

 

[mongo]

At the end of Q4 17 they announced a burst rate of 1000 but averaged only 800 for Q1, including the burst rate of 2100 in the final week. If you subtract the 2100 they averaged 689 in the 1st 11 weeks of Q1 below the 1000 burst rate.

Likewise they averaged 2100 in Q2 for the first 11 weeks while making 5k in the final week. Mid June, Musk wrote in an internal email that they achieved 3500/wk. That means they did 12000 in the last 3 weeks alone (3.5K+3.5K+5K) which gives us an average of only 1842 for the first 9 weeks of the quarter, lower than the 2100 burst rate.

Also if you read my latest article i discovered VIN registrations predict total production by 45 days which generates an approximate average of 5K/wk no sooner than the month of august based on the recent VIN registration velocity.

The point being is that the sustained rate is not immediately achieved after the burst rate. Even in Musks latest interview with Bloomberg he says these are proof of concept burst rates achieved under extreme conditions which eventually become sustainable in the following weeks. There is also a direct quote in my last article , from Musk himself, that says a target burst rate of 6k/wk by end of quarter will move them closer to a sustained rate of 6K/wk a few months later.

None of that is the quarter's average production rate. Fremont was shutdown for about 2 weeks of Q2, yet you are dividing by total weeks of a subdivided quarter...
The 5k mentioned by @Krugerrand was the last week's production, not a burst rate, not an extrapolated rate, a full week build rate.
Final 7 days of Q1 was 2,020 not 2,100.

End of Q4 was 793 over the last seven days.

And, no, not going to go out to SA to see how this is supposed to be reasonable.

 

[Reciprocity]

Highly, highly doubtful Tesla would engage a Chinese battery manufacturer for Shanghai GF. Nuts actually. They will either partner with Panasonic again or do the battery manufacture themselves in the GF.

Are you asserting that 20 years of China stealing western IP is just "western propaganda" ? This is understood and documented by every tech company and security agencies in western countries.

Tesla contracting to have a Chinese battery company make it's cells would just save Chinese security agencies from the effort to try and pry it out piece by piece. Having better cells than other EV companies is a key element of Tesla's competitive moat.

There are no moats when you have a cyborg dragon.

I agree that this might be the time that Tesla diversifies away from Panasonic. Maybe their own cells with perbix and grohmann making the automation equipment. It could also be that Tesla has Jeff dahn breakthroughs coming for this factory, though I would assume they would use those in Nevada as well. Could be Samsung sweetened the deal. But, It's probably going to be Panasonic.

The point is that it kind of doesn't matter. Tesla's moat is not the cell chemistry, anyone can buy a model 3 and reverse engineer the cells and the pack. Tesla's moat is that they have the balls to do it all full boar and push the limits of automation. What can't be reverse engineered is the machine that makes the machine. It's not there yet but soon. The cell chemistry matters little and everyone will have it soon enough and Tesla doesn't care, they want more EVs. They also want to lead the EV revolution, which is what they are doing. Not because of cell chemistry but because they had the balls and will to make a gigafactory 4 years before they needed it so that it would be ready for Tesla to double production every 9 months.

There are no moats. Only cyborg dragons.

 

[MikeC]

Do you know what the specs of the car he ordered was? AWD, Performance, etc?

RWD, LR, white,18s

 

[ggr]

There are no moats when you have a cyborg dragon.

I agree that this might be the time that Tesla diversifies away from Panasonic. Maybe their own cells with perbix and grohmann making the automation equipment. It could also be that Tesla has Jeff dahn breakthroughs coming for this factory, though I would assume they would use those in Nevada as well. Could be Samsung sweetened the deal. But, It's probably going to be Panasonic.

The point is that it kind of doesn't matter. Tesla's moat is not the cell chemistry, anyone can buy a model 3 and reverse engineer the cells and the pack. Tesla's moat is that they have the balls to do it all full boar and push the limits of automation. What can't be reverse engineered is the machine that makes the machine. It's not there yet but soon. The cell chemistry matters little and everyone will have it soon enough and Tesla doesn't care, they want more EVs. They also want to lead the EV revolution, which is what they are doing. Not because of cell chemistry but because they had the balls and will to make a gigafactory 4 years before they needed it so that it would be ready for Tesla to double production every 9 months.

There are no moats. Only cyborg dragons.

I'm not sure I agree with this. If I took a sample of scrambled egg, could I know what the original egg looked like, or how to cook it? Sure, a good lab could tell me exactly what chemicals were in the cell. Actually we already know they're using 8:1:1 NCA, they've told us. (Maybe I got that wrong, but...) That doesn't tell us how to produce the film, how to bake it, how to roll it, is there something in there that bakes out, I can't even begin to ask the right questions.