Tesla, TSLA & the Investment World: the Perpetual Investors' Roundtable

[RobStark]

View attachment 483714
Typical Tesla - taking all the credit of those hard working Merinos.

It should say Made on Earth by Sheep
with a little help from Human friends.

 

[Singuy]

I didn't think about how attractive a bulletproof truck would be in areas like Mexico...

A mayor in Mexico put down his own money to reserve 15 Cybertrucks, and he's hoping for a discount from Elon Musk

More attractive if it's bulletproof vs a 7.62x39mm as ak47 are the weapon of choice when I visited Mexico awhile back.

 

[Singuy]

This.

If you ever want to get mad, check out Scot Work's twitter page (@ghost_scot) and see what they do. They search out prospective customers and try to scare them off, and search out any owners with any sort of problems, try to convince them that everyone has their problem and hates Tesla, and to convince them to return their cars, sue Tesla, and/or complain to the media. They of course never bother mentioning that they're short-sellers.

If any of you ever have any spare time, by all means check his page regularly to see who they're targeting and let their victims know what game they're being played with.

LoL, that's like trying to stop a tidal wave with a bucket. All that time wasted to work on 1-2 customers every few days while 5k around the world took deliveries during that time.

Good strategy back in the days when Tesla were selling like 300/month.

 

[Buckminster]

Did that 82 pages ago. 83,000. Final answer. Spread the word. (And hold)

Lucky you have 9 lives with a WAG like that.
New poll:
Poll - Q4 M3 production

 

[kengchang]

Great writeup and very helpful.

On the arguments against, I would add that there are other items we have not seen that we would likely see if production was going to increase 20% or more QoQ.

• No word on additional shifts - Tesla would have had to have increased efficiency dramatically to production to ramp with the same number of shifts. This should have been noticed through increased hiring
• Substantial energy product increase - Rumours of increased powerwall and megapack production have circulated this quarter, potentially using more cell supply from Panasonic and trimming the potential to increase vehicle output as Pana increase production.
• Not enough leaks from suppliers - Given the number of suppliers, shouldn't we have heard from more than just one supplier that orders have increased

1. Model 3 is already running 3 shifts 24/7
2. Storage also uses cells from Samsung/LG
3. I mean the supplier "doubled" the production in August, isn't that enough?

 

[Fact Checking]

International ship loading days up 60% QoQ QTD and assuming we are now finished for the quarter, will finish up 49% QoQ. This suggests International Shipments of 50-55k Model 3s this quarter. This is real data by the way and not "tea leaves", some people are spending a huge amount of time tracking all the shipments.

Franco Mossotto on Twitter (he maintains the Tesla Carriers spreadsheet) is warning that ship loading time in this quarter might have been influenced upwards by the high number of ships compared to Q3:

I.e. number of ships increased from 12 to 16, a ~30% increase - and due to the significantly higher frequency of ships the individual ships might have been loading less efficiently, waiting for batches of cars from Fremont.

My interpretation:

• In particular the outlier of 4.07 loading days for "Grand Pioneer" might have been partially idle time, it would be conservative to correct this back to ~3 loading days which seems to be the broad max.
• The Q4 China loading times are also uncharacteristically high for individual ships: 3.24 loading days of the "Grand Quest", especially due to the close proximity of the arrival of both the previous and the next ships - so this too might have been 1 day shorter in reality.
• The ship trackers can see a ship docked, and that's the basis of the 'loading days' - but they cannot reliably measure the efficiency of ship loading. Once moored, a ship doesn't leave Pier 80 until fully loaded, even if it's idle.
• @Doggydogworld noted that the Thanksgiving break might have created more idle time (dock workers not working) than usual.

So to scale back some of my previous 'ship loading days' enthusiasm: it's not reliable to scale Q3 ship loading days to Q4 ship loading days: the Q4 international shipments number is possibly lower, and closer to the 30% increase in ship count than the 50% increase in loading days.

So I'd say an increase in production from 80k to 85k or maybe 90k is well supported, but 100k+ bets less so. In addition to @tschmidty, @ValueAnalyst is warning against too high expectations based on ship loading days as well.

I'd guess around 107k deliveries for Q4: 89k Model 3's and 17k S+X.

This is comfortably ~1k above guidance - and Tesla would push most other deliveries into Q1.

 

[Fact Checking]

3. I mean the supplier "doubled" the production in August, isn't that enough?

Well, this was a not particularly reliably sourced leak, also translated from Chinese (?), plus we don't know whether this doubling wasn't just some preemptive increase in more expensive and more tariff sensitive parts of the supply chain to get high density, low volume parts stockpiled in by the threatened steep tariff hike by December 15.

In August they might also have started stockpiling for the GF3 ramp-up and the Q2'2020 Model Y ramp-up - in exchange for better supplier prices. Tesla does have some cash to invest into better margins after all.

At these $330 price levels I think buying anything but straight stock or maybe Theta-heavy options expiring Q3'2020 and beyond is a significantly elevated risk (and even those are risky, as usual - this is TSLA after all) - and the recent jump in the stock price and the risk of a breakout made most options pretty expensive.

A number of cautious TMC bulls got burned by options in our Q4'2018 euphoria, which came crashing down and met cold reality in Q1'2019. Even buyers of the stock had a couple of stressful weeks in June ...

 

[EV Promoter]

Franco Mossotto on Twitter (he maintains the Tesla Carriers spreadsheet) is warning that ship loading time in this quarter might have been influenced upwards by the high number of ships compared to Q3.

I.e. number of ships increased from 12 to 16, a ~30% increase - and due to the significantly higher frequency of ships the individual ships might have been loading less efficiently, waiting for batches from Fremont.

Cannot find the graph showing the plateau for Model3 deliveries in Eu due to the accident of the headlight (perhaps), but i can add that in my EU country, some few orders placed between Nov 2 and 15, are consistently given a delivery time of 3 months, somewhere in February with no further specification. I wonder if any other EU country is experiencing the same, to validate if US delivery is given the preference, in spite of what some few people said here about the preference to EU deliveries.

 

[Fact Checking]

Cannot find the graph showing the plateau for Model3 deliveries in Eu due to the accident of the headlight (perhaps), but i can add that in my EU country, some few orders placed between Nov 2 and 15, are consistently given a delivery time of 3 months, somewhere in February with no further specification. I wonder if any other EU country is experiencing the same, to validate if US delivery is given the preference, in spite of what some few people said here about the preference to EU deliveries.

Yes, that's correct all new EU orders, in all countries have been pushed to February end of last week (including the highest priced Performance trims) - but this does not tell us anything but that there's enough EU buyers for all the cars on the ~8 EU-headed ships in Q4.

There is still new Model S and Model X inventory.

In Q3 they sent 7 ships to Europe, so there's an increase but not a huge increase in EU orders. They might send one more (smaller) ship to the EU, from the east coast. Most of the ship count increase in Q4 is towards Asia: from 5 in Q3 to 8 in Q4 - two of which are for Korea and are presumably smaller ships with fewer cars.

 

[Alphacrux]

TSLA premarket price does not seem to be available? Any reason why not? Does anyone have it?

Edit: ok, there it is : $331.50

 

[TradingInvest]

Well, this was a not particularly reliably sourced leak, also translated from Chinese (?), plus we don't know whether this doubling wasn't just some preemptive increase in more expensive and more tariff sensitive parts of the supply chain to get high density, low volume parts stockpiled in by the threatened steep tariff hike by December 15.

In August they might also have started stockpiling for the GF3 ramp-up and the Q2'2020 Model Y ramp-up - in exchange for better supplier prices. Tesla does have some cash to invest into better margins after all.

At these $330 price levels I think buying anything but straight stock or maybe Theta-heavy options expiring Q3'2020 and beyond is a significantly elevated risk (and even those are risky, as usual - this is TSLA after all) - and the recent jump in the stock price and the risk of a breakout made most options pretty expensive.

A number of cautious TMC bulls got burned by options in our Q4'2018 euphoria, which came crashing down and met cold reality in Q1'2019. Even buyers of the stock had a couple of stressful weeks in June ...

I assume joining S&P is still in the card. Tesla should satisfy the earnings requirement the earliest after Q1, latest after next Q3. TSLA will join S&P sooner or later, it's only a matter of time.

For normal companies, maybe it's easy to buy 20 million shares from the market. For TSLA, I don't think longs will sell their shares even if the price goes higher, especially when TSLA starts to have a string of positive quarters. So who will sell shares to the index funds?

Staying cautious is good, things don't always go smoothly. Overall I think shorts are not in a pretty situation. I just keep saving cash and get ready to add more shares whenever shorts work hard.

I ordered a set of snow tires from Tesla. I don't compare price with other sellers. I just want to make sure Tesla can earn a few hundred dollars of profit from me. As Tesla's scale increases, the financial situation will get better. I buy and hold shares and want to see shorts cry in the next 10 years.

 

[kengchang]

Well, this was a not particularly reliably sourced leak, also translated from Chinese (?), plus we don't know whether this doubling wasn't just some preemptive increase in more expensive and more tariff sensitive parts of the supply chain to get high density, low volume parts stockpiled in by the threatened steep tariff hike by December 15.

In August they might also have started stockpiling for the GF3 ramp-up and the Q2'2020 Model Y ramp-up - in exchange for better supplier prices. Tesla does have some cash to invest into better margins after all.

At these $330 price levels I think buying anything but straight stock or maybe Theta-heavy options expiring Q3'2020 and beyond is a significantly elevated risk (and even those are risky, as usual - this is TSLA after all) - and the recent jump in the stock price and the risk of a breakout made most options pretty expensive.

A number of cautious TMC bulls got burned by options in our Q4'2018 euphoria, which came crashing down and met cold reality in Q1'2019. Even buyers of the stock had a couple of stressful weeks in June ...

 

[lklundin]

More attractive if it's bulletproof vs a 7.62x39mm as ak47 are the weapon of choice when I visited Mexico awhile back.

Yes. That would require a non-trivial upgrade (of not only the glass) since the kinetic energy is about 4 times higher than for a 9mm.

 

[Fact Checking]

But since Model X's drag coefficient has been alternatively reported at 0,24 and 0,25, relative to the Model S's 0,24, why would you expect a 0,02 gain for the Model Y vs. the Model 3?

I was just guessing wildly - it might be 0.24 or 0.23.

But the Model Y range is 300 miles vs. 322 for the Model 3, i.e. 5-7% lower depending on how much nerfing there is in the published Model Y numbers. Which brings me to the next point:

Efficiency is a parameter that buyers hardly value at all. The knock-on impacts of efficiency, however, are highly valued by customers. These dramatic knock-on effects (range, charge times, longevity, operating costs, etc etc) are what has given Tesla their success in the marketplace vs. less efficient vehicles.

And efficiency is a major part of TCO. Efficiency is how much you pay for power at home, how much you pay on the road, and how quickly you rack up cycles on your (very expensive) battery pack, as well as how deep you're cycling said pack to go a given distance.

But range doesn't only depend on Cd, it also depends on mass and, in large part, battery capacity - which depends on the other factors such as production costs. If Tesla can save say $2,000 in overall material, labor and equipment depreciation costs, they could put that $2,000 into the battery pack and offer the CyberSUV at Model Y prices - with a $2,000 larger battery pack, which buys it about 20-30% more battery capacity in ~2 years time and thus significantly more range.

My point, I disagreed with your very first characterization of Cybertruck-derived vehicle designs:

Musk thinks that with "extreme effort" they might be able to get the Cd "down to" 0,30. By contrast, Model 3 is 0,23. So even if you could get a Cybertruck-shaped Model 3 LR all the way down to 0,30, it would only have a range of 236mi (if rolling drag was also hurt due to those offroad-style tires; if tires are left unchanged, somewhere approaching 250mi). Worse if they can't get all the way down to Cd=0,30. E.g.: a Cyber-M3 LR would get worse range than an actual M3 SR+.

Tesla was able to offer a $39,000 Cybertruck that is a shocking ~40% larger than the Model 3.

We cannot say anything about the economic viability of the CyberSedan or the CyberSUV based on the Cd alone, it's not even close, because the cost of goods appears to be so radically different - even taking ~2 years of +5%/year efficiency improvements in battery tech into account.

In fact I'd venture that based on these numbers the ~40% price advantage will likely enable a less expensive CyberSUV or Cybersedan, despite the knock-on effects of a 20%-30% worse Cd of 0.30 which are undoubtedly there.

 

[lklundin]

I don’t think people should count on a blowout quarter and I definitely don’t recommend people betting more than they can lose on short term stock price gains from Q4 deliveries/earnings, however it’s also stupid to ignore that there is at least a possibility of a large increase in production this Q, particularly in light of accumulating evidence. You are never going to make money on the stock market if you discard new evidence just because it doesn't match current consensus - consensus is priced in. You make money when you are right and consensus is wrong.

I’m sticking with a base case deliveries estimate of 110k (hitting minimum FY guidance requires towards 105k), but I think it is possible production could be anywhere from 104k to 120k and deliveries up to 130k given the 15k Model 3 inventory at the end of q3.

We have many data points which alone I would largely ignore, but together are painting a consistent picture: Jerome’s July email, CFTC supplier comments, Q3 “commercial negotiations” COGs reduction, Elon’s Q2 FY guidance, Galliers and Ellinghorst due diligence last year, Panasonic public comments, Carsonight leak, International shipments tracking, US delivery data in Troy’s spreadsheet.


Arguments for a large step up in Model 3 Fremont production:

• In July Jerome told employees they were preparing to raise output at Fremont. "The company is “making preparations” to raise output at its auto plant in Fremont, California, Tesla’s automotive president, Jerome Guillen, said in an email to employees on Tuesday. “While we can’t be too specific in this email, I know you will be delighted with the upcoming developments.”
• In July CFTC announced doubling of Model 3 components starting in August 2019. 'Precision stamping service provider China FineBlanking Technology (CFTC) will increase monthly shipments of components for relays used in Tesla Model 3 from about 20,000 units currently to 40,000 beginning August 2019, according to industry sources.” Presuming this is for both US and GF3, this could be towards 10k per week Model 3 production capacity.
• During Q3 Tesla achieved a very significant reduction in COGs which they partially attributed to "commercial negotiations with suppliers”. A significant reduction in supplier prices is consistent with a significant ramp in orders which delivers the supplier staff and depreciation cost leverage.
• In the Q2 call Elon said he expected Fremont Model 3 production to get to 8.3-8.6k per week by year end (but of course he may have been misspeaking and thinking about global production). Joseph Osha: And just as a follow-on then, could we see you manage to make 8,000, 7,500, 8,000 Model 3s in Fremont by the end of the year you think? Elon: Yes. I mean I feel confident it's -- let's just say that the trend is very clearly towards being able to get to 10,000 vehicles a week of which that would be -- there is rough numbers like 8,300 to 8,600 Model 3s and the balance in S and X. So, there's sort of 1,600 to 1,800 SX. In round numbers 8,500 3s, 1,500 SX per week, but probably a bit more than that.”
• Last year the Galliers and Ellinghorst report said Fremont could get Model 3 production to 8-10k per week with limited capex. “Based off our tour and what we saw, we see no reason why Stamping and General Assembly should not be able to handle [seven to eight thousand cars per week] today, and even potentially 10k units, with very little incremental Capex. We believe the same is also true for the Paint Shop when it comes to reaching 8k units a week, with some incremental capex potentially required to get to 10k units. For Body, our understanding is that incremental capex is required (our impression is in the tens and not hundreds of millions) in order to get to both 8k units and eventually 10k units.
• Panasonic published a graph showing a predicted 20% QoQ GF1 cell production increase in Q4 in a public presentation last week.
• Carsonight reported 20% QoQ GF1 Pack production increase to 8k per week. Cleantechnica reported 7k packs were sent to China in the 12 weeks to mid October before shipments were finished. If we assume this is true and there has been no further pack stockpile build since, then about 10 out of 12 weeks were produced in Q3 and 2 of 12 in Q4 (so 1.2k packs sent in Q4). This puts Q4 packs going to Fremont at 13*8k -1.2k = 104k -1.2k = 102.8k Model 3s produced at Fremont
• International ship loading days up 60% QoQ QTD and assuming we are now finished for the quarter, will finish up 49% QoQ. This suggests International Shipments of 50-55k Model 3s this quarter. This is real data by the way and not "tea leaves", some people are spending a huge amount of time tracking all the shipments.
• US delivery date entries in Troy's tracker are up 40% in the first 64 days of the quarter (up to December 3rd) vs the first 64 days of Q3. I have seen many people assume Tesla has been producing 100% international cars this quarter, but this doesn't look to be true at all. It seems US production has also increased QoQ so far this quarter. There are 149 delivery date entries in the spreadsheet in this period in Q4 vs 114 in the same period of Q3, 40% QoQ growth. This did include a large acceleration in the last 7 days. If we assume Tesla stopped International production after 57 days this Q we can instead take just the first 57 days of Q4. Here there are 107 entries or +4% growth vs 103 in Q3. InsideEVs estimated 44k US deliveries in Q3 (I don't trust their monthly breakdown) and in Troy's tracker 1/3rd of Q3 US delivery entries were in the first 57 days - suggesting c.15k US deliveries in the first 57 days of Q3. Taking 4% growth suggests 15.5k deliveries in this period of Q4. If we assume half of these were inventory and half new production it suggests Tesla potentially produced 58-63k Model 3s in the first 57 days - which extrapolates to 93k-101.5k total Q4 Model 3 production and largely reconciling with the datapoints we have from GF1.

Arguments against a large step up in Model 3 Fremont production:

• We haven’t heard stories of significant production shut down or new equipment installation at Fremont.
• We haven’t heard any specific leaks that Tesla is aiming for a huge step up in delivery targets this quarter. (But these emails are more common when Tesla is fighting to meet targets and may not be necessary if they are already almost sold out for the Q).
• Why did Tesla not ramp up to this production rate before if they had the capacity with limited capex? (Most likely because they needed to wait for Panasonic cell production rate to catch up and for word of mouth marketing plus new region rollouts to get quarterly demand to the next level.)
• Ramping up Model 3 production in the US shortly before ending exports to China and also shortly before potentially cannibalising demand with Model Y launch could lead to excess production for US and European markets next year. However it’s possible this ramp is only temporary and they will redirect components to GF3 and Model Y next year when the new production lines are ready for them.
• Towards 100k Model 3 production in Q4 just feels too good to be true.

Just before reading the above I picked up a handful of 320$ Jan. 17 calls, which I now feel pretty good about.

 

[CorneliusXX]

1. Model 3 is already running 3 shifts 24/7
2. Storage also uses cells from Samsung/LG
3. I mean the supplier "doubled" the production in August, isn't that enough?

1. Last I heard they had closed the tent. But the statement still stands. A huge efficiency increases is needed to squeeze the lemon that much more.
2. Indeed. But we don't know if that's where the extra cells are coming from.
3. It doesn't matter if one supplier increases 100x. Cars are complicated and multiple suppliers need to increase supply for production to ramp.

Not saying production hasn't ramped. Just that we are trying to tie many threads together to come to a conclusion. If there is information (or a lack thereof) suggesting otherwise it should be considered.

 

[Tslynk67]

TSLA premarket price does not seem to be available? Any reason why not? Does anyone have it?

Edit: ok, there it is : $331.50

On CNBC it only updates from the previous day's close when there's a certain number of pre-market shares traded. I don't know what that figure is, 100 or so, I think.

You can always see it on the Nasdaq page though even if it's just one share traded.

 

[Fact Checking]

3. It doesn't matter if one supplier increases 100x. Cars are complicated and multiple suppliers need to increase supply for production to ramp.

I believe Tesla has a supply chain of over 1,000 individual companies (over ~100 major ones), for the thousands of Model 3 parts they acquire from external sources. The maximum speed of production is determined by the weakest link, the slowest to ramp supplier.

"Complicated" is not the right word I believe.

 

[Fact Checking]

View attachment 483811
View attachment 483812

Thanks, this should make the "CFTC leak" reliable, backed by financial reports, right?

I don't think this is contradicting the "Tesla is stockpiling parts sensitive to tariff hikes" hypothesis though - even though I do agree that it's not the only explanation.

 

[lklundin]

Tesla deliveries in Austria 2019:

210: November
53: October
2513: Jan-Sept


Kraftfahrzeuge - Neuzulassungen
http://www.statistik.at/wcm/idc/idc...electionMethod=LatestReleased&dDocName=120255