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

[KarenRei]

Sitting from outside you or I couldn’t have foreseen the things Deepak could have, and that is the point of my criticism. This is one of the main jobs of the CFO.

For future they can do a few things.

• Do some stress testing of balance sheet and liquidity, play what-if scenarios.
• Remember capital and liquidity is important to any business, whether it is a start-up or mega banks. So keep sufficient cash. Don’t spend all of 2.6B, put at least half for buffer.
• Never put yourself in a box by saying we will pay bonds in cash or we will not raise capital. Say when market is appropriate we will raise capital if needed.
• Operations should work with finance..which means the “Wave” decision is not just operations decision but very much a finance decision.
• Timing is important in liquidity. Space out big capital outlays. So if FSD is the big investment, GF4 can wait a year (unless you are awash with cash, in that case deploying unused cash faster is the main concern).

In the end, the board should also step up and hold the CFO accountable. We focus too much on Elon, but being a clever finance/business guy is not his strong suit.

I think there's really only one "overall" lesson to be learned (apart from the various tactical mistakes of the quarter... which will always happen at random as the company grows). And that lesson is: "You're not maintaining enough cash in the bank to stop institutionals from freaking out if anything bad happens."

You don't have to be like Apple, but if you want institutionals to stop periodically dumping your stock whenever anything bad happens, you need to keep more cash on hand.

 

[WarpedOne]

Short squeeze will start when SP reaches 270?

No.
Short squeeze will start when SP reaches 720.

 

[ZachF]

Hey look at this!

 

[hcdavis3]

every day at 10:15am edt tmc goes down now??
locked out for 1/2 hour

Yup. It’s back now. I rebooted my iPad just in case.

 

[Drax7]

Sitting from outside you or I couldn’t have foreseen the things Deepak could have, and that is the point of my criticism. This is one of the main jobs of the CFO.

For future they can do a few things.

• Do some stress testing of balance sheet and liquidity, play what-if scenarios.
• Remember capital and liquidity is important to any business, whether it is a start-up or mega banks. So keep sufficient cash. Don’t spend all of 2.6B, put at least half for buffer.
• Never put yourself in a box by saying we will pay bonds in cash or we will not raise capital. Say when market is appropriate we will raise capital if needed.
• Operations should work with finance..which means the “Wave” decision is not just operations decision but very much a finance decision.
• Timing is important in liquidity. Space out big capital outlays. So if FSD is the big investment, GF4 can wait a year (unless you are awash with cash, in that case deploying unused cash faster is the main concern).

In the end, the board should also step up and hold the CFO accountable. We focus too much on Elon, but being a clever finance/business guy is not his strong suit.

Good points, from here onwards they should prepare for massive contractions ,
And assure themselves of ample liquidity. They did anticipate a contraction
But the magnitude was beyond their expectations. This will serve as a forcing function.

 

[PhaseWhite]

From above it looks like the FCA payments are at least 140m$/quarter.

So you're saying Tesla will report "One time" credits every quarter? Analysts/Reporters love the narrative that Tesla is juicing earnings with one time credits, but it seems this is a legit revenue stream that might even increase in the future as other automakers drag their feet on meeting emissions targets.

 

[Prunesquallor]

Peter Campbell on Twitter

Lots of questions on the Tesla/CO2 deal. Latest - does it get you over the CO2 line?


Manley estimates FCA would have picked up €390m fine this year.


Lays out v clear roadmap of how FCA intends to meet CO2 rules over 2020-22...

Peter Campbell on Twitter


Roadmap of FCA CO2 compliance:


2020

20% conventional ICE tech rollout

80% credit pooling


2021

40% ICE tech

45% from EV/hybrid rollout

15% credits


2022

50-60% EV/hybrid

40% ICE tech


Manley: "If there is need for pooling [in '22], it will be v v small"

Ok so. 390m eur fine dodged this year. Could be 140m $ total paid for 2019 to avoid.

I deleted an earlier reply to this post because I decided I didn't like the assumptions I was using. Here's the update.

My interpretation of the 2020 statement is that FCA thinks they can reduce their fleet average CO2 emissions from 120 g/km (2017 value) to 114 (20% of the way to their target of 91) using some kind of "ICE Tech". In order to get the rest of the way to compliance, they would need 190,000 ZEVs in the pool. (This assumes FCA fleet size of 912,000).

The first 37,000 ZEVs would be very valuable, since the Supercredit rules let you count each car x2. Each of these would be worth about 20,000€ in penalty reduction. Beyond that, each ZEV would be worth about 10,000€. So the total of 190,000 ZEVs would have a penalty reduction value of ~2B€

We're going to need a bigger boat.

 

[EVNow]

OT :

But solving the next 9 does not have a longer long tail than the previous 9. If the first 9 took 5k scenarios to solve, then the next 9 should also take c.5k scenarios. This is because the decreased frequency of events now you have moved further down your problem list scales in-line with the 10x smaller problem you are now solving (now only need to solve problems summing to 0.1-0.4%, rather than 1%-4% previously).

Lets assume this to be correct i.e. each 9 needs fixed number of scenarios to be solved. Then, we still have the question - how can this be solved exponentially quicker.

At different stages of the development cycle, Tesla is collecting high level summary data on 1) when shadow mode executes a manoeuvre which varies significantly from the human, 2) when a human disengages Autopilot and 3) full sensor data when a car has an accident. From the shadow mode and disengagement summary data, Tesla can try to group the events into categories of the most common problems (this could just be type of location and speed and type of disengagement data, or it could be more detailed info about what the NN sees).

I think this categorization will not be adequate. Edge cases will happen when the object recognition is incorrect or when the control software responds incorrectly to the movement of objects nearby. I don't know whether they collect data around movement of objects and send it with disengagements/shadow mode errors. May be this is what they need to do.

Tesla can use this to build a priority problem list of these subcategories. Tesla can then write heuristics for cars to collect data in these situations, or it can train a basic NN to recognise these situations in sensor data as a trigger to start collecting data. When Tesla start getting data back, they annotate this and feed it back to the recognition neural net. This makes the cars better at detecting the situation they are targeting, so the quality of data sent back by the fleet improves. At this stage they should be getting very high quality, relevant information from the fleet for the problems at the top of its priority list. Now humans have to review and annotate the data, and update the heuristics and retrain the neural nets.

Yes - lots of manual work here, which would make it linear.

As Tesla starts progressing through the march of 9s, shadow mode/real world disengagements will become less and less frequent. This means Tesla can start collecting more and more information from each of these events as their baseline (if these events are now 10x less common, Tesla can collect 10x more data from each event without increasing its overall data usage). This will make the priority problem identification easier as they move forward.

Yes - once you start getting real problems as disengagements - rather than routine ones, it starts becoming more manageable. Still very difficult to prioritize - but they are all low probability scenarios anyway, they can probably won't have to be very strict about prioritization. Just solve the issues as reported.

In all this Tesla hasn't talked about collected "bug reports" from drivers. They should do this. Ask drivers to do some categorization when sending bug reports. This would help them in prioritization. Show drivers the list of disengagements they had so they can pick the problematic ones and let them categorize (what kind of error and what they think caused it). Essentially offloading some of the categorization problems to drivers.

 

[MarcusMaximus]

Ok, so each capital raise gets us another $10 in SP. They don’t even need FSD, just raise capital 400 more times!

 

[dc_h]

yeah the way i read it is keep stk under 309.83 conv price if you want cash, otherwise convert to stock

any anti-rally effort, to me, is still just to suppress the stock and keep recent derivatives plays ITM, regardless of notes (that aren’t even convertible yet). that convertibles play comes down the road more in 2023 heading into feb 2024 before it turns into a straight conversion until maturity may 15 2024

It would be interesting to have a stock valued at 40 billion in 2023 with over 200 billion in revenue and 20-30 billion in profit. If the shorts have that much power, I may as well move to my shack down the street from Ted Kaczynski.

 

[schonelucht]

The SP will be natually capped at the convertible Senior notes price (set at May 2nd close or $244.10/share).

With the previous convertible there seemed to be a cap in place at round the conversion price, we touched $360 area multiple times but never really breached it. Wondering if that was real and if that cap was related to the convertible and if so, will we see the same kind of long-term cap again with this convertible? But then again, the 2018 convertible sailed right through it's conversion price.

 

[JBRR]

So you're saying Tesla will report "One time" credits every quarter? Analysts/Reporters love the narrative that Tesla is juicing earnings with one time credits, but it seems this is a legit revenue stream that might even increase in the future as other automakers drag their feet on meeting emissions targets.

read my other posts or the twitter thread I linked.

 

[Remus]

Wow, almost like the shorts are wrong in that the bankruptcy is imminent and nobody wants Tesla equity /s

 

[generalenthu]

I deleted an earlier reply to this post because I decided I didn't like the assumptions I was using. Here's the update.

My interpretation of the 2020 statement is that FCA thinks they can reduce their fleet average CO2 emissions from 120 g/km (2017 value) to 114 (20% of the way to their target of 91) using some kind of "ICE Tech". In order to get the rest of the way to compliance, they would need 190,000 ZEVs in the pool. (This assumes FCA fleet size of 912,000).

The first 37,000 ZEVs would be very valuable, since the Supercredit rules let you count each car x2. Each of these would be worth about 20,000€ in penalty reduction. Beyond that, each ZEV would be worth about 10,000€. So the total of 190,000 ZEVs would have a penalty reduction value of ~2B€

We're going to need a bigger boat.

It's a missed opportunity to not use the supercredits by having more manufacturers in the pool. Super credits are good for 7.5 gm/km reduction per manufacturer.

For instance they could add PSA and someone like maybe BMW to the pool. They can figure a figure a formula to split the avoided fines and come out better. I would be very very surprised if FCA is the only manufacturer in the pool for 2020.

 

[JBRR]

I'm not seeing any chatter on the fact that FCA confirmed that they are buying 1.8b eur worth of credit - most likely solely from Tesla, most of it in the next 18 months.

Are the analysts waiting to put this out bc they want to give their clients a chance to buy first?

 

[Artful Dodger]

Middlle Bollinger Band just touched 260.0

 

[gene]

A good way to raise stock price is to be sure Jay Inslee of Washington makes it into the democratic debates. His platform is mostly climate change. I'm a Bernie guy but I still just gave Inslee $50. He needs a certain number of donors to be automatically included in the debates. I think it is more the number of donors, not the amount of money so even $10 or $20 will do.

Inslee for America

The Democrat’s vision for a decade-long mobilization to combat climate change includes having the nation’s utilities powered entirely by carbon-neutral power by 2030. It also calls for all new smaller vehicles and buses to reach zero emissions and for new commercial and residential buildings to produce no carbon emissions by 2030. It would also close the nation’s coal fleet by 2030, while proposing support for communities and workers affected by those closures.

 

[Remus]

I don't know whether they collect data around movement of objects and send it with disengagements/shadow mode errors. May be this is what they need to do.

Yes they do!
kaparthy talked about cut in prediction, how they collect the data about vehicle movement before the cut in action. They also collect the data where the computer predicted a cut in but it didn't happen. All these are labeled by the car and sent to the mothership

 

[EVNow]

kaparthy talked about cut in prediction, how they collect the data about vehicle movement before the cut in action. They also collect the data where the computer predicted a cut in but it didn't happen. All these are labeled by the car and sent to the mothership

This is in a known situation. I'm talking about all disengagements / shadow mode errors.

 

[CorneliusXX]

With the previous convertible there seemed to be a cap in place at round the conversion price, we touched $360 area multiple times but never really breached it. Wondering if that was real and if that cap was related to the convertible and if so, will we see the same kind of long-term cap again with this convertible? But then again, the 2018 convertible sailed right through it's conversion price.

My read is that this is the tail wagging the dog. $1B of convertibles isn't much compared to the outstanding stock.