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

[anthonyj]

What happened to the pickup partnership rumor that Karen first broke to us?

 

[neroden]

I wish we could take a timeout and talk about more quantitative and financial modelling perspectives. Has this forum produced any post-results model? I would be more comfortable if I knew what it took exactly to stay cash flow positive.

Very very vaguely speaking, it takes ~5000 Model 3s / week, ~2000 Model S/X per week, and no inventory-in-transit buildup. More Model 3 means less model S/X needed, obviously. There are lots of other things which move the cash flow one way or another but basically they need to be making enough cars.

Sales volume has to go up dramatically to counter what will probably be descending model 3 per unit profitability, but it would be good to be more precise.

Wish I could be more precise, but basically, they need to be making enough cars. When they hit 10,000 Model 3/week they'll be very comfortably generating plenty of cash. Sadly it appears that's not until early 2020 at this point.

 

[Krugerrand]

But if there is a demand problem...then the sky is falling.....sorry couldn't help myself

That’s two more I need to buy Monday. I’m going to need a Scrooge McDuck sized vault.

 

[EinSV]

Good to know.

But going forward, it is best if this forum doesn't become too conceited about TSLA FSD. There's going to be a first mover advantage here and GM or Waymo will probably be first. Upon completion GM will probably partner with Uber. So TSLA will be fighting an upward battle to gain market share. Part of the reason why it is valued at 0. There's really no benefit at taking #3 market share in robotaxi.

Let's assume for the sake of discussion that Waymo and Cruise are approved first for some form of self driving (probably in a few cities). Optimistically let's assume they have a few tens of thousands of robotaxis on the road.

Then let's assume that six months later, the "laggard" Tesla flips a switch and has 500,000-1,000,000 Model 3s on the road that can be used as robotaxis on the Tesla Network, and is producing another 500,000-1,000,000 FSD enabled cars per year (mostly 3/Y). Those robotaxis last 1 million miles, and cost $50,000+ less per car to manufacture than the Waymo/Cruise-mobiles with expensive computers and LIDAR.

Who wins then?

 

[Singuy]

Very very vaguely speaking, it takes ~5000 Model 3s / week, ~2000 Model S/X per week, and no inventory-in-transit buildup. More Model 3 means less model S/X needed, obviously. There are lots of other things which move the cash flow one way or another but basically they need to be making enough cars.


Wish I could be more precise, but basically, they need to be making enough cars. When they hit 10,000 Model 3/week they'll be very comfortably generating plenty of cash. Sadly it appears that's not until early 2020 at this point.

It seems that Fremont is maxing out at 7.5k/week right? So they will not hit 10k until GF3 is online. But based on how Elon is pretty bullish on energy storage, I think there will be more cells produced than they need for cars by the q3 and q4. We should expect an explosion of revenue coming from storage at that time.

 

[M3Rider]

The tax credit cliff tested


Pent up US demand ? Already lots of inventory SR+ available in Chicago.

Well, orders were opened 2/28, deliveries prob. a week later, I believe insideevs estimates U.S. deliveries in March ~10k and not all of them were SRs. So, I thought that is not all that there is to it with so many people bugging Elon on twitter about his $35k promise, so many of them should be getting deliveries now or still getting their finances in order, 200k high trims were delivered so far, so I think there's got to be more to it than 10k SRs.

Also, those inventories you see they may not be stagnant, but getting replaced daily with new cars and getting assigned daily to those people whose turn came up. If you see cars on the lot it doesn't necessarily mean there's no demand.

I believe the new process was said to be that all cars are delivered to stores first and assigned to inventory(which you see), only then they are matched to buyers.

 

[neroden]

Okay this level 235$ is CLEARLY being defended. There might be some truth to those rumors after all. If that is true they have tipped their hand and it will be a losing battle for the near term.

Wait, what rumors? Who?

 

[MarcusMaximus]

On top of that, who is to say that they are not using cameras and NN learning as well.

They don’t have the quality or quantity of data Tesla has; not even close. And without data, neural networks don’t work well.

 

[Singuy]

Wait, what rumors? Who?

Something about short sellers doing napkin math and figured that's Elon's margin call.

 

[Krugerrand]

Aren't you losing $$ in # of trades?

There is a demand issue
There is a demand Issue
There is a demand Issue

(not, not not )

I considered that and thus buying in bulk once per day to cover the 5,032 posts I’ve already read and including extra shares for the 23,750 posts I’ve yet to read. Plus, it doesn’t cost me very much to buy. They like me.

I’m doing some hard maths to guestimate my extrapolation of averages and medians et al...

FYI, your post sadly does not count - no conviction, irrationality, ire or wrath and only mildly humorous. Ok, half a share for the humor.

 

[Beaver]

With all the doom and gloom today I just found something on Reddit that got me excited.

Tesla is opening a huge store and repair center on Golf road in Schaumburg Illinois. Golf road in Schaumburg is a car buying mecca of the NW Chicago suburbs. You can visit almost every type of car dealer within a couple of miles. Map with dealers:
Google Maps

Tesla will be just west of the Alfa Romeo dealer. I see this driving up sales in our area a lot. People going to test drive cars will stop by to test drive for the hell of it. I know from our experience once you do that, you are sold. Right now the Chicago locations aren't exactly convenient to get to.

Reddit thread on this location:
New Tesla Facility on Golf Road in Schaumburg, IL : teslamotors

Purchased for just over $13M:
Tesla (NASDAQ: TSLA) Service Center

 

[Beaver]

The tax credit cliff tested


Pent up US demand ? Already lots of inventory SR+ available in Chicago.

Chicago was in the low teens earlier this week. They received what appears to be two large shipments the last few days. It would be nice to see no inventory but it is moving.

 

[Krugerrand]

Ya feb 2016. I can't remember what happened anymore. Was it Model X burning too much cash that the ER was too ugly?

Don’t rightly recall for sure. So, so much has happened over the years it sort of all runs together. Thought a lot of it involved the beginnings of Brexit and Tesla was simply hit 10x harder. But you could be right that that was Model X related, though seems too early in the year. I do remember the panic on here. I also got a bit of heartburn myself for an hour or two - bad chili.

 

[Late2theGame]

It seems that Fremont is maxing out at 7.5k/week right? So they will not hit 10k until GF3 is online. But based on how Elon is pretty bullish on energy storage, I think there will be more cells produced than they need for cars by the q3 and q4. We should expect an explosion of revenue coming from storage at that time.

There was talk about Fremont potentially having a 10k/week capacity if they spent money on it:
Tesla, TSLA & the Investment World: the 2019 Investors' Roundtable
Has this changed? Would it cost too much right now?

 

[neroden]

The VIEs did not create a "cumulative cash benefit" of 750m+. If you dig into the old SCTY presentations you see four phases of solar leasing:

1) Installation - financed ~100% by non-recourse lenders and "tax equity" + "cash equity" investors.
2) First ~10 years of lease - monthly cash receipts go almost entirely to repay lenders and tax/cash equity investors.
3) Second ~10 years - monthly cash receipts mostly flow to SCTY
4) 10 year renewal period - monthly cash receipts flow to SCTY

Phases 1 and 2 were pretty much cash-neutral for SCTY. The VIE "cash flows" on the cash flow statement are mostly reversals of non-cash GAAP profit or losses, themselves driven by the incredibly arcane lease/VIE accounting. They don't represent actual cash inflows or outflows for SCTY/TSLA.

In fact, the VIEs are designed (this is standard practice) to generate a slight cash flow to SCTY/TSLA to cover management/overhead costs.

SCTY opex (mostly SG&A), on the other hand, is a real cash outflow. It was approaching $1b/year by the time SCTY hit the wall.

Correct.

SCTY continually issued stock and recourse debt to cover opex. Tesla should have slashed SCTY opex 80-90% within six months, but they blew half a billion or more dragging it out 18 months.

Correct, they really didn't shut that down fast enough. And they *really* didn't wind down leasing fast enough.

Again, this comes straight from the old SCTY presentations. No analysis required.

Tesla also took on SCTY's 1.6b of recourse debt. Since almost all was due before phase 3 cash flows kick in, Tesla has been paying that out of pocket. Whether Tesla will get their cash back depends on cash flows during phase 3 and 4. I found SCTY's assumptions reasonable for phase 3 and laughable for phase 4

FWIW, some of the leases are already in phase 3 (the earliest ones, from 2008) and more and more should be each year. I haven't dug up the old numbers for how much SolarCity leased each year; I know the 2008 (first year) leases were pretty insignificant financially but the 2009 ones are more substantial.

All told, I figure they'll eventually get their cash back plus a little. Of course they'd be better off to have that cash now instead of getting it 5-18 years down the road.

The timing is interesting to look at.

 

[Krugerrand]

Surely in at least one of the multiverses.

Sorry to say, you’re being hunted down in my simulation. Not sure how much longer you can run and hide. But you get an A for effort.

 

[neroden]

Interesting insight from AutoLine. I live in a *huge* apartment building on the San Francisco Peninsula. In my building's garage there is zero Tesla. None. The reason is very simple... there is no EV charger in the building at all.
Also... There is no EV charger in my office building's garage. In the past 3 years our office moved 3 times, but we stayed in SF. None of the office buildings had EV chargers... because they are sooo cramped that you can't even fit a bicycle rack in there.
And this is SF, fairly close to the tesla HQ.

Maybe I'm the minority, but it would be fairly inconvenient for me to drive a Tesla.

You can install a charger in the garage. You can literally force your landlord to let you -- it's California law. Consider talking to them about it.

 

[Carl Raymond]

Good to know.

But going forward, it is best if this forum doesn't become too conceited about TSLA FSD. There's going to be a first mover advantage here and GM or Waymo will probably be first. Upon completion GM will probably partner with Uber. So TSLA will be fighting an upward battle to gain market share. ANd without a good cash flow, TSLA will lose the cash bleeding competition like Uber lost against DIDI in China.

Part of the reason why it is valued at 0. There's really no benefit at taking #3 market share in robotaxi.

To start profiting from a driverless robotaxi they have to perfect vision. Nobody is close to where Tesla is now on this front. What the competition have is a lead to the halfway mark via lidar, but no way to reach the finish line. Tesla may not be winning, but only Tesla can win!

 

[neroden]

SEC settlement reached: Elon Musk and SEC reach agreement over Tesla CEO's use of Twitter

This is a perfectly reasonable and coherent settlement.

 

[neroden]

The source of the wave is transitioning assembly to produce the unique aspects of the exports - at full speed. I don’t know how they smooth that out other than accumulating inventory then shipping it at a constant rate. They need a buffer in the system.

OK, so the leaks are showing that they are transitioning between different regional models very fast now, like multiple times per day, without slowdowns.