TSLA - Tesla Motors

[Dalal.Holdings]

Old school value investors will always underestimate cult stocks...I mean, Buffett only started buying AAPL in recent years because of its brand value/moat which has arguably existed for decades (a bit late to the party...). If you can’t see any “brand equity” in the fervor of Tesla’s customer base then I can’t help you.

 

As Peter Thiel says, great companies can look very cult-like (esp to outsiders). Apple fans seemed very “elitist” to outsiders for a long time as well...and people also used to say “good for them, not everyone else can pay x for a computer”. Obviously there is a lot of uncertainty and room for failure for Tesla, but viewing fanatacism of its customer base as a weakness is definitely wrong.

 

Anyway, the thing that is easiest to make fun of is those who believe the risk-reward set up for shorting this stock is somehow an enticing proposition...I am not long, but I can make fun of “value oriented” shorts. How’s Einhorn been doing lately?

[Grant]

The gross margin increase for Models S/X you uncovered here (5-6 percentage points over the last year or so) didn't strike me as too out of the ordinary.

 

It was a 5% gain from Q2 to Q3. Q2's (estimated) gross margins were already around the best the car had seen. S/X margins have been flat since 2017 Q1 (around the same levels they were at early 2014 before the X launch).

 

ERICOPOLY had a useful post a few pages up where he shared some numbers re: battery costs he got from a Tesla service center employee.  According to him, the price of a base Model S battery pack went from 48k to 25k in the ~6 years since product launch.  So that's a ~4k/year reduction in cost per vehicle.  That alone gives you a ~5 percentage point increase in gross margins per year, assuming each car sold for ~80k and the battery cost reduction was not passed through to customers.

 

This is the post:

 

I went to the Rocklin service center today with questions on the battery packs.

 

Info collected

1). Battery packs will soon come down 30% in price

2). A refurbished Model S 85kwh replacement battery pack costs $17,000

3). A new Model S 85kwh replacement battery pack costs $25k now

4).  Originally, a new Model S battery pack was about $48,000

 

They are planning on soon making available 100kwh replacement packs as upgrades for cars with smaller packs.

 

He said the Gigafactory is responsible for the anticipated 30% upcoming price drop

 

These sound like retail costs, so will be much greater than the costs to Tesla. Manufacturers typically mark parts up a lot. Anyway, if there are big jumps down in price instead of steady declines, that could certainly explain the jump. As could the Model 3 ramp adversely affecting S/X production.

[SHDL]

The gross margin increase for Models S/X you uncovered here (5-6 percentage points over the last year or so) didn't strike me as too out of the ordinary.

 

It was a 5% gain from Q2 to Q3. Q2's (estimated) gross margins were already around the best the car had seen. S/X margins have been flat since 2017 Q1 (around the same levels they were at early 2014 before the X launch).

 

ERICOPOLY had a useful post a few pages up where he shared some numbers re: battery costs he got from a Tesla service center employee.  According to him, the price of a base Model S battery pack went from 48k to 25k in the ~6 years since product launch.  So that's a ~4k/year reduction in cost per vehicle.  That alone gives you a ~5 percentage point increase in gross margins per year, assuming each car sold for ~80k and the battery cost reduction was not passed through to customers.

 

This is the post:

 

I went to the Rocklin service center today with questions on the battery packs.

 

Info collected

1). Battery packs will soon come down 30% in price

2). A refurbished Model S 85kwh replacement battery pack costs $17,000

3). A new Model S 85kwh replacement battery pack costs $25k now

4).  Originally, a new Model S battery pack was about $48,000

 

They are planning on soon making available 100kwh replacement packs as upgrades for cars with smaller packs.

 

He said the Gigafactory is responsible for the anticipated 30% upcoming price drop

 

These sound like retail costs, so will be much greater than the costs to Tesla. Manufacturers typically mark parts up a lot. Anyway, if there are big jumps down in price instead of steady declines, that could certainly explain the jump. As could the Model 3 ramp adversely affecting S/X production.

 

I do agree with your points.

 

More generally I'm starting to think the bull case here (not saying I'm bullish on the stock at this time, just to be clear) is that the company actually has in hand a potentially very profitable product lineup whose potential was masked up to this point by various production/design issues that needed sorting out.

[ERICOPOLY]

Eric, as far as I’m concerned I would be glad to hear your explanation about how they achieved those margins.

 

They had the best margins in the sector in one of the most challenging quarters (production hell & delivery hell). Also they have some clear disadvantages compared to other automakers:

- location of the plant (higher cost)

- proprietary ERP system that is not fit for mass manufacturing

- no railway connection from the plant to the rest of the country - all cars have to be moved by truck (Musk his idea of directly delivering to a customer’s house is definitely more costly than what traditional automakers do)

- ...

 

Funny thing is that all this also happened in a quarter (just after) the VP of manufacturing and the CAO left, but probably these people just had probably issues with Musk his ‘way of working’ .

 

Btw: someone received confirmation from Tesla IR that inventory is treated on a car by car basis. This could mean that cars that required rework (with for instance 5% margins) were held as inventory in Q3 while ‘good cars’ (with for instance 15% margins) were sold. This underbuilds the thesis somewhat that certain costs are still hiding within inventory.

 

 

I'm not capable of fundamental analysis really, so you are asking the wrong person your questions.  I cannot tell you how much the stock is worth either.  I don't think any of the short sellers can tell you what the stock is worth -- they are relying on their own biases ("it can't be possible") to reach valuations.  I recall Buffett saying that he prefers boring companies with a long financial history that make boring staple products with a wide moat -- I believe he says this because forecasting the future is enormously difficult and you need every bit of help you can get.  But here we have legions of value investors ignoring the great sage's advice and presuming to know that the worth of Tesla is lower than here, even though Tesla does not have a long financial history, etc...

 

I have taken this attitude towards shorting the stock since it was price at $20 or $30.

 

One other piece of advice from Buffett that the shorts aren't following is to not follow a company quarter by quarter.  Valuation works in both directions doesn't it?

[ERICOPOLY]

Panasonic has already taken credit for holding up the production rate of the Model 3:

 

"The bottleneck for Model 3 production has been our batteries. They just want us to make as many as possible.”

 

https://electrek.co/2018/09/26/panasonic-ahead-schedule-tesla-gigafactory-1-battery-cell-production-lines/

 

Tesla Gigafactory 1, which started in 2013, has already grown into the biggest battery factory in the world with an annual production capacity of over 20 GWh.

 

Panasonic said that it planned on adding three production lines to increase the capacity to 35 GWh by the end of the year to support the production ramp.

[ERICOPOLY]

Eric, as far as I’m concerned I would be glad to hear your explanation about how they achieved those margins.

 

I will take a stab at it from 10,000 foot view.

 

1) . A German firm says that it should cost $28,000 to build a Model 3

2).  I have not done my own teardown of the car, am not an automotive engineer, and cannot refute them.  I have no reason to dispute their analysis.

3).  Average selling price is far higher than $28,000 but Tesla at the end of Q2 talked about having a lot of costs of production left to cut.

 

 

https://qz.com/1294282/the-tesla-model-3-cost-28000-to-build-german-engineers-say-and-it-still-may-not-be-profitable/

 

[Grant]

This is the most up-to-date story on Munro's Model 3 teardown: https://www.bloomberg.com/news/videos/2018-10-17/what-engineers-found-when-they-tore-apart-telsa-s-model-3-video

 

Munro estimates the car (a LR RWD Model 3 I think) would cost $34,700 to build, if it was built at a conventional car factory. To hit that number at Fremont, Munro says it'd have to function with 35-40% fewer employees. Then add in a few more grand for warranty reserves, sales and delivery.

[ERICOPOLY]

Grant, didn't they include the price of a long range battery?

 

The $35,000 model will have a short range battery and costs are perhaps soon coming down 30%.

[ERICOPOLY]

JB Straubel sometimes corrects Musk on the spot.  I don't believe this company is intentionally trying to deceive anyone and I think Straubel would be gone if there was no "there... there".

 

I think Musk is a bit histrionic and perhaps bipolar could explain his behavior -- I don't know.  He made that crazy tweet about the Saudi's when he was reportedly not getting any sleep.

 

Straubel looks to be a real engineer and not a huckster.

 

People point a lot to Musk's erratic behavior but why not point instead to Straubel's steady behavior?  Is there a bias involved in that choice?

 

[Grant]

Yes, as I said Munro tore down a LR model.

[ERICOPOLY]

Yes, as I said Munro tore down a LR model.

 

They tore apart a rear-wheel drive LR model but only the dual-motor LR model can be ordered today (so I can't find the price for the model that Monroe tore apart).

 

However, they are offering the mid-range battery rear-wheel drive model today for $45,000.

 

Had Monroe tore apart a mid-range battery model it would have been significantly lower cost than $34,700 I presume.

 

There is no other luxury car being produced in these volumes on a single production line, is there?  Economies of scale?

[Grant]

Yes of course the MR model would be cheaper to produce. It sells for $46,000 by the way, $3,000 cheaper than the LR RWD did.

 

I'm not sure on production volumes from individual factories. Other manufacturers build multiple models sharing the same platform on a single line, and the Model 3 is certainly not the best selling platform.

[ERICOPOLY]

Okay, suppose the cost then on the $46,000 car is about $35,000 to produce with the mid-range battery and warranty reserves factored in -- manufactured at reliability level of big-name firms.

 

Those are decent margins so why aren't they doing it already?

[John Hjorth]

No skin in this game for me.

 

Here, I'm just trying to shed some light on the Tesla accounting - data driven, based on what information we already have - with regard to the margin discussion, which I right now consider a bit - humm-ho - "circular", as of now in this topic [all, please take no offense here].

 

Basis: Tesla 2018Q3 Update Letter, specificly p. 7: "Condensed Consolidated Statements of Operation".

 

Activities:

 

1. Automotive sales

2. Automotive leasing

 

3. [= 1+2] Subtotal automotive

 

4. Energy generation and storage

5. Services and other.

 

[Total turnover = 3+4+5]

 

- - - o 0 o - - -

 

So, in a way, what you're looking at in the income statements in the 2018Q3 letter, is a "collapsed" income statement with regard to activities. It's an aggregation of the activities 1, 2, 4 & 5, except for turnover and cost of revenues. For each period, they could actually be perceived as sum of columns over activities.

 

- - - o 0 o - - -

 

Now, let's "zoom in" on the activity "Automotive sales":

 

Margins [measured on turnover] for Automotive sales :

 

2018Q3 [isolated] : [tUSD 5,878,305 - tUSD 4,405,919]/tUSD 5,878,305 = 25.04 %

2018Q2 [isolated] : [tUSD 3,117,865 - tUSD 2,529,739]/tUSD 3,117,865 = 18.86 %

 

2018Q3 [YTD] : [tUSD 11,558,051 - tUSD 9,027,055]/tUSD 11,558,051 = 21.18 %

2018Q2 [YTD] : [tUSD 5,679,746 - tUSD 4,621,136]/tUSD 5,679,746 = 18.63 %.

 

- - - o 0 o - - -

 

Q : Now, what is a margin?

A: That depends [on who you're looking at]. Many definitions exist. A margin is basically defined by the entity that you're looking at [by its own reporting of margins]. - So you need to understand how margins are defined by the one that you're looking at, basically, based on its reporting.

 

- - - o 0 o - - -

 

Tesla highly likely runs an ERP-system based on SMC [standard manufactoring cost].

 

And because Tesla is capital intensive, it is most likely defined and set up based on this:

 

[Everything below is per unit produced]:

 

Margin =

 

a. Sales price, minus

 

b. Direct labor ["standard labor"]+ standard materials [bOM, = Bill of Materials], minus

 

c. Indirect variable [variable by production output, that is] production costs, minus

 

d. Indirect fixed productions costs, allocated per unit.

 

ad b:

 

"Standard labor" is defined as what's demanded by each employee at each stage of the production process. [Visible and most likely also agreed with by employees].

 

BOM is budgeted prices of all components going into each unit [based on model and configuration]

Including expected warranty costs [based on historical statistical information, combined with expectations].

 

ad c:

 

Examples here:

 

Tooling depreciation [specificly mentioned before in this topic by Grant],

Electrical power to run the plant,

 

[so you get an idea of what it's about, conceptually].

 

ad d:

 

Examples here:

 

Maintenance costs PP&E,

Depreciation PP&E, ex. depreciation tooling, ref. above,

Real state taxes production properties [if any],

Insurance costs PP&E,

Toilet paper consumed at toilets in production area [Just to bent it in neon! : Not production output related],

Compensation site/plant management and administration.

 

- - - o 0 o - - -

 

If you think of it this way, there is no puzzle, wonder, mystery or - perhaps even worse : -  red flag in margins going up, when Tesla all of a sudden behaves like an old fashion glass bottle of Heinz ketchup with regard to spitting out Tesla vehicles.

 

In short, it's all about the built-in definition of COGS, and when the factories - with regard to production output -  are underperforming, a part of fixed costs included to COGS simply hit the P/L instead of being booked as part value of inventory of finished goods. [in "SMC language", it's called "under-absorption" of fixed costs included in calculated costs of unit produced [... which just hit the P&L].

[Gregmal]

https://seekingalpha.com/news/3401819-fbi-probe-tesla-intensifies

 

How dare they question Tesla. Fake news!

[SHDL]

https://seekingalpha.com/news/3401819-fbi-probe-tesla-intensifies

 

How dare they question Tesla. Fake news!

 

Fake, no, old, yes... 

[Grant]

John, as I mentioned earlier, the mystery isn't so much Model 3 margins. Yes they were higher than expected and guided, but we have no prior data to compare them to. The bigger mystery is S/X margins. There's also the mystery of the supplier rebates: how significant were they and where did they go in the income statement?

 

From the last 10-K:

 

Cost of automotive sales revenue includes direct parts, material and labor costs, manufacturing overhead, including depreciation costs of tooling and machinery, shipping and logistic costs, vehicle connectivity costs, allocations of electricity and infrastructure costs related to our Supercharger network, and reserves for estimated warranty expenses. Cost of automotive sales revenues also includes adjustments to warranty expense and charges to write down the carrying value of our inventory when it exceeds its estimated net realizable value and to provide for obsolete and on-hand inventory in excess of forecasted demand.

 

It doesn't look like depreciation of Fremont itself (or other shared infrastructure) is included in COGS. Just tooling and machinery.

 

Eric, in my opinion the ICEV manufacturers (and almost everyone else) underestimated the speed of which BEVs would replace ICEVs. They also probably still make more money on ICEVs.

[Dalal.Holdings]

"Authorities are homing in on whether the company made projections about its Model 3 production knowing it would be impossible to meet the goals, people close to the situation say."

 

https://www.wsj.com/articles/tesla-faces-deepening-criminal-probe-over-whether-it-misstated-production-figures-1540576636?mod=hp_lead_pos2

 

Guess they should've arrested Steve Jobs back in the day when he made "unrealistic projections" about the Macintosh in the early '80s...

 

As far as the margin question: maybe has to do with fixed costs. After all, Tesla is relatively vertically integrated (at least as far as battery production is concerned). Those fixed costs (COGS component) which were shared by the S and X now are also shared with the 3...additionally, battery pricing is a rapidly decreasing component of COGS (unlike say, steel). Tesla also does not do major product refreshes of its Models (which requires expensive retooling, etc), but rather incremental updates to its products which is also likely favorable to overall margin profile.

[CorpRaider]

Sounds like "Operating Leverage"

[John Hjorth]

John, as I mentioned earlier, the mystery isn't so much Model 3 margins. Yes they were higher than expected and guided, but we have no prior data to compare them to. The bigger mystery is S/X margins. There's also the mystery of the supplier rebates: how significant were they and where did they go in the income statement?

 

From the last 10-K:

 

Cost of automotive sales revenue includes direct parts, material and labor costs, manufacturing overhead, including depreciation costs of tooling and machinery, shipping and logistic costs, vehicle connectivity costs, allocations of electricity and infrastructure costs related to our Supercharger network, and reserves for estimated warranty expenses. Cost of automotive sales revenues also includes adjustments to warranty expense and charges to write down the carrying value of our inventory when it exceeds its estimated net realizable value and to provide for obsolete and on-hand inventory in excess of forecasted demand.

 

It doesn't look like depreciation of Fremont itself (or other shared infrastructure) is included in COGS. Just tooling and machinery.

 

Eric, in my opinion the ICEV manufacturers (and almost everyone else) underestimated the speed of which BEVs would replace ICEVs. They also probably still make more money on ICEVs.

 

Thanks, Grant,

 

Please study my highlights of your post, and please compare to my original post. In fact, to me, you're actually confirming my thesis for explanation of the margin changes in last quarter.

 

"Overhead" is not "just" "Overhead" - you have to distinguish between indirect variable production costs and indirect fixed production costs here.

 

To phrase it another way:

 

Please take a look at income statement in the 2018Q3 update letter - what do you have there of P&L  components with regard to Automotive sales?:

 

1. Automotive sales

2. Cost of revenues - Automotive sales

3. R&D

4. SG&A

5. Restructuring & other

6. Interest income

7. Interest expenses

8. Other income (expense) net

9. Taxes

 

- - - o 0 o - - -

 

Now, where do you think the fixed, not-production-output-related costs "land" in Tesla P&L? - There is only one answer!- And that is my only point with regard to reported margins - nothing else.

 

 

 

 

[Grant]

Manufacturing overhead and depreciation are COGS shared between all models of Tesla. My model suggests you can get back to expected Model 3 and S/X margins (15 and 26%) with a $280M reduction in COGS, most of which is applied to the Model 3. You need about $100M taken off of S/X COGS to get back to a 26% gross margin. In your model which confirms your thesis, how large is manufacturing overhead and depreciation?

[Dalal.Holdings]

Add Larry Ellison to the Tesla "fanboy" club:

 

https://www.bloomberg.com/news/articles/2018-10-26/oracle-s-ellison-says-tesla-is-his-second-biggest-investment

 

Remind me how unlikely it is a tech firm would buy this thing if it ever got to $20-$30B market cap? Seems like whatever upside exists to a short position on this only gets smaller and smaller.

 

“This guy is landing rockets,” Ellison said. “You know, he’s landing rockets on robot drone rafts in the ocean. And you’re saying he doesn’t know what he’s doing. Well, who else is landing rockets? You ever land a rocket on a robot drone? Who are you?”

  :D :D :D

[Grant]

Fake, no, old, yes...

 

The "new" bit appears to be:

 

In recent weeks' data-ipsquote-timestamp=' FBI agents have contacted former Tesla employees asking them for testimony in the criminal case. The former employees received subpoenas earlier in the probe, and FBI agents recently have sought to interview a number of them, the people said.[/quote']

 

The bigger question is why would the FBI leak this now?

[Dalal.Holdings]

Sounds like "Operating Leverage"

 

The leverage seems to be higher up on the income statement: at gross margins, not just operating margins.

[John Hjorth]

Grant,

 

Naturally, the calculated margins over the product mix matter - perhaps a lot.

 

That, however,  was not my ["isolated?] "mission" here. My "mission" with my posts today here in the Tesla topic was just to emphasize, that "a margin isen't just a margin". I was trying to emphasize, that increased margin in last quarter does not necessarily imply a "red flag", based on accounting considerations, and changes in production activity.