Tesla Gigafactory Investor Thread

[bwa]

Panasonic so far only invested a few million USD in the Gigafactory (GF). In any case, it won't be $3 billion (Tesla once intended to invest $2 billion on its own until 2020) in total:



http://www.reuters.com/article/2015/06/08/panasonic-autos-batteries-idUSL3N0YU31Z20150608

More importantly, Tesla's latest GF investment numbers don't make any sense to me and are very low in comparison to the $2 billion raised in 2014:



Emphasis in capital letters mine. Less cash used in last six months than in three months within the same period?

In any case, not more than $400 million (taking the highest numbers, six months should probably read three months in the latest 10-Q) will be spent by Tesla on the GF until January 2016, that is $206 million until the end of Q2 2015 plus ("up to" according to Tesla's 10-Q) 190 million in Q3 and Q4 2015.

At the same time, most if not all cash on hand will be gone by then in my opinion (with just the latest line of credit from June 2015 left as a cushion).

I can only see one result: A massive capital raise with new equity soon or very soon - because most of the $2 billion raised in 2014 (convertible bond) weren't used on the GF so far and the rest will be burned within six months on other items.

After working in one of Intel's "hotels" (they rent manufacturing space to "tool owners" who actually own the factory machines making chips) and many other big building construction projects (Facebook, Apple, HP, etc.), I'm almost certain the big costs for tooling the factory are yet to come. Less than a billion on their building seems reasonable in this picture; think house vs. furnishings, then take out scarcity of homes and add in super expensive furnishings, and you have a miniature factory analog. Those machines in factories are expensive, and in addition, there's a lot of building finish work yet to do, and a lot of factory facility work too like process plumbing and areas.

 

[jhm]

After working in one of Intel's "hotels" (they rent manufacturing space to "tool owners" who actually own the factory machines making chips) and many other big building construction projects (Facebook, Apple, HP, etc.), I'm almost certain the big costs for tooling the factory are yet to come. Less than a billion on their building seems reasonable in this picture; think house vs. furnishings, then take out scarcity of homes and add in super expensive furnishings, and you have a miniature factory analog. Those machines in factories are expensive, and in addition, there's a lot of building finish work yet to do, and a lot of factory facility work too like process plumbing and areas.

Awhile back, we had a post that broke down expenses. If memory serves me, $1.1B was for land and building, $0.3B for core systems, and the rest, $3.6B, for equipment. So I assume Tesla is fully on the hook for the first $1.4B and just a fraction of the $3.6B. Does anyone recall Panasonic's contribution was to be? My impression was that Panasonic and other partners would buy their own equipment to the tune of $2B.

The tricky thing has been trying to understand how quickly Tesla would need to burn its own cash. So my simple mental model has been that Tesla would need to set down about $1B for land and initial construction then roll out $40M per incremental GWh. So on 50 GWh this is a total investment of $3B. Under this model, the one quarter pilot facility is fully equipped at a total cost of $1.5B, that is, $1B upfront plus $0.5B for 12.5GWh.

Completion of the pilot plant is the crossover point for cash flow. If we assume that that Tesla is able to raise incremental cash of $40/kWh or more (and I do believe $50 is more realistic), then the pilot facility is able to generate at least $500M in incremental cash per year. This is sufficient to complete the remaining 37.5 GWh of the GF, an incremental investment of $1.5B, within 3 years. So completion of the pilot facility is the point at which the Gigafactory becomes self-funding.

In terms of timing, my hope is that the pilot is complete by the end of 2016. And completion of 50 GWh capacity is achieved by end of 2019. This implies linear growth in capacity one 4 years. My view is that if Tesla wants to accelerate growth, they will start launch multiple GF campuses.

Surely my basic model could be fine tuned, but I think the insight about the roll of the pilot facility is really key. That is, Tesla spends about half their full contribution on a quarter scale pilot, which is sufficient to self-fund the completion of the entire facility. It think this is a very Musk-like way to view the world. Investor view the $5B price tag and easily miss that, through partner contributions and self-funding post-pilot, the capital requirements to Tesla are quite low. Tesla only needed to raise $1.5 plus a margin for safety, which they accomplished last year. Now if they want to accelerate gigafactory expansion, faster than originally planned, this may require additional capital, but this is strictly for acceleration.

 

[chickensevil]

What I would be careful of is assuming margins that high from just the pilot plant. I believe they had always tried to stick to the "30% cheaper" target when the factory is fully operational, not from the outset. It may be that they get nice numbers like 50$ profit per kWh produced, but I wouldn't calculate everything on that as it seems generous. They did say that the TE part of the business would go gross profitable once they move into NV, but haven't locked in a short term number yet. I know we are hoping they can self fund the one factory so they can use extra money elsewhere, but we should really plan for it not hitting those types of numbers until 2018/2019.

 

[MitchJi]

Hi,

A large portion of the following savings will happen quickly. And I think it's safe to assume that they can negotiate the contracts for raw materials, so that they get the prices from the beginning.

Tesla GF Costs
It will also reduce the ‘per kWh’ cost of production of battery packs by over 30%. The automaker expects the factory to provide economies of scale and reduce production costs based on innovative manufacturing techniques, reduced logistic wastes, optimization of co-located processes, and lower overhead costs.

 

[jhm]

What I would be careful of is assuming margins that high from just the pilot plant. I believe they had always tried to stick to the "30% cheaper" target when the factory is fully operational, not from the outset. It may be that they get nice numbers like 50$ profit per kWh produced, but I wouldn't calculate everything on that as it seems generous. They did say that the TE part of the business would go gross profitable once they move into NV, but haven't locked in a short term number yet. I know we are hoping they can self fund the one factory so they can use extra money elsewhere, but we should really plan for it not hitting those types of numbers until 2018/2019.

My thinking here is that $60/kWh is 30% reduction on auto battery packs. $50/kWh is 20% on Powerpacks. $40/kWh is an extra bit of conservatism on the two preceeding numbers. Also the one quarter pilot will have been operating for about 12 months before it reaches its 12.5 GWh capacity in 2017. So hopefully that will be sufficient experience and scale to get most of the efficiencies hoped for at full scale.

 

[tftf]

After working in one of Intel's "hotels" (they rent manufacturing space to "tool owners" who actually own the factory machines making chips) and many other big building construction projects (Facebook, Apple, HP, etc.), I'm almost certain the big costs for tooling the factory are yet to come. Less than a billion on their building seems reasonable in this picture; think house vs. furnishings, then take out scarcity of homes and add in super expensive furnishings, and you have a miniature factory analog. Those machines in factories are expensive, and in addition, there's a lot of building finish work yet to do, and a lot of factory facility work too like process plumbing and areas.

The model calculations (done for subsidy calculations in Nevada by third parties) indeed have the building/construction part (shell) at "only" $1bn or $1.1bn until late 2017.

That is not the issue.

The issue is that Tesla will have soon spent (because they had some cash on hand before the early 2014 capital raise) about 90% of the $2bn+ convertible bond on other items and not on the Gigafactory project.

That is just the beginning because R&D on the Model3 will soon kick in.

The biggest sell-side bull analyst (Adam Jonas, Morgan Stanley) just issued a new report that another $14 bn (!) of cap-ex requirements for 2015-2020 are needed.

We will see where all this money is coming from and how much is needed to complete the first battery factory - even if Panasonic chips in 30-40% of the total until 2020.

I predicted a huge equity raise in earlier posts - the recent rounded of $650 m from last week was just the appetizer in my opinion.

Tesla will need to raise billions and billions until 2020 - since they already have convertible bonds and a line of credit (collaterals pledged) outstanding the only way do it will be once again new equity offerings imho.


PS: When will Tesla fix their 10-Q in relation to GF expenses or is everything ok with it?

 

[Krugerrand]

I predicted a huge equity raise in earlier posts...

You predicted nothing, but I did predict you'd lay claim to it. The equity raise had already been stated by Elon/Tesla at the ER. All you did was repeat what everyone who listened to the call had heard themselves.

 

[tftf]

You predicted nothing, but I did predict you'd lay claim to it. The equity raise had already been stated by Elon/Tesla at the ER. All you did was repeat what everyone who listened to the call had heard themselves.

I wrote the $650 million from last week was just an appetizer. It was not the huge equity raise. Tesla will need (fresh) billions in the next 6-18 months imho. That was my prediction and it hasn't happened yet - except for the appetizer (first of many rounds, maybe they wait until the new CFO is in place for the main course).

As I wrote above, even uber-bullish sell-side analysts like Adam Jonas predict a cap-ex of about $14bn until 2020.

Either you...

a) disagree with him (e.g. Tesla will need much less, but where will the growth come from then in a cap-ex intensive industry like cars ?)
b) think Tesla can magically generate these funds with internal cash-flows from now on (14 billion ?!)
c) think Tesla can tap other sources again (new bonds, new line of credits, new partnerships...)

or otherwise Tesla will have massive equity offerings for the completion of the Gigafactory and other projects like the Model 3.

 

[Krugerrand]

I wrote the $650 million from last week was just an appetizer. It was not the huge equity raise. Tesla will need (fresh) billions in the next 6-18 months imho. That was my prediction and it hasn't happened yet - except for the appetizer (first of many rounds, maybe they wait until the new CFO is in place for the main course).

Uh, huh. You do realize people can scroll back and reread old posts, right? They can even view old posts in other threads.

 

[tftf]

PS: When will Tesla fix their 10-Q in relation to GF expenses or is everything ok with it?

If readers missed the details of my question here they are again:

Tesla's latest Gigafactory investment numbers don't make any sense to me (and are very low in comparison to the $2+ billion raised in early 2014):

Latest 10-Q (just released): "During the SIX months ended June 30, 2015, we used cash of $54.6 million towards the construction of this project and expect to spend up to $300 million for the full year."

Former 10-Q for Q1 2015: "During the THREE months ended March 31, 2015, we used cash of $56.3 million towards the construction of this project and expect to spend up to $300 million for the full year."

Emphasis in capital letters mine. Less cash was used in last six months than in three months within the same period?

Even if Tesla meanwhile received cash-backs from Nevada I don't see how basic accounting rules would allow to directly subtract cash expenses and somehow arrive at a lower number (and I don't see other explanations).

There was also no 10-Q/A or similar filed so far.

 

[RobStark]

Rare Earth And Bacanora Sign Lithium Supply Deal With Tesla Motors

On condition that the Sonora project reaches certain performance milestones in the next two years, Tesla will buy lithium hydroxide to feed the manufacturing of batteries at its Gigafactory in Nevada. One of the conditions will be that the Sonora project can supply lithium hydroxide in accordance with volumes and time frames which Tesla will determine.

The supply deal is for an initial five years, starting from when Tesla makes its first order, with an option for this to be extended by another five years.

"The selection of the Sonora lithium project as one of the lithium suppliers to the Tesla Gigafactory is a landmark transaction that will support the development and commercialization of the Sonora lithium project. " said Rare Earth Chairman David Lenigas.

http://www.lse.co.uk/AllNews.asp?cod...h_Tesla_Motors

 

[WarpedOne]

Share price (BCN.L) just went from 70 over 90. +30% overnight.

 

[32no]

I'm a bit late to the party, but on the last page tftf claimed that Morgan Stanley says Tesla needs to spend $14 billion in Capital Expenditures through 2020 to meet their goals. This is false, and I am pretty sure tftf deliberately left out a caveat about that $14 billion, because I have seen him post it correctly before. That $14 billion is for Capital expenditures AND R&D. I'd say about $7-$8 billion of that is Capital expenditures.

 

[Benz]

Seeking Alpha:

Tesla Motors signs lithium deal with Bacanora

• Tesla Motors enters a long-term lithium supply agreement with Bacanora.
• The deal includes performance milestones which must be met by Bacanora before lithium hydroxide is purchased to supply Tesla's Gigafactory.

 

[Johan]

BCN.V: Summary for BACANORA MINERALS LTD- Yahoo! Finance

 

[Ampster]

I'm a bit late to the party, but on the last page tftf claimed that Morgan Stanley says Tesla needs to spend $14 billion in Capital Expenditures through 2020 to meet their goals. This is false, and I am pretty sure tftf deliberately left out a caveat about that $14 billion, because I have seen him post it correctly before. That $14 billion is for Capital expenditures AND R&D. I'd say about $7-$8 billion of that is Capital expenditures.

In my mind R & D ARE capitalized so in a broad sense the total is still $14 billion of cash to raise to cover those needs.

 

[Johan]

Bacanora up 43%. Did anyone buy just for sh**ts?

 

[Cosmacelf]

Interesting that Bacanora is still an exploratory stage company. Looks like they have proven reserves, but have not yet set up actual mining operations. I wonder how quickly a company can start producing lithium hydroxide starting from scratch?

 

[Krugerrand]

Wasn't there a second company also involved: Rare Earth Minerals PLC?

 

[tftf]

In my mind R & D ARE capitalized so in a broad sense the total is still $14 billion of cash to raise to cover those needs.

Yes, and I didn't leave out the R&D deliberately, $14 bn in cash are needed in total including R&D, I simply didn't spell it out in every post. Here's the exact quote from the report so 32no hopefully has a clear piece of mind:

Page 2:

"Tesla's pace of capex and R&D spending through 2020 vastly exceeds its current size and production footprint. We have modeled $14bn of combined spending from 2015 through 2020, for only 2 model families (Model S/X and Model 3), 1 assembly factory, and 1 very large battery factory. On a per-unit basis, this level of spending is nearly 10x that of Ford. "

I find it somewhat funny when my forum posts are scrutinized in every detail, yet when I repeatedly asked about/pointed out errors (or at least unclear numbers) in Tesla's 10-Q (Gigafactory expenses for Q1 and Q2 2015, see my previous posts in this thread) I get no answer.

So may I ask for a last time:

1. Do you think Tesla's latest 10-Q contains no errors in the quoted sections related to the GF expenses? After all, this is a public company with a huge market cap, reported numbers in SEC filings matter.

Latest 10-Q (just released): "During the SIX months ended June 30, 2015, we used cash of $54.6 million towards the construction of this project and expect to spend up to $300 million for the full year."

Former 10-Q for Q1 2015: "During the THREE months ended March 31, 2015, we used cash of $56.3 million towards the construction of this project and expect to spend up to $300 million for the full year."

2. Even if the 10-Q contains no errors, don't you find it strange that about 80-90% of the funds ($2bn+) raised in early 2014 were spent on other items (ex-Gigafactory), especially since new money was again raised in 2015 (twice if one adds the line of credit)?