"Tesla. A Much Needed Reality Check (in depth)" - Julian Cox's Expert Analysis Posted

[aznt1217]

Yes, how can they not. Doing a parts explosion on the model S should make it very clear. You can't use Maserati as a comparison due to their aggressive 2010-14 product plans. That and it's an entirely different animal in regards to resource sharing between Ferrari and Fiat. For all we know, Maseratis R+D is shared resources for the other companies and can be a corporate tax play.

Yes, the battery is the most expensive part, but you do realize that capacities to make the number of cells required is there now between Samsung and Panasonic in 2015 there will be enough using current trajectories. Tesla has massive economies of scale here.

Not too sure where you are with the warranty claims and winter, but... It's been a little over a year and quite a few years with roadsters. Nothing major yet...

Nobody is going to change your mind but your comparisons are faulty and the rather pessimistic view on technological development is proven flawed.

 

[brianman]

The major cost component of all Tesla models is the battery pack. Right now these battery cells are in tight supply and there is no real reason to believe prices will come down. Tesla has no comparative advantage here, its all in the hands of Samsung and Panasonic.

Are you really saying that Tesla has no advantage over other auto competitors when it comes to locking in future production from these (and other) battery suppliers? I find it (a) hard to believe and (b) hard to believe that anybody believes that.

After "not taking Tesla seriously enough", I think many suppliers have learned their lesson. And those that didn't need to learn it (like Panasonic) are likely to remember that they made the right call and strengthen the relationship with sweeter deals going forward.

 

[Convert2013]

Panasonic and Samsung get only the specs from Tesla (OEM) to design and build the battery assembly, but integrating it into the vehicle as a larger system as a whole (including battery management in one high level word for simplicity's sake) is not a supplier core competency, even though they are provided some bits and pieces of the system environment for battery verification purposes. That integration capability will take time for GM, Ford, BMW to replicate. They will eventually do it as great engineering talent is available at all major OEMs. In my experience, my best guess is it will take 5 to 7 years minimum for the traditional auto makers to catch up to that type of systems integration capability, considering the various new revolutionary subsystems in the Model S. The assumption is that Tesla kept its subsystem-integration core competency capability in house, which is most likely.

 

[Soflason]

After "not taking Tesla seriously enough", I think many suppliers have learned their lesson. And those that didn't need to learn it (like Panasonic) are likely to remember that they made the right call and strengthen the relationship with sweeter deals going forward.

Brianman, impressive... your prediction just came to fruition:
http://www.forbes.com/sites/markrogowsky/2013/10/30/teslas-huge-battery-deal-suggests-sales-could-quintuple-within-4-years/?partner=yahootix

 

[brianman]

Brianman, impressive... your prediction just came to fruition:
http://www.forbes.com/sites/markrogowsky/2013/10/30/teslas-huge-battery-deal-suggests-sales-could-quintuple-within-4-years/?partner=yahootix

The numbers tell the story:

What’s remarkable about the Panasonic-Tesla agreement is that the entire lithium-ion cylindrical battery industry (the type Tesla uses) was consuming 660 million cells a year and this agreement represents 500 million annually for Tesla alone.

500/660 > 75%
Dominant consumer? Kinda.

 

[blakegallagher]

The numbers tell the story:

500/660 > 75%
Dominant consumer? Kinda.

This would imply that Panasonic will be adding a lot of production right? I can not imagine they currently have anywhere near 75 percent of the "entire lithium-ion cylindrical battery industry (the type Tesla uses)" production capacity.

Also worth noting that Panasonic took a 10 percent stake in Tesla a few years ago. I think they will get great deals from Panasonic and keep that partnership for a long time.

 

[brianman]

This would imply that Panasonic will be adding a lot of production right? I can not imagine they currently have anywhere near 75 percent of the "entire lithium-ion cylindrical battery industry (the type Tesla uses)" production capacity.

Also worth noting that Panasonic took a 10 percent stake in Tesla a few years ago. I think they will get great deals from Panasonic and keep that partnership for a long time.

Agreed. Seems like their stock should bump a bit on this news. But who knows these days.

 

[kevin99]

Julian, Bravo on your recent article! How do we get it posted on SA? The short blitz campaign on SA is really rattling me (not the mention the market.) Q3 earnings report ought to put that to a stop, but would like to see more bull articles like this hit the mainstream!

I am catching up with the discussion here.

We had a bad experience with SA and looks like it is not gonna to happen.

As with Mercury News, no getting back after a couple of contacts shuffling.

So this is a crowd-effort project. If you can help to get the article on other media, that will balance the short blitz out there. The short theory is quite pressing recently.

I see a lot of medias choose to highlight or rephrase important viewpoint of other sources, perhaps that is way to go. Here is an example:

Tesla Motors Inc (TSLA) Is Way Overvalued: Damodaran

- - - Updated - - -

@FluxCap and Julian - I think what some people are upset about is that this article could have been great, but that it has at least a few simple (yet major) errors that completely negate anything positive in the rest of the article.


As it stands now you have people in this thread questioning the assumptions, as well as on Reddit. I think that everybody is going to use these flaws to completely dismiss the rest of the article; and that is what is happening.

We are quick to point out the tiniest errors in the articles written by the shorts, but then are asking others to ignore some big mistakes in an article written by a long.

Julian put in a lot of effort to write this article, but promoting it to the outside world does us longs a disservice. The shorts will point out the obvious flaws and dismiss the rest of the article. Which is the exact same treatment we have been giving the shorts with their articles.

I can't agree with this. Knowing this article covers a lot of territory and depth, I wouldn't say readers are upset because there is a flaw here and there.

Just like how we treat the short theory, I see many people longs do not hesitate to point out some valid points when they see them in the short theory.

There are too many gems in JC's article that a notion of dismissal because a flaw is laughable.

TSLA is a fascinating and difficult subject. It is inevitable to have flaws in a analysis like this. As in a Chinese saying: The truth becomes clearer with more debates. So let's debate more and don't be shy in just pointing out the flaws, contribute to it to build up the theory.

 

[DonPedro]

I couldn't disagree more. This is a war of credibility, and you win credibility by avoiding hyperbole and flawed arguments.

Why not wait until someone produces an article that brings up the good points from this one, without including complete misunderstandings and flaws? If there are genuinely new arguments put forward here - do we really want them to be DOA because the article in which they are presented shows that the author does not understand basic concepts like cost of capital and dilution?

 

[kevin99]

I couldn't disagree more. This is a war of credibility, and you win credibility by avoiding hyperbole and flawed arguments.

Why not wait until someone produces an article that brings up the good points from this one, without including complete misunderstandings and flaws? If there are genuinely new arguments put forward here - do we really want them to be DOA because the article in which they are presented shows that the author does not understand basic concepts like cost of capital and dilution?

Doa or not, it is as subjective as other gushing praises, which is plenty.

As there are many arguments laid out independently, one flaw, if there is one (I am no domain expert but I do enjoy the debates) does not necessarily destroy others.

And who is to judge "someone produces an article that brings up the good points from this one, without including complete misunderstandings and flaw"? Are we counting on you as the authority to declare that when such piece arrives?

 

[DonPedro]

And who is to judge "someone produces an article that brings up the good points from this one, without including complete misunderstandings and flaw"? Are we counting on you as the authority to declare that when such piece arrives?

I think that the flaws have been collectively pointed out by many key contributors on this forum. If you write up the remaining points, you will have something like what I am describing.

 

[bonnie]

Kevin, DonPedro has a right to state his opinion. The aggressive tone of "are we counting on you as the authority to declare" is uncalled for. Please tone it down. Everyone here is speaking from a place of good intent.

 

[FluxCap]

As Bonnie said, we discussed this for twelve good pages last week and seemed to reach a reasonable set of conclusions. Let's try to leave this one be for now, and remain in good spirits. There will be plenty more to write about post-earnings.

Happy Halloween, everyone!

 

[kevin99]

Kevin, DonPedro has a right to state his opinion. The aggressive tone of "are we counting on you as the authority to declare" is uncalled for. Please tone it down. Everyone here is speaking from a place of good intent.

Well I was sensing the absurdity of DOA and tone like that.

 

[DonPedro]

Here is a direct attempt at a contribution: I've imported the article into Word, gone through it, and marked up the problematic parts:

Dropbox - Reality check article markup.docx

I have tried to explain the problem in each part and why it has to be taken out or changed. The main thing is that you can do that and still have virtually all the argument intact - there is really no good reason to keep these errors in there. (Especially since they are really bad, as in you would fail Finance 101 if you wrote something like that).

As for the DOA, that was meant quite literally. Both Sleepyhead and I had the same reaction - we read until we got to the cost of capital bit, and then we stopped reading because we lost all faith in it. Now that I have read the rest, I think it has some really, really great points. Fixing the errors means getting those points out there with credibility, rather than the opposite.

I hope this can be a contribution to the article's success.

 

[Convert2013]

Doa or not, it is as subjective as other gushing praises, which is plenty.

As there are many arguments laid out independently, one flaw, if there is one (I am no domain expert but I do enjoy the debates) does not necessarily destroy others.

And who is to judge "someone produces an article that brings up the good points from this one, without including complete misunderstandings and flaw"? Are we counting on you as the authority to declare that when such piece arrives?

I like JC's article. No analysis is perfect, even in engineering. JC's article is a strong point of view backed up with arguments. Every article on Tesla that I've read so far be it long or short had inaccuracies in it.
BTW I have contacted two columnists at LA times, no response so far. Now will have to try somewhere else...

 

[kevin99]

Here is a direct attempt at a contribution: I've imported the article into Word, gone through it, and marked up the problematic parts:

Dropbox - Reality check article markup.docx

I have tried to explain the problem in each part and why it has to be taken out or changed. The main thing is that you can do that and still have virtually all the argument intact - there is really no good reason to keep these errors in there. (Especially since they are really bad, as in you would fail Finance 101 if you wrote something like that).

As for the DOA, that was meant quite literally. Both Sleepyhead and I had the same reaction - we read until we got to the cost of capital bit, and then we stopped reading because we lost all faith in it. Now that I have read the rest, I think it has some really, really great points. Fixing the errors means getting those points out there with credibility, rather than the opposite.

I hope this can be a contribution to the article's success.

Much appreciated.

 

[jeffhre]

In which case he is wrong. It is just a matter of making assumptions that reflect your beliefs.

Then we are all wrong, since that encapsulates what we are all doing. How can you peg the mid to long term CoC of a company that grew revenues 1500% in one year, based on it's one and only, newly available model, and assume there is some possible potential for predicting present value of future cash flows. WITHOUT assumptions that reflect your beliefs? Rather than say preposterous, I will say that each person will contribute observations from their point of view that will fortify a model that is bound to be wrong, no matter how expertly conceived. Though of course, at some point, someone will create an outlier of a model, or probably several, and discard the ones that prove to be false. They will then at a future point be hailed as the hero who nailed Telsa's future values, outputs and EPS. Which will be as exactly right as the broken clock that tells the time twice a day.

 

[DonPedro]

Then we are all wrong, since that encapsulates what we are all doing. How can you peg the mid to long term CoC of a company that grew revenues 1500% in one year, based on it's one and only, newly available model, and assume there is some possible potential for predicting present value of future cash flows. WITHOUT assumptions that reflect your beliefs? Rather than say preposterous, I will say that each person will contribute observations from their point of view that will fortify a model that is bound to be wrong, no matter how expertly conceived. Though of course, at some point, someone will create an outlier of a model, or probably several, and discard the ones that prove to be false. They will then at a future point be hailed as the hero who nailed Telsa's future values, outputs and EPS. Which will be as exactly right as the broken clock that tells the time twice a day.

The statement you are quoting was in response to the following statement:

I think wcalvin is saying the latter: quantitative models don't account for the qualitative factors that are, in this case, driving the price.

What I am saying is that the qualitative factors wcalvin refers to can be reflected in the quantitative models by baking them into the assumptions. Therefore he is wrong that the quantitative models do not account for them. They do - to the extent that you make assumptions (e.g. about sales growth or margins) that reflect your judgment of these factors.

Whether or not you use a quantitative model, you have to assume SOMETHING if you want to have an opinion on the share price. Might as well be a bit rigorous about what you do with your assumptions, no?

The long term cost of capital, that you cite as an example, is in fact the EASIEST part to make a reasonable assumption about. Reasonable assumptions will be in the range 7-12%, as it is for most companies that aren't too heavily indebted (small and big, profitable or unprofitable, growing or declining). The point is that this is about the cost of money, not the cost of the business model. An investor doesn't care whether a 10% return comes from this or that stock (except that he would prefer to have that return with as low beta as possible).

The assumptions that cause the biggest challenge for valuing Tesla are in my opinion:
- Sales projections
- Margin projections
- Dilution

I frankly do not understand this advocacy for not using quantitative models to judge an investment. See if you can find a way of disagreeing with this statement:

For any investment, there exists a price level at which it becomes unprofitable.

Unless you disagree: How are you going to determine what is "too expensive" for TSLA, if you are opposed to quantitative models in general?