As the title suggests, in this article I will outline five of the major reasons for which I'm short the stock of Tesla Motors (TSLA). I'm sure that many will disagree, but I hope that at a minimum it will provide food for thought and discussion.
There are several possible reasons to short a stock: a belief that the company is a scam (like Enron and many BB companies), a belief that it will ultimately fail due to outdated products or poor management (e.g. Kodak or Kmart), or a belief that its market valuation is simply too high. The valuation argument can take two forms: over-valued relative to its peers, or a valuation that replaces the appeal of the company for any realistic analysis of its economic prospects. In an earlier article I argued that this latter category applies to Amazon's (AMZN) current valuation, and I believe that it's also the case with TSLA.
One way to see this is to ask, "At what price would TSLA proponents think the stock fairly-valued (let alone over-valued)?" The answer in my experience is always "much higher". If the stock is trading at $25, then fair value is $50; but when it gets to $50 it's $100, then $200 etc. In reality, there's no limit because "how can you put a price on being part of a revolution?" This attitude is what, in my opinion, gives rise to a shorting opportunity.
Now of course, substituting the user's perspective for that of an investor's is much more understandable in TSLA's case than in AMZN's, as its product is by all accounts, sexy, safe and high-performing. The company's flagship Model S sedan (currently the only car the company is producing) earned Consumer Reports' highest ever rating, scoring 99 out of 100 points! Moreover, Tesla is said to convert 1 out of every 4 test-drivers to a sale (though of course test-drivers are a highly self-selected group). Moreover, the technology is genuinely disruptive, and TSLA's CEO Elon Musk is widely-regarded as an engineering genius. There are other positives as well, but it's not necessary to rehash them all as they have been well presented across the web.
There have also been many articles here on SA arguing that TSLA is fundamentally over-valued. I tend to agree in general, but in the following my focus will be on the factors that have led to this over-valuation, and what could change in the future such that the company will no longer be awarded such stratospheric multiples. The arguments are broadly captured in 5 categories, which I discuss below. But to help us get started, here are Yahoo's summary valuation stats at Tuesday's closing price of $149.58:
Before proceeding, let me also state upfront that I recognize that any analysis of a company by outsiders is often dismissed by the rebuttal "who are you to judge the people actually doing the work and creating progress?" Yet that retort goes to my basic idea that there's an investing opportunity here. Investors - whose real job should be one of asset-allocation - have become so enamored with the product that they're now essentially cheerleaders rather than cold, hard decision-makers. Or to put it another way, the idea that investors (from both the short and the long side) are making judgments about companies in which they're not actually involved is almost always true. That's just the nature of our job wherein we try to decide how capital can be most productively employed.
With that said, here are five major reasons why I'm short TSLA, presented roughly in order of importance.
#1 The Company's Promotional Approach & Investor Sentiment
In my view the company tends to be overly-promotional, perhaps with a goal of propping up the stock price in the short term. In a way it's understandable, since there was a time (though no longer) where the company's solvency was in question. As a result, boosting the stock price was critical to raising equity at the best prices possible. But somehow it seems that approach is still with the company (for another possible motivation see also #5 below).
Personally I'm much more comfortable investing in a company that takes Warren Buffett's approach of not caring (directly) about the stock price. Yet that's certainly not TSLA.
Similarly, instead of announcing news when it becomes available, the company dribbles it out for maximum effect, often pre-announcing announcements! A few examples:
Now of course all of this could simply be put down to personality, i.e. that Musk has a bit of "Tony Stark" in him. But it gets more serious when the same approach is used to present the financials. After switching from a sales model to a combined sales & lease model, Tesla tried to have its cake and eat it too by promoting what a prominent critic here at SA refers to as "fairytale" non-GAAP accounting numbers. I'll discuss the substantive liability issue in point 2 below, but for now I just want to highlight the promotional aspect of the company's behavior.
The most recent letter to shareholders does everything it can to dissuade shareholders from considering GAAP accounting numbers; from the initial paragraph which only presents non-GAAP numbers to the final section, "outlook" which likewise is given in exclusively non-GAAP terms. Yet GAAP accounting is the result of 100's of years of industry experience trying to come up with standards that best present a company's financials (including the basic matching principle whereby one tries to align the timing of revenue and expense recognition). A willingness to flout that in order to present shareholders with more appealing numbers is a noteworthy warning sign in my personal opinion.
Nonetheless, all the promotion (plus a sexy product) has succeeded in whipping up a frenzy among supporters, many of whom are also shareholders. One can see it by any number of (anecdotal) measures: the number of page views and comments generated by any article on TSLA, the forums where proud buyers post photos of their new babies (Tesla S's), the quasi-rock star status of Mr. Musk, the gleeful sparring with short-sellers, etc.
The result of this crescendo in investor sentiment has been (in my opinion) to push the stock price way ahead of itself. Of course stocks can remain overvalued indefinitely, but in my experience, when the price is partially driven by an overly promotional approach, future promotion tends to lose its effect and, simultaneously, the company becomes more vulnerable to selling off, not only on genuinely bad news, but also for simply meeting expectations. This potential vulnerability is the first reason why I'm currently short TSLA.
#2 Lease Liabilities
That - after less than a year of sales - the company saw the need to help consumers buy the car by creating a lease program to lower initial payments might say something about the strength of underlying demand. Perhaps, but that's not the point I want to focus on here.
My concern is the liabilities that the company is assuming in launching this program and the risk that this entails. In point #1 I referenced a recent SA article ("Tesla's Non-GAAP Fairy Tale") which does an excellent job of clearly and concisely explaining the full set of facts and context - so I highly recommend that readers consult it, plus the author's many replies in the comment section.
In the following I'll simply summarize the gist of the problem and how it contributes to the short case.
Consider TSLA's own words in the risk factors from its latest 10-Q:
[..]Furthermore, while we do not assume any credit risk related to the customer, we are exposed to the risk that the vehicles' resale value may be lower than our estimates and the volume of vehicles returned to us may be higher than our estimates. Currently, there is only a very limited secondary market for our electric vehicles in particular, and electric vehicles in general, on which to base our estimates, and such a secondary market may not develop in the future. Our residual value and return volume estimates could prove to be incorrect, either of which could harm our financial condition and operating results.
For a company that's barely profitable, particularly as subsidies are phased-out, such increases in costs could represent a significant burden, especially since 30% of buyers last quarter took advantage of the program and the company expects that percentage to increase.
But the assumption of a high resale value isn't just problematic in the sense of adding unknown liabilities to the balance sheet, it also doesn't quite fit the rest of the bullish narrative.
On the one hand we're told that just as the Roadster was, so too the Model S is still simply a step along the product development path, a path which will come to full fruition in several years with much improved batteries, technology and manufacturing capabilities. The promise of the Gen 3 car rests squarely on this development. Yet on the other hand, we're supposed to believe that the Model S will retain its value better than any luxury car on the market.
If everything is going to be so much improved in three or four years, won't that make today's Model S a technological throwback or a collector's item? And if the company has to retrofit the cars it takes in on lease with new state of the art batteries and computers in order to resell them, won't that be a much bigger cost than is currently being accounted for?
Though I don't actually agree with the bulls' case about the speed with which the technology will improve, I do think that high resale values won't hold as competition from newer Teslas as well as from other manufacturers emerges. This then sets TSLA up for massive financial restatement(s) down the road, especially if more and more customers avail themselves of the lease option and then turn the cars in after three years.
This potentially growing liability is the second reason why I'm short TSLA. Of course, since this whole issue won't be resolved for at least three years and likely will take longer, it may not have any short term effect on the stock's price. But longer term it's definitely worth keeping an eye on.
#3 The Early Adopter / Rapid Scaling Paradigm
An important part of the TSLA bullish story is that - in re-inventing the car business - the company has modeled itself on companies selling new technologies. As such, one can't compare it to stodgy old car companies in particular, nor even to industrial companies in general.
Instead, (as I understand it) the paradigm we're supposed to accept is best represented by Apple (AAPL) when it introduced its disruptive iPods, iPhones and iPads. This paradigm has (at least) two essential premises. First, that one should target affluent first adopters in order to get the product out. This allows one to get valuable feedback from a sympathetic audience and to iron out any issues before one massively scales up sales. Second, if the product is met with success from first-adopters, scaling will be simple, and a market will have been primed such that exponential growth is all but assured.
Conversely, eschewing this approach is tantamount to failing before one starts. Indeed this is a reason often given (by TSLA bulls) for why Chevy's Volt (PHEV) and Nissan's Leaf (EV) programs are nothing but old thinking with no value other than compliance. As such, they argue, no legacy ICE manufacturer can mount real competition to TSLA.
While I agree that there's much to be learned from the success of innovative companies, and that many older manufacturers don't sufficiently seek out the benefits of cross-pollination from other industries, I'm still quite skeptical that the paradigm put forth by TSLA's advocates is appropriate. Here's a sampling of my doubts:
First off, I think people fail to appreciate just how quickly the three Apple products were adopted and thus how unusual they were. This site lists the early quarterly sales and includes the following cumulative sales graph for the iPod, iPhone and iPad:
And here's a graph of the quarterly unit sales of the worst-selling of the three, the iPod:
Within two years, sales of the iPod (both quarterly and ttm) were almost an order of magnitude higher, and within five years they were two orders of magnitude higher! We can't yet draw a comparable chart for model S sales (or total TSLA vehicle sales), but we can try to appreciate what it would mean for the cars to sell like the worst of Apple's three disruptive products. Taking the Oct-Dec 2012 period as the first real commercial quarter for the model S, and using a combination of actual sales and projections to get a twelve month figure, we get initial quarterly and ttm sales of ~2,400 and ~17,500 units respectively.
If the company wants to replicate the iPod's growth path, it will have to sell ~24,000 units in the last quarter of 2014 and ~240,000 units in the last quarter of 2017. Similarly its trailing twelve month sales at the end of Sep 2015 will have to be ~175,000 vehicles and ~1,750,000 in the twelve months ending Sep 2018!
That this is impossible is not in any way a knock on TSLA - it simply goes to show that the expectations of many of today's investors are ludicrously unreal. (As a reference point, the company is projecting demand for 2014 at 40,000 units according to a recent conference call. Assuming that they could meet this hypothetical demand, that would put the quarterly sales for the last quarter of 2014 at ~12,000 units, only half of Apple's growth. And it's debatable whether they can scale manufacturing to even achieve that number, assuming the demand actually exists.)
Modeling growth on a once in a generation firm like Apple is to set expectations way too high; eventually leading to future disappointment. (For anyone interested in pursuing this line of thought further, I'd point them towards another article of mine where I compare the growth and performance of MSFT to that of AMZN.)
But the new proposed paradigm doesn't only fail because it sets completely unattainable target growth rates, there's also the issue of scaling.
While Tesla's cars feature a lot of software - and by all accounts it's well done and instrumental to the car's appeal - the differences between a car and an iPod are too substantial for the paradigm to work. Once software has been debugged and optimized, there's no issue in replicating it. Moreover, electronic goods such as the iPod and iPhone have benefited from outsourced high volume manufacturing for decades now. Thanks to the history of mass production of electronics from the Sony Walkman, to computers and laptops, to video game consoles, etc. there's an industrial ecosystem ready to take a new product like the iPod and build it very quickly. None of this is true for TSLA.
Yet to hear TSLA bulls tell it, once the first-adopter stage is complete, scaling is a "no-brainer". Just run three shifts per day and use 100% of the factory floor space and you'll be able to scale output by 15x without any issues!
Of course there are many reasons why this can't be done so simply, but to play devil's advocate let's accept the proposition and ask the following question: If indeed scaling is a "no-brainer", AND demand for the Model S is growing by leaps and bounds, AND the gross margins for the product are on the order of 25%, why hasn't the company already ramped up production significantly?
The accepted answer is that ~1% of the car's components create the bottleneck and if they could simply procure that 1%, production would be significantly higher. But is that credible? Let's say the average selling price of a Model S is $110,000 (it was $107,000 last quarter). With gross margins of 25%, the cost of the car is ~$82,500. If the 1% bottleneck is proportionate in costs (i.e. assume by 1% they don't mean that there's a scarcity of batteries), then they're holding up substantial extra production in the name of ~$825 worth of parts. Let's say - as the bulls do - that real annual demand is for 40,000+ vehicles right now. Then by obtaining $16.5M more in parts (20,000x$825), the company could generate $2.2 billion(!) more in sales. In my manufacturing experience, paying a 25% expediting fee normally got one all the extra parts one could use, but in this case even paying 200%, or 500% would make economic sense - if indeed the numbers and situation were as the bulls say. To me something is very wrong with this story, since as of late, TSLA hasn't ramped up production significantly.
So the proposed paradigm doesn't work on growth rates and scalability, but here's yet another aspect to consider. Automobiles, both as a product and as a market, are much more regulated than are computers and electronics. As a free market proponent, I wish this weren't true, but in today's world it is - and it has negative ramifications towards any company trying to think and work outside of the "system".
For instance, one criticism often levied against "dinosaur" automotive firms is that they're always tweaking inessential parts such that they never truly benefit from the economies of scale. While there may be something to this, it also overlooks the fact that the automotive industry is subject to a host of (arbitrary?) rules and regulations, including constantly updated fuel efficiency and safety requirements. These changes often cause automobile companies to make changes for reasons other than their own design and manufacturing preferences. As a new manufacturer, Tesla may not be too impacted by this initially, but it's likely that as it becomes a more established player, it will be. Similar things can be said about the battle to sell cars directly (where many state laws require consumers to go through dealers), or the goal of staying union-free in a pro union state like CA.
Finally, there's the issue of the benefits accrued from affluent and educated first adopters. In Apple's case, there was a type of beta-testing performed by its initial users. The types of problems that these advanced users found were relatively esoteric and difficult to find, making their involvement particularly useful.
But this may be less relevant with a car, and from what I've seen there's also a downside, viz. that TSLA aficionados are likely to give the company a pass where more mundane users wouldn't. That is to say, early adopters treat the car as an interesting science project, but once that demographic is exhausted, others won't be as forgiving. At that point it's unlikely the car will retain its "perfect vehicle" image. (Please note that I'm not saying that the Model S has many mechanical problems, it doesn't. But like most cars it has some, and some of its mystique will evaporate once it's no longer exclusively in the hands of fans.) To get a sense of what I mean, refer to some of the Tesla forums. For example consider the context this user feels obligated to post: "Luckily it is less than what I expect from a new car company, with their first car, from their newly opened factory, using cutting edge technology. I have no complaints." Will regular owners have the same perspective?
My third reason for shorting TSLA is that while I believe TSLA has a bright future ahead of it, it will never come close to achieving the growth rates nor the easy scaling and mass market adoption that is currently priced into the stock.
Many people have discussed the issue of new competition coming from the likes of BMW and others. I don't have any additional insight into this other than agreeing it will be a factor. But I'd also add that though there are disadvantages to being a "legacy" ICE manufacturer, there are also ways that these companies' existing infrastructure could be used to mount a competitive threat to TSLA. One idea that got my attention was when BMW said it might lend ICE vehicles to purchasers of its EVs in order to combat range anxiety on those few occasions a year when it's a concern. That type of 'out of the box' thinking, in which existing resources are used to enhance a company's EV offerings, could easily pose an unexpected threat to TSLA's assumed dominance.
Similarly, the way that TSLA bulls dismiss Nissan (NSANY.OB) (and its top-selling EV as a potential competitor is not particularly persuasive to me). As much as the paradigm of "starting the learning curve with a luxury vehicle and then working down the value chain" is espoused by TSLA's proponents, it's not at all clear to me that it's necessary or even particularly beneficial to go from luxury car to mass-market car. My guess is that Nissan and Ford (F) have learned directly many important lessons relevant to the mass market - including being ahead on the manufacturing and supply management learning curve (not just from overall institutional knowledge, but specifically from their experience building EVs and PHEVs for the lower priced market). Moreover the feedback they get from working in the actual market seems as valuable - if not more so - than that which TSLA gets from its affluent Model S owners / aficionados.
As such, the potential rise of important competition is the fourth factor supporting my short position.
#5 Potential Dilution
The final factor in my short case is that there are a large number of options outstanding (though not yet all exercisable) that will eventually dilute the stock by about 20%. Or to put it another way, the valuation metrics presented at the outset of this article actually understate the fully diluted over-valuation by about 20%. Here's the most currently available breakdown of options from the company's 10-K report (TSLA chooses not to update this table in its 10-Qs).
(The large number of employee options outstanding might also shed additional light on the company's choice to be self-promotional, as outlined in factor #1 above.)
Trading Position and Approach
Those then are five primary reasons for which I'm short TSLA. But to conclude, let me now put the short position in its trading context. For various reasons (which I'm not going to get into) I'm generally bearish on the market as a whole for the next few years. As such I want to have a portfolio of shorts to hold. As outlined above, TSLA fits my criteria for a promising short and will make up somewhere between 3-4% of my short portfolio at an average entry price of about $147.
It's important to note, however, that none of the factors above lead to any short term catalyst, so the timing of any eventual price drop is uncertain. Instead, I simply expect the stock to trade to less than 50% of its current fully diluted market cap at some point over the next 2-3 years. If that occurs, and depending on my overall outlook, I will probably scale out of the position on the way down to my target price.