Let me start by saying I found the Tesla (NASDAQ:TSLA) Model 3 very pretty on the outside. A lot prettier than GM's (NYSE:GM) Chevy Bolt for sure. The interior seems weaker, but I believe it will be revised considerably from its prototype stage to being a commercial product. At the very least, I expect to see a HUD (Heads-Up Display) complementing the console-less design - a step that would make it much more acceptable.
With the Model 3 being so pretty and with reservations just requiring a refundable $1,000 deposit (versus the $5,000-$40,000 in earlier Model S and Model X iterations), it's not a huge surprise reservations are incredibly higher. However, this story doesn't end at the car being pretty and reservations being massive. There's a whole lot more to be said, and this article covers some of it.
The Car People Saw Won't Cost $35,000
This is the first thing to consider. Yes, the Model 3 shown was very pretty, but what was shown is not representative of the base $35,000 model. Consider the following:
- The cars shown had 20" wheels. These will be premium wheels. The standard wheels will be 17-18". The price difference on the optional wheels shown will be around $2000-$4000, if we go from Model S options' pricing (using a lower price given the different wheel size).
- The cars shown had premium interiors. The base model will have textile seats. The price difference on the interior is likely to be ~$2,500.
- The cars shown had a glass roof everyone thinks will be standard. This is unlikely to be so, with a regular metal roof being in the base model instead. Ultimately the glass roof will be an option - going from the Model S, a $1,500 option. If the glass roof was cheaper to build, the Model S would come with it as standard. Unless we see an updated Model S with the glass ceiling as standard (and metal as an option), it's likely that the Model 3 will also have glass as an option.
- The cars shown had premium paint including options not available today. At the very least, the paint jobs shown will be $1,000-$1,500 extras and quite possibly even more for the matte color.
- The cars that were test-driven had an AWD powertrain and likely larger batteries and certainly better acceleration than the base car. This won't be the case with the base Model 3, which will be RWD with the smallest available battery. AWD itself is likely to be a $5,000 option, and the battery would add to that (but I won't be considering the difference right now).
- The $35,000 car Tesla is promising, but does not include full autopilot. It only includes the safety features, such as emergency braking, which will be fairly standard when the car hits the market. Full autopilot will entail a further payment. This I'm only adding because a lot of people seem to think the car comes with full autopilot (not to speak about the fantasy of full or close to full autonomy).
- The $35,000 car Tesla is promising is supercharging capable, but won't have access to free supercharging. This is a common misconception. Tesla said it was "capable" because it can use the superchargers. But it never said it would have free access to the supercharging network. It saying "capable" is an indirect way of saying it can - for a fee. This fee will either be upfront or usage-based and explains the changes to the TSLA 10-K removing the "free" claim from supercharging.
Putting all of this together, the cars shown in Tesla's presentation were at the very least (ignoring possible larger battery sizes) $42,000 (if we consider all aesthetic changes at their lower ranges and don't include AWD) to $49,500 (including AWD and options at their higher ranges).
Of course, we should also add that the car costing $35,000 won't be as pretty or capable as the one shown due to the differences above. Many of the people putting reservations on the Model 3 might not be aware of these differences.
The Model 3 Stands A Good Chance Of "Osborning" The Model S
What do I mean by "Osborning"? It's a reference to Osborne Computer, which is said to have gone bankrupt on account of previewing a much better computer model well before it became available. The preview supposedly led to a dramatic plunge in orders for Osborne's existing products, which in turn led to its bankruptcy.
The Osborne story is controversial and there are other explanations for its failure. But still, the prospect of a severely cheaper and as-capable product can clearly have an impact on existing products.
In the case of TSLA, there are a couple of factors which make it worse, including:
- TSLA promising the car is less than two years away, which can fall within many people's tolerance to keep their present vehicles a while longer.
- The TSLA Model S having many "stretch buyers," who bought it for its unique qualities and because no other similar (but cheaper) vehicles were available.
Perhaps it could be said that a BMW (BAMXY) 5-Series has a different market than a 3-Series. But the problem here is that the entry price on a Model S ($70,000 - a 70 with RWD) is exactly double the promised entry price on a Model 3 ($35,000). A 5-Series BMW, on the other hand, has an entry price ($50,200) that's "just" 51% over the price on an entry 3-Series ($33,150).
TSLA supporters usually put this effect aside by saying that the Model S is much faster and has a much larger range. But that isn't so when we compare the entry versions of each model and likely won't be so when we compare the higher versions either. For instance, regarding the entry versions:
The main positive difference for the Model S would probably be just in terms of superior cargo space, as the living space looks to be rather similar given the Model 3 tradeoffs (smaller front and cargo space). Also, the 70 RWD does include free access to the supercharger network, which will be an extra for the base Model 3.
This price difference clearly seems too large for what it brings to the table. On top of it, the Model S is made to look somewhat dated by the Model 3 visuals. Putting it all together, the Model S does indeed look like it will be cannibalized on the approach to the Model 3 delivery. The debate on this possibility is already ongoing.
Tesla is aware of this. Changes to the Model S lineup and pricing are expected by mid-April. Initially, there were rumors that there would be price increases (which might have favored March orders), but given the already-massive price differential this looks to be unlikely. More likely, is the improvement of the Model S feature set along with a slight face-lift, so as to minimize the cannibalization. Still, again, the price difference looks too large for there not to be a significant impact among stretch buyers.
Not only is there the chance that the Model 3 will Osborne the Model S, but I also find it highly unlikely that it will have positive margins. The reason is simple and does not rely on a large tortured model (which in my opinion considers too little weight, too small a battery for the range, and too weak a hp rating for the acceleration given that the weight will be unrealistic). The Model S 70 RWD is very low margin, and the Model 3 has nearly all of the same content. Size differences are not going to matter as much as thought because even though the Model 3 will be smaller, using much less aluminum and more steel will make it relatively heavier (that is, the weight savings from being smaller will at least be partially eaten by using steel). The weight not being as low due to aluminum substitution will impede possible savings elsewhere. For instance, a Nissan Leaf with a smaller air-cooled battery and less powerful motors, also using a steel and aluminum construction, carries a curb weight of ~3,300 lb -- already higher than the model above considered for the Model 3.
Even if we take out 20% of the entire Model S cost and see the 70 at a 15% gross margin, both optimistic assumptions, this would put the base Model 3 cost at $47,600. A 20% smaller battery would see the Model 3 battery at ~55kWh, which seems realistic.
Also useful to remember, the whole idea of a cheaper Model 3 was predicated on batteries being built from the ground up in the new Gigafactory. Well, there are news on that front as well …
Gigafactory Is Giga No More
During September 2015, I wrote an article titled "Honey, I Shrunk The Gigafactory (And More)", where I concluded that the original Gigafactory project had already been substantially shrunk. In the meantime my thesis has received significant confirmation both by the fact that the factory grew no more, and by the fact that a TSLA spokesman has already said that the construction of the whole Gigafactory won't happen until 2020, whereas this was supposed to happen until October 2017.
As always, this has been denied by TSLA, but the denial doesn't hold water because it speaks about a different reality - that of production, both when it will start (early 2017 as per TSLA) and when it hits its full capacity (2020). You see, what is being said about the Gigafactory is not that it won't produce or hit full production by 2020. Instead, what's being said is that the Gigafactory won't be built to the previously-divulged size, and this has further implications.
In this case, the implications seem to be that it will no longer be a vertically-integrated factory taking in raw materials and churning out battery packs. Instead, at its much-reduced size, it simply seems the factory will assemble vehicle battery packs and battery powerwals and powerpacks, while taking in complete battery cells as an input. Obviously, the lack of vertical integration means all previous ideas of a massive cost reduction on the battery packs it produces are now gone.
They're probably gone for a good reason, too. This is so namely because of the fact that producing batteries is cheaper in Asia (but Mexico could also compete) due to lower labor cost and better availability of the required value chain. It's thus not a surprise that Panasonic (TSLA's partner in the Gigafactory) is investing in a new EV battery factory. In China.
I believe this new reality will filter out slowly through time. There are no PRs for negative developments. You never saw a PR by TSLA canceling one of the two "hugely successful" Powerwall products. You won't see one for this, either.
It's Not Just The Gigafactory
During the Model 3 unveil, and presumably to handle the demand which the Model 3 would bring, TSLA announced that it would be doubling the number of superchargers and service centers by the end of 2017. This would seem like a relevant development, enabling TSLA to handle the Model 3 volume. I said "it would seem" for a reason, though: because it wouldn't (enable TSLA to handle the Model 3 volume).
The reason is simple. The Model S+X installed base will grow by ~170% between 2016 and 2017. Hence, merely doubling the supercharger network and service center network wouldn't even keep up with the needs of the Model S and X, never mind the Model 3.
Moreover, doubling the supercharger network and service network will also increase TSLA's SG&A cost base heavily, since these networks' operating costs flow through that line. This, in turn, means that for the next 2 years economies of scale are not likely to provide a significant uplift to TSLA. There could be economies of scale in R&D, but that would ignore the fact that the Model S is aging and will also require significant investment to replace.
On The Positive Side - Significant Deposits Lower TSLA's Liquidity Needs
If TSLA gets 0.5-1 million $1,000 deposits on the Model 3, such will amount to a $0.5-$1.0 billion interest-free unsecured loan. This will allow TSLA to not borrow the same amount from elsewhere at a much higher cost, or have less need to issue more equity.
It should also be interesting to check Q1 2016 quarter-end customer deposits. We'll have to subtract $118 million from those to know how underlying Model S+X deposits evolved. This is so since TSLA still received at least 118 thousand Model 3 reservations within Q1 2016.
On a slightly negative note, the high number of Model 3 reservations might actually increase TSLA risk a bit. This happens if the reservations (which aren't permanent capital or liquidity) convince TSLA management that the company is fully funded on account of the deposits and that TSLA doesn't take advantage of the amazing opportunity to issue more equity. Equity, in a business as risky as TSLA, ought to be issued when it's possible to do so, not if it ever becomes absolutely necessary to make such issuance.
When Will Model 3 Deliveries Start?
The official line is that Tesla will start delivering the Model 3 in late 2017. This seems highly unlikely to be achieved. Let's take a look at how Tesla performed with its past models:
- First prototype ~2004-2005, though first public prototype hit only on July 19, 2006
- First delivery (to Elon Musk) in February 2008
- 3-4 years from first prototype to first delivery. However, the Tesla Roadster isn't really comparable to the Model S and X due to it being built on a partially-constructed body based on the existing Lotus Elise. This significantly reduced the time needed to deliver a consumer car.
- First public prototype on March 26, 2009
- First deliveries (hand built) on June 22, 2012
- 3 years, 3 months from first prototype to first deliveries
- First public prototype on February 9, 2012
- First deliveries (hand built) on September 29, 2015
- 3 years, 7-8 months from first prototype to first deliveries
- First public prototype on March 31, 2016
- First deliveries on ???
For the Model 3 to be available in late 2017 (say, December 2017), it would have to go from first prototype to first deliveries in 1 year, 9 months. That would be nearly half the time it took for the other 2 models, which seems highly unlikely.
In Tesla's favor, we have:
- Greater experience in building cars/EVs in general.
- A simpler concept than the Model X, though more or less in line with the Model S in complexity.
- Greater experience with the whole powertrain. Indeed, the ability to show several running Model 3 prototypes certainly came from using a derivative of the Model S/X powertrain (the AWD version, to boot).
Not all is favorable, though, because:
- The car needs to be built to a much tighter budget and in much higher volumes, which implies redesigning nearly the entirety of it. As an example, the Model X shares 30% of the parts content from a Model S (down from an expected 60%), whereas the Model 3 is said to share very few components with either the Model S or X.
- The car will have a much higher steel content and so TSLA will be faced with a different construction technology with which it has much lower experience.
Overall, it seems highly unlikely that the Model 3 will actually start shipping before 2018. Worse still, were TSLA to take 3 years (which would be faster than either the Model S or X), and that would put first deliveries into early 2019.
A Slight Doubt Regarding OTA (Over-The-Air) Updates
Little has been said regarding whether the Model 3 will receive updates like the Model S and X do. This is not as irrelevant or certain as it might seem. There is uncertainty because using OTA updates implies TSLA paying mobile operators for each and every Model S and X circulating out there. Given Model 3's challenging price and margin profile, it would seem hard to believe Tesla would assume a costly telecom bill on every one of those cars as well.
OTA is a simple concept to implement. Many other automakers haven't implemented is as a free feature not out of their inability to do so, but out of the ongoing cost it implies for them. Like with supercharging, there will be a push towards charging an ongoing cost for this feature (though here TSLA charging right away is very uncertain as well, unlike with supercharging).
The $7,500 Conundrum
The fact that the $7,500 federal tax credit starts being phased out in the quarter after an automaker delivers its 200k vehicle has already been beaten to death. However, there's something else which hasn't been beaten nearly as much.
This something is what happens when TSLA loses access to federal tax credits during 2018, while several new EV competitors are still warming up their EVs. This event will mean each TSLA EV will be saddled with an additional 10-20% cost disadvantage on whatever cost inefficiency TSLA still has versus traditional automakers at that point.
To this problem we have to add that other automakers can live with breakeven margins on EVs, because they produce ZEV credits which they would otherwise have to buy (most likely from TSLA).
TSLA hasn't been able to capitalize on having unique EVs while it still has the federal tax credit and little competition. It would seem that it not having the federal tax credit when others still have it, together with other automakers initially not requiring positive margins on their cars, would argue for a very difficult future for TSLA between now and at least 2020.
When the present euphoria dies down, what will be left is:
- TSLA with massive reservations for a car it can't deliver profitably.
- A Model 3 which will arrive late.
- When the Model 3 arrives the $7,500 federal tax credit will be on its way out. TSLA will then be fighting without a tax credit versus many other automakers still having access to the tax credit.
- The other automakers will be incentivized to price their EVs at little to no margin, both due to ZEV credits and market share.
- And in the meantime, the Model 3 might also dent ongoing Model S/X demand.
All of these put together mean that right now, TSLA's risk of failure seems even higher than it was before. This is so because the deposits might give it renewed confidence in not needing more equity, and also because its current lineup will suffer to some extent from the "Osborne effect".
As a small aside, I should also add that no customers wanting to buy the Model 3 at $35,000 can possibly expect to receive the tax credit. This is so because TSLA will prioritize higher-end versions when it starts delivering the Model 3. Likewise, none of the reservation holders for the base model can expect to be saving a place in line to receive their Model 3. This is so because TSLA will both discriminate geographically and by model price (just like it did with the Model X).
Disclosure: I am/we are short TSLA.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.