2014 Q3 Earnings Report and Conference call discussion thread

[Robert.Boston]

This line in the shareholder's letter caught my eye:

For example, our Model X body center will initially be independent from our Model S body center, allowing us to start building Model X bodies without impacting Model S production.

The talk has been about a unified body center, with the clever robots switching production on the fly. It looks like (a) Model S demand is so high that it's fully utilizing the expanded capacity and (b) there are enough differences between the S and X that building them on the same line is inefficient.

 

[ItsNotAboutTheMoney]

It's a tough job, which is all the more reason to evaluate if there are partners who would be able to do a better job than yourself. There is no reason to assume that a 'BEV with 18650 cells' is different than other innovative high tech consumer goods when it comes to manufacturing. Yes, feedback between engineering and manufacturing is essential but they don't have to be in the same building.

Margins in manufacturing are tight. Tesla has understood the benefit of colocation. As Elon Musk remarked, when asked about having Chinese manufacturing, when they looked at buying from China it ended up being more expensive because of shipping costs. Tesla doesn't make everything, and doesn't try to make everything; Tesla focuses on the high value items where expertise is essential.

The continuous upgrading actually is not a strength but a weakness. (And it is a testament to Tesla's marketing which you dismiss that Tesla customers perceive so differently) Where seamless continuous upgrades on the software side generally is welcomed by consumers, not so on the hardware side. In fact, some of the strongest words of ardent Tesla supporters on this forum towards the company are made wrt hardware updates complaining about a lack of transparency on policy and implementation of them.

The continuous upgrading is absolutely a strength. Tesla's policy is absolutely obvious: you get what you ordered, hardware upgrades might be possible but don't count on it and be prepared to pay a lot, software upgrades will happen but don't count on when. The upgrading means that they can catch up faster and that they don't get leapfrogged. Without inventory, dealers or marketing there's no reason for the typical cycle.

That's absolutely not true. For example, the development of the supercharger network is foremost a marketing rather than a technological achievement. It's technologically well executed but not actually terribly innovative. But on a marketing level, it changed the whole message about driving long distance with an electric car. The ability to asses what the market needs for electric driving is something that Tesla understands as no other company and is in my eyes their biggest competitive advantage.

What other charging network gets people to pay a large amount of money up front in a lump sum? What other network has no account management or transaction handling? What other network pays the site owner to host chargers? What other network puts 6, 8, 10 or even 12 slots in one location? The innovation isn't just marketing, it's about building an actual, coherent network with the lowest possible cost.

 

[chickensevil]

This thesis depends on Model S demand being larger than the number of batteries they can produce, which I believe is ~200K pre-GF (i.e. current panasonic deal is to grow to ~200K batteries per year without GF). I am not so sure that Model S demand will ever be that high; they need Model X to get to this number (pre Model 3). So while a short delay now should not impact profitability, further significant delays or scrapping it altogether certainly would.

200 is being optimistic. They said 150k would be doable MAYBE 200 if they were to push Panasonic hard but both of those numbers were stated under the umbrella of a question about how much could you do if you COULDNT get the gigafactory.

The secondary evidence pointing more toward 150 is the production capacity for cells in the gigafactory is going to be 35GW but the number of packs produced will be 50GW. That missing 15GW comes from Panasonic and comes out to around 150k cars.

All that being said the timeline to get there is ~2017. So not 2015 which they will be lucky to be getting enough cells to hit 70k.

The X will surely be out by the time they are going for 100k+ cars.

 

[blakegallagher]

I couldn't disagree more. I think your whole argument hinges on the assumption that bringing Model X to market would increase Tesla's profitability, and I find fault with that assumption.

The reason (of course) is that Tesla is production constrained, not demand constrained. People here seem to understand this when we only talk about Model S, but whenever we talk about Model X they assume that a second model obviously means that Tesla can sell more cars. They can't.

Model X uses the same batteries, sourced from the same partner, assembled in the same space, by the same team. Its design center needs to be build and maintained by the same software developers. It needs to be sold by the same sales people, in the same stores, where it is displayed on the same floor. It is shipped around the country and world on the same trucks and boats, inspected and delivered to customers by the same delivery experts. It is maintained in the same service centers by the same service people.

But here's the catch: it's not the same. It is a different machine. Not completely different, but different enough. Different enough that building x Model S and y Model X is much much more costly than building x+y Model S. And it has to be one or the other. The Model X program steals resources from the Model S program.

I hope that is clear. It is crystal clear to me.

The same logic can be applied to the Supercharger rollout plans and to the battery swapping stations. Both are meant to boost demand for the cars by making them more convenient to own than they would otherwise be. The battery swaping stations seem to be completely unnecessary (nearly everyone is content with Superchargers) and it turns out we don't need nearly as many Superchargers as we thought in order to sustain this level of production.

When Model X was first announced Tesla thought they needed it. They thought that demand for the Model S was around 20k units a year world-wide. But that turned out to be wrong. The truth is (as was discussed on the call today) that Tesla doesn't need the Model X at all. There seems to be enough demand for Model S for Tesla to grow at an enormous rate---about as fast as is possible---until Gen III.

Now, they didn't find that out until well into the Model X program, so there is no point in scrapping it now, but it is very prudent to take their time with it. On the other hand, think of those other Model S derivatives that Tesla talked about around the time of the IPO. There was supposed to be a cabriolet as I recall. Again, they thought they would need it. But they were wrong.

I think what you see as poor execution I see as just good business decisions in the allocation of resources. These predictions were all made with a certain level of demand in mind. A good CEO should change course once they find that reality doesn't fit their initial assumptions. That's what they have done here.

I totally understand if Model X reservation holders are upset about the date being moved back, as a customer, I certainly would be. But, I can't imagine how any investor could possibly be upset. Releasing Model X could do nothing but hurt profitability. Worse, it complicates operations and puts the Model S program (which is doing so well) at risk.

As long as Elon delivers the kind of stock appreciation I'm looking for, I don't much care which levers and knobs he manipulates to do it. It looks to me like he has decided to turn the Model X dial down so that he can crank Model S to the max. For what it's worth, I totally agree with that decision.

Totally agree. Whenever they bring X to market margins will suffer ...a lot. It will take awhile to iron the kinks out like it did with the model S. (edited ... use to say Model X)

 

[pGo]

I liked the conference call. The tone was upbeat and Elon seemed very confident on demand. What keeps bothering me is production. What in the world is taking so long to figure out production? It has been almost more than 2 years since they started production. Its a good enough ramp to figure out the best processes.

If you look at ICE, because of its more moving parts, it is supposed to an issue there, but everyone cranks out huge amount of cars. Tesla has a lot of electric components, and that should be easier to manufacture in masses. Before, battery was given a reason for slow production, but now I don't get it. Any thoughts?

 

[pGo]

Model S, in Q1 13, at about 4900 units made the quarter profitable while manufacturing processes still not being perfect. I think Model X doesn't have to go through a lot of those. They don't need duplication of teams and resources as a lot can be used for both products. I still don't get this.

 

[bonnie]

I liked the conference call. The tone was upbeat and Elon seemed very confident on demand. What keeps bothering me is production. What in the world is taking so long to figure out production? It has been almost more than 2 years since they started production. Its a good enough ramp to figure out the best processes.

If you look at ICE, because of its more moving parts, it is supposed to an issue there, but everyone cranks out huge amount of cars. Tesla has a lot of electric components, and that should be easier to manufacture in masses. Before, battery was given a reason for slow production, but now I don't get it. Any thoughts?

They completely shut down the production line and revamped it - and consequently took a one month hit. And until the gigafactory is up and running, batteries will continue to be a constraint.

 

[Right_Said_Fred]

Another thing to point here, Elon is preparing the market for 2015 expectations. A few months back, this board was talking in the tunes of 75K. Now, it looks like somewhere around 48K (50% more model S) + ~10K (model X being delayed till Q3 2015) = 58K. I like 2000/week numbers, but the math does not add up unless they quickly raise Model X production. But that's been difficult so far.

The fact that this board was talking about 75K in 2015 doesn't mean that it's something Tesla was planning for. If they deliver 33K this year, then an 'over 50% increase' means at least 49.5K in 2015. Add to that 10K Model X and indeed you would get to just under 60K.

But that doesn't mean it doesn't add up with the 2000/week number at the end of 2015. That's just because Model X production starts up later in the year than planned. If Model S sales/production increases by another 50% in 2016 (Tesla: "by over 50% ... for probably several more years to follow"), then Tesla will produce at least 75K Model S in 2016, plus Model X, for which 25K is probably low-balling it. Together that makes 100K, or 2000/week. But Model X production will probably be much higher in 2016, so 120K is a reasonable number to expect for that year, based on Tesla's expectations.

 

[jhm]

I liked the conference call. The tone was upbeat and Elon seemed very confident on demand. What keeps bothering me is production. What in the world is taking so long to figure out production? It has been almost more than 2 years since they started production. Its a good enough ramp to figure out the best processes.

If you look at ICE, because of its more moving parts, it is supposed to an issue there, but everyone cranks out huge amount of cars. Tesla has a lot of electric components, and that should be easier to manufacture in masses. Before, battery was given a reason for slow production, but now I don't get it. Any thoughts?

My impression is that Tesla is coming to terms with a complexity constraint. The more variants they have to make, the more flexible the system has to be. But this flexibility comes at a cost that impedes efficient high volume production. So the solutions at this point are to reduce the number of configurations, i.e., no more 60D or P85 and consolidated options. Moreover, Model X will get it own production line, not because Tesla can't be more flexible, but because the complexity of ramping up MX could put MS production at risk. So this last quarter Tesla actually witnessed this sort of risk. The assembly line upgrade took about a month and set production back 2000 cars. And this is why we are guided to 33k, down from 35k. This risk was taken on to accommodate higher volume, D and Autopilot hardward, and Model X. Thus, it was a necessary risk, but Tesla is learning how to mitigate those risks. Having a separate line for MX and reducing options/variants are ways of confronting this complexity constraint.

 

[aznt1217]

In regards to model X it was pretty clear what the delay is about. They just want to make the ramp up faster. A lot of this will come in Q1 with the new body and paint center up and running. We know Tesla isn't one to waste time, so while making these mass production improvements, why not tweak the product to perfection. They said on the CC that the team needs to work on not being perfectionists. I believe this 1000%. Just think about Elon's comment about the sunvisor sticker... If only GM had this attitude on THAT SWITCH. If only BMW had this attitude about that stupid idrive system. Its about getting the details right.

I also share sentiments that the team wants real world testing of the data for the dual motor and regulatory approval of side mirror cameras to work on range for the X. I also believe they want to scale faster to keep margins the same by pricing the base X slightly above the Model S.

 

[Robert.Boston]

Increased margins are not just coming from reduced costs. I noticed that the AWD option is now up to $5,000 -- up $1,000 from it's announcement last month. My guess is that demand for the D is so high that they decided to up the margin on it. Understandable but a little tacky.

 

[JRod0802]

The fact that this board was talking about 75K in 2015 doesn't mean that it's something Tesla was planning for. If they deliver 33K this year, then an 'over 50% increase' means at least 49.5K in 2015. Add to that 10K Model X and indeed you would get to just under 60K.

But that doesn't mean it doesn't add up with the 2000/week number at the end of 2015. That's just because Model X production starts up later in the year than planned. If Model S sales/production increases by another 50% in 2016 (Tesla: "by over 50% ... for probably several more years to follow"), then Tesla will produce at least 75K Model S in 2016, plus Model X, for which 25K is probably low-balling it. Together that makes 100K, or 2000/week. But Model X production will probably be much higher in 2016, so 120K is a reasonable number to expect for that year, based on Tesla's expectations.

So you think production will increase by more than 100% from 2015 to 2016 (under 60,000 in 2015 to 120,000 in 2016)?

 

[Zzzz...]

I couldn't disagree more. I think your whole argument hinges on the assumption that bringing Model X to market would increase Tesla's profitability, and I find fault with that assumption.

The reason (of course) is that Tesla is production constrained, not demand constrained. People here seem to understand this when we only talk about Model S, but whenever we talk about Model X they assume that a second model obviously means that Tesla can sell more cars. They can't.

They thought that demand for the Model S was around 20k units a year world-wide. But that turned out to be wrong. The truth is (as was discussed on the call today) that Tesla doesn't need the Model X at all.

I could not agree more with you! No point in bringing X to market if it would not actually increase sales.

 

[geturchiru]

From the bears point of view, they did not meet the Q3 guidance(minimal effect considering it a marginal miss), poor guidance for Q4 and for 2014(lesser by 2k cars which is 6% compared to previous guidance and the total revenue will be less by ~200M), not able to meet the previous guidance so what is the guarantee that they will in 2015, no ZEV credits for Q4(again 100M loss) and those who is looking for short term profits read as funds might be disappointed with the ER.

 

[Robert.Boston]

I was initially stunned by the 50% growth per year scenario. Then I actually ran numbers and realized that these are pretty much in line with what we've been kicking around here on TMC for some time. Specifically:

Year	Production
2014	35,000
2015	53,000
2016	79,000
2017	118,000
2018	177,000
2019	266,000
2020	399,000

Looks about right, or borderline conservative if the Model 3 is as great as seems likely.

 

[aznt1217]

I was initially stunned by the 50% growth per year scenario. Then I actually ran numbers and realized that these are pretty much in line with what we've been kicking around here on TMC for some time. Specifically:

Year	Production
2014	35,000
2015	53,000
2016	79,000
2017	118,000
2018	177,000
2019	266,000
2020	399,000

Looks about right, or borderline conservative if the Model 3 is as great as seems likely.

I think the 50% was strictly talking about the higher margin vehicles for the next 2-3 years. When Gen III rolls around I think the goal was 500k annually by 2020.... somebody correct me if I'm wrong.

 

[chickensevil]

I think the 50% was strictly talking about the higher margin vehicles for the next 2-3 years. When Gen III rolls around I think the goal was 500k annually by 2020.... somebody correct me if I'm wrong.

Initially I wanted to agree with you, but I think the way it was stated was almost an "at least" 50% we will ramp faster if we can but ramping up manufacturing is hard. So I would take that to mean that 50% is almost as fast as they feel comfortable in saying they can go across the board for everything. However at some point I think they would need to go just a bit faster if they want to hit 500k by 2020.

 

[Zzzz...]

50% grows rate of Model S could continue only for a limited number of years. Worldwide Model S sales already exceeded worldwide Panamera sales(~33k vs ~30k units this year). Moreover MS in it segment already outsells every model, Audi A8/A7, BMW 6 and 7 series, merc CLS, Lexus LS in US so far this year, with exception of S-class.
Large Luxury Car Sales In America - September 2014 YTD - GOOD CAR BAD CAR

And it do look like next year MS will have a good shot at outselling S-class not only in US, but worldwide. Rest models would be left far behind in rear view mirror. But how far Model S could realistically grow to? I doubt MS from Teslas will outsell combined power of brands from Japan and Germany.

 

[Newb]

Elon clearly stated that 50% growth rate in orders and deliveries refers to Model S only (!). Check out conference call at 16:30-17:05

 

[ckessel]

One thing that bugged me in the call (sorry if this was mentioned somewhere in the preceding 11 pages), was the complete dodge on the question of current run rate. Elon hesitated and Deepak said it "wasn't relevant". How on earth is run rate not relevant? It's directly related to meeting projections. And it's a question they've always answered before.

I can't think of what value they'd have in dodging the question. The gave Q4 projections and that's close enough you can derive some run rate information. They also drained every possible car from supply and apparently ZEV credit to make Q3 non-GAAP profitable. Tesla isn't usually afraid to be honest and they're Q3 ER statement is pretty blunt about being willing to sacrifice short term for long term, so why the combination of dodging on the run rate and such an extraordinary effort to sell every car?

Probably nothing, but it stuck out as odd.