特斯拉 (TSLA) 2015 Q1 法說會逐字稿

Good day ladies and gentlemen, and thank you for standing by. Welcome to the Tesla Motors first-quarter 2015 financial results Q&A conference call.

(Operator Instructions)

As a reminder, today's conference may be recorded. Now, my pleasure to turn the floor over to Jeff Evanson. Sir, floor is your's.

Thank you Hughie, and good afternoon everyone. Welcome to Tesla's first-quarter Q&A webcast.

I'm joined today by Elon Musk, Tesla Chairman and CEO; JB Straubel, our CTO; and Deepak Ahuja, Tesla's CFO. We announced our financial and operational results today in a shareholder letter that is available at the same link as this webcast, And a replay of the webcast will be available later today at the same link.

The shareholder letter includes GAAP and non-GAAP financial results, as well as reconciliations between the two. Our non-GAAP measures add back deferred revenue and related expenses for cars delivered where cash has been or will very soon be collected.

These non-GAAP results also exclude stock-based comp and non-cash interest expense. Revenues and costs associated with cars leased directly through Tesla are treated the same in our GAAP and non-GAAP financial information.

During our call we'll be discussing our business outlook and making other forward-looking statements which are based on our predictions and expectations as of today. Actual events or results could differ materially due to a number of risks and uncertainties, including those mentioned in our most recent form 10-K filed with the SEC.

Now, Hughie, if we could please have the first question.

(Operator Instructions)

Dan Galves, Credit Suisse.

If we assume Model S volume of 50,000 in 2015, deliveries need to average over 14,000 per quarter in Q3 and Q4. What's the biggest challenge to get there, and do you have order rates today that support that level of volume in the back half?

Yes, that's about right. Actually, with Model X production ramping up quite heavily in Q4, depending upon how that ramp goes, and obviously it's difficult to predict that with perfect clarity, but our volume essentially doubles in Q4. Depending upon how the ramp goes, I want to emphasize, sometimes people don't totally appreciate is that there are several thousand unique parts in a car. And if even one of those parts is not available for any reason, then you cannot ship -- you cannot scale production.

Essentially the production ramp goes according to the unluckiest and worst-performing supplier or part of Tesla. That said, we do expect to see a significant ramp in Q4 for the X. And have something that is may be as much as 2 times of the quarters in Q4.

As far as demand for that, we do not see that being a problem. There are huge advance orders for the X, and we see a steady climb in demand for the S.

Okay. Got it. Just a follow-up. You've improved the Model S a whole lot in the last couple of years. Does that make it easier to launch Model X at a high quality level? Does it allow you to move engineering and design resources toward Model 3 faster than you moved resources to Model X?

Yes. In the case of the X, if the X ended up being a lot more different than the S than we originally anticipated, so that the development took a lot longer and we were distracted solving all sorts of issues with the S during that time, which made it difficult for us to allocate engineering resources to the X when there were issues to be solved with the S. I think we'll do a lot better with the X.

We are paying close attention to some of the things that are different about the X to make sure that they are not an issue, particularly buckling door and the second row seats. I'm feeling pretty good about things.

Because that production ramp scales exponentially, depending upon where that exponential curve falls across a quarterly boundary can actually make quite a significant effect on the production and deliveries in that quarter. That's why it's a lot easier to predict the continuous than if it's discreet with arbitrary quarterly cutoffs.

I really think the X is going to be -- it's really going to be a great car. I just drove the latest prototype today. And it was like, wow, this is by far the best SUV.

Sounds good. Thank you.

Go ahead, Hughie. Next question, please.

John Lovallo, Bank of America Merrill Lynch.

The first question is, do you guys earn four or nine ZEV credits per vehicle, and how many vehicles are participating in your battery swap beta program?

The amount of credits, ZEV credits, we earn depends on the size of the pack. Different for what we are doing for the 60 versus 85, and now the new one is at 70.

That has been varying over time, the amount of ZEV credits we earn. I'll need to confirm that before I give the official number here.

The ZEV credits thing, it sort of moves things by like 2%. It's not super-material. I'm not sure what the point of your question is.

Here's the point of the question, is that assuming that you guys sell 10,000 vehicles in ZEV states this year, that would mean that you would need 400 program -- sorry, 400 vehicles involved in this battery swap program, if you assume that each of them swap 25 times each. A minimum of 400 vehicles. I'm just curious if there are 400 vehicles involved in the battery swap program.

You do realize that ZEV credits don't sell for 100 cents for the dollar. They sell for like $0.50, or sometimes less. There are not always customers for the ZEV credits.

I'm just asking about in terms of what you guys are actually earning per vehicle. Okay. And I guess the next question (multiple speakers)

How'd I do? Yes. And as more of our production goes overseas, obviously there are no ZEV states overseas. As our sales increase outside of California, also -- or Canada, those are not ZEV states. So this is like -- ZEV stuff is an increasingly small part of the picture over time.

Okay. That's fair. Go ahead.

We do have hundreds of vehicles in the battery swap pilot program. It's not like 10 or something like that. It's hundreds of vehicles.

Yes. We're steadily increased the invitation list. We found that there is not a lot of interest in people doing pack swap. We make the invitations and we got a very small percentage that actually take us up on the invitation.

Right. No, the only point is that it has a difference of five ZEV credits per vehicle. That was the only point.

Second question would be, and final question. There is some news report saying that Solar City has said they are not going to use the 7-kilowatt hour battery. In fact, I think there was a quote from a spokesman, whether it's true or not we don't know, saying that it doesn't make economic sense. Can you guys comment on that, please?

Let me just talk more broadly about the response to the Powerwall and Powerpack because I think that's really the questions you should be asking. The response has been overwhelming, like, crazy.

In the course of less than a week, we've had 38,000 reservations for the Powerwall, 2500 reservations for the Powerpack. The Powerpack, it should be noted, typically this is brought by utilities or large industrial companies, for heavy industrial work.

Typically, Powerpack, it's at least 10 Powerpacks per installation. So if there's 2500 reservations, actually 25,000 Powerpacks.

Powerwall also is, we suspect, is probably an average of the number of Powerpacks is probably 1.5 to 2 per installation. 38,000 reservation is more like 50,000 or 60,000 actual Powerwalls.

So there is no way that we could possibly satisfy this demand this year. We are basically sold out through the middle of next year in the first week. It was just crazy.

We had 2500 requests from companies that want to distribute and install Powerwall and Powerpack. We can't even respond to them. We have to, like, triage our response to those who want to be a distributor. It's crazy off the hook. It seems to have gone super-viral.

For the specific case of Solar City, what they are referring to is that there is two versions of the Powerwall. There's the daily cycling version and there is the power back-up version. One is energy optimized in one is daily cycling optimized.

For the daily cycling optimized one, the economics, it is true in the US, with rare exceptions, are more expensive then utility. If somebody wants to do a daily cycling, basics, they go off-grid. It's going to be more expensive than being on-grid.

This doesn't mean that people won't buy it, because there are people who want to go off grid on principle or they just want to be independent. That's what the Solar City comment is about.

It might also be worth noting that Solar City doesn't yet operate in Europe. And the main target application for the daily cycling battery pack was actually -- were several markets not in the continental US, and particularly Germany and Australia are very strong markets where it does make economic sense today, based on the feed-in tariff and the electricity rate structures in those countries. Solar City's comments, I think, need to be put in the regional context.

Solar City's only operating in the US. The Powerwall -- it will be available from Solar City and from other installers in both configurations. But if someone is doing a daily cycling application, they are doing it because they specifically want grid independence. There's some number of people who will want to do that, and that's good.

It's also important to appreciate for, even to say the power backup systems so that you always have power in the event of power outage. Let's say that appeals to 2% of households in the US, or 1%. That is one million households. Like, is there one person in 100 care about having battery backup in the event of utility outage? Probably.

We couldn't even support a small fraction of that right now. So its kind of a moot point.

Okay. John, I hope that answers your questions. Hugh, I guess we're ready for the next question, then.

Andrea James, Dougherty & Company.

Just to build on the Tesla energy conversation. What are your revenue and gross margin targets on that business? How do we look at the 2015 ramp?

The gross margin revenue obviously is going to change with time. When it's low volume, made in three months, it will be relatively low margin.

Once we get to Gigafactory up and running and high-volume, get the economies of scale working, this is just a guess right now, but maybe it's somewhere around 20%. This is not like -- it's like we don't have enough information to say exactly what that would be, but probably 20% is a reasonable guess.

In terms of volume, we are going to try to scale it as fast as we can -- as the slowest manufacturing constraint. It's easier to say what's the long term without saying exactly which quarter is it going to be in? I don't know.

I think we'll see demand for stationary storage, as measured in megawatt hours or gigawatt hours, to be approximately double that of the car. That's our best guess for long-term demand. Yes.

How do we think about your capacity, or even your cost per kilowatt hour, because it's different than a car, right? The duty load is way different for a stationary storage application. Could you just educate us a little bit on how we look at analyzing that business and the difference between that and the energy needs of the car business per pack?

There is two applications which are quite different. One is backup power, or peak-up -- the equivalent, on a utility scale of like a peaker plant, which is a high-energy application. And there is the daily cycler application. There are different chemistries, depending upon what you have.

The backup power chemistry is quite similar to the car, which is a nickel [cobalt aluminum] or a cap load. The daily cycling control constituent is nickel manganese cobalt. It's quite a lot of manganese in there.

One is meant for, call it maybe 60 or 70 cycles per year. And the other one is meant for daily cycling -- daily deep cycling, so it's 365 cycles a year.

The daily cycler one, we expected to be able to daily cycle for something on the order of 15 years. Obviously warranty period would be a little bit less than that. We expected to be something that's in the 5000-cycle range capability.

Whereas the high-energy pack is more like around the maybe, depending on how it's used, anywhere from 1000 to 1500 cycles. That's -- they have comparable to calendar lives.

Now for the high-energy one, it's important to appreciate that this is actually -- has a lot of interest from utilities because utilities have to maintain these things called peaker plants, and when there is a sharp increase in usage. You can imagine the highest energy day in California on a hot summer day where there's a heat wave, the energy consumption there is very high compared to a pleasant spring night where nobody's air-conditioning or heating is on, or for very little and commerce is not happening and people are asleep and lights are off.

That can be a huge delta. Depending upon the situation, it can be anywhere from a 5 to 10 times difference. Having a battery pack that can take out those very weird sharp peaks, like the heat wave day. You can either have a battery pack, which requires basically no maintenance and doesn't require fuel, and it's going to peak shave those troublesome days. Or you can have a power plant that requires fuel and maintenance and it's always got to be maintained and gets -- it takes time. You can't just start it up in three seconds; you've got to let it have a bit of notice.

The high-energy pack is actually very economically competitive in those sorts of situations. The high cycling pack is really great for if you've got some sort of wind or solar situation, that's where the high cycling one is really great on the utility scale.

I should say we expect most of our stationary storage sales to be at the utility or heavy industrial scale. It's probably, again just a guess because early days, 5 to 10 times more megawatt hours will be deployed at the utility and heavy industrial scale then at the consumer scale.

So, the Powerpack would be the one that is the heavy duty, the big production sales one, not so much the Powerwall. Powerwall is great, but it's, like I said, it's probably only 10% or 20% the size of the Powerpack demand.

Maybe one point on the cost structure. There's definitely a lot of commonality in the supply chain, and even the manufacturing base on how we do the modules themselves, for the Tesla Energy products along with the vehicle products.

As we ramp up production on the Tesla Energy products, there definitely are benefits to the vehicle costs of energy storage, from scale and from generally moving more material and kilowatt hours through that whole chain. Both business benefits each other.

Clearly given the very high demand that we are seeing for Tesla Energy products, we're actually trying to figure out if we can go from our current production target of 35-megawatt hours at the cell level and 50 at the pack level in our Nevada plant to maybe 50% more than that, or even higher. Because just the sheer volume of demand here is just staggering. We could easily have the entire Gigafactory just do stationary storage.

Right. So you have some capacity. I think you said maybe 15% of your capacity might be reserved for stationary storage. Maybe I read that in an article somewhere, I'm not sure.

15 gigawatt hours, or one-third of the 50 gigawatt hours that targeted at the pack level. 15% is 15 gigawatt hours, or roughly 30%.

Okay. You might add, now, you're looking at doing another 50%, and growing the capacity 50%. You have the space, I guess, there in Nevada to do that?

We do. I don't want to make that a prediction. That seems like the thing we should do, so we're investigating that. It seems like the logical thing to do. So we are going to try to do that.

We're fairly confident at this point, like I said, that the entire Gigafactory output could just do stationary storage. Then it's like, wow okay. But we need to make cars, too. We just try to make the factory bigger.

That's the total logic. (laughter) It's not more complicated than that.

But given the choice between making -- sorry, and then I'll get off. Thank you so much for your time. But given the choice between making a pack for a car and making a pack for a stationary storage application, how do you pick which one gets the priority?

Well, I guess we'd pick cars, because we've got this whole other plant here in Fremont making cars. Cars would get the priority.

Yes, that would be the logical priority. It really feels like, man, the stationary storage demand is just nutty worldwide, it's just crazy.

Great. Thank you.

Hughie, we actually have a question e-mailed in from one of the analysts that is out of the country. I will read those next. We'll start with the energy storage questions.

Guys, do our years in R&D work in vehicle batteries contribute benefits into this Tesla Energy business? Whereas we've drawn some learnings from the vehicle business for Tesla Energy?

We learned a lot about battery packs, obviously, and how to make them and have them last for a long time and work in high temperature ranges and that kind of thing. I think there are a few points of clarification that should be made regarding the Powerwall, Because a lot of people confused about the whole inverter issue.

The Powerwall does include a DC-to-DC inputter that can interface directly with a solar panel installation. If somebody has a solar panel installation, they already will have a DC-to-AC converter for the solar panel system. No incremental DC-to-AC inverter is needed.

In some of the analyses we've seen online, by people who think they're experts, they don't seem to realize that there is a DC/DC inverter. We haven't been, I guess, quite clear about that, which we need to be.

If you already have a solar installation or you're going to get one, the AC inverter is already there. That's an important point in considering the cost of the system.

Then the operating temperature, it's actually capable of operating at a much wider band of temperature. We've got to fix that specification that's stated on the website.

In general, most of the learnings from the car R&D and engineering carry over. It's part of why we can do this so quickly.

We knew where the car works, the pack would work. The car works pretty much everywhere. In fact, it does work everywhere. The pack will work everywhere, obviously.

If anything, it's quite over-engineered for a battery pack that doesn't have to move and gets mounted on your wall. I think that's going to be a great benefit for reliability and longevity of the pack.

Then one for Deepak. Would you be willing to provide any breakdown of the capital -- the CapEx spending in Q1?

The CapEx spending was primarily driven by the capacity expansion that we're doing for Model X, including the tooling spend. Then of course we are putting in the new paint shop, which is a very significant investment. That will be ready in Q3.

Then we have also had investments in the Gigafactory and our sales and service network. That's probably the order in which we've been spending.

Great. Thank you. Okay, Hughie, let's go back to the queue, please.

Colin Lang, UBS.

My first question, staying on the stationary storage topic. How do you view the -- you mention there is two different chemistries. How does that work with the Gigafactory?

Does that require all separate lines? Is there any more complexity? Is it all within the same facility? How do you view your chemistries versus other competitive options out there, there are other people who are bullish on things like [Sanken], other more basic chemistry?

They could still both be built in the Gigafactory. We have multiple lines operating in the Gigafactory. If one line is building nickel cobalt aluminum chemistry and another is building nickel manganese cobalt, that still works just fine. A lot of the cell assembly and then the processes that happen after that are almost identical.

They look the same at the module impact level. You can't really tell just by looking at them. It's the internal chemical constituents.

If you think of the -- right down the palm, the cell is a chemical engineering problem. It's like a little can of chemicals. And then the module and pack problem is mechanical engineering, electrical engineering and software engineering problem.

If you have a small number of cells, then as an overall engineering problem, it's mostly about the cell. Once you have a lot of cells, than the intellectual property challenge -- the intellectual challenge becomes more at the mechanical, electrical and software level, which is where Tesla does quite well.

That's way don't really see -- nobody else has a pack like ours. The cells are relatively generic. The module and pack is not.

We've looked at pretty much every chemistry couple that could possibly be relevant for this and are confident that what we are using is going to be the best. If there's something better, we can also adjust and change over time. There's a lot of flexibility.

I'm not sure if the exact (inaudible) would be, if you look at a laptop, and you can say, a laptop's really just like a Intel CPU and some micro and DRAM chips. Big deal.

(Laughter) what is any computer company actually doing? They are doing a lot.

What do you view as your competitive edge? It sounds like you think it is the software and electronic integration, the competitive advantage. In your presentation you actually mention you encourage other companies that build -- to the market, as well.

I think in terms of the electrical engineering, the mechanical, the software and the overall aesthetics, and having something that really just works and is easy for consumers and utilities and large industrial applications to just order and it just works, and it's just is there as a fully integrated system. That's the value that Tesla's adding.

I think there's also substantial benefit to the track record of the automotive fleet. That's something that a lot of the newer start-up companies or different technologies in this space really struggle with, especially in the utility application where you need to be sure it's going to work for 15-plus years. We have a sense of scale and a great track record on the vehicle fleet that you can really pull from data and pull data from, and understand how it's going to work.

I shouldn't say like we are wedded to a particular chemistry or anything like that. We just want to use the best chemistry. Whatever little can that contains the cathode and anode and separator and electrolyte, whatever the best constituents of that can are, that's what we'll want to use.

We'd love it if someone could come up with a better internal chemistry for the cell. It's worth noting, like, nobody has sent us anything, a sample cell that's better than the cell we're producing, or something that we will produce in the Gigafactory.

We'd love it if somebody would do that. They just haven't. So there's al these things which are big on promise and short on delivery when it comes to battery chemistry.

It's just a real hard problem. Hardly a week goes by that there's not some alleged breakthrough in batteries. What they'll do is they'll cite the power but not the energy, or they'll forget to mention that it only lasts for 50 cycles, or uses an incredibly exotic raw materials.

Or one component out of the battery pulled separately.

Yes, exactly. It's not like the full picture. It's not like we don't want a better chemistry to exist than what we're using. We would love it if there was such a thing, and I'm sure there will be improvements over time. We'll implement them as soon as they are remotely production-ready.

Thanks for the color. Just one last final question. Any color on -- a financial question -- other expenses is up quite a bit. It was $22.3 million. What was really just driving that in the quarter?

That was essentially, revaluation of primarily our foreign currencies that we had at the end of the quarter. The dollar was pretty clearly strong against foreign currencies there.

It's also, in most cases, unrealized losses. As the currency moves in Q2 in different directions, that could have an impact, as well. A positive impact, is what I'm --

Okay. So we should consider most of those one-time in nature, because of FX, balance sheet FX?

Yes. They are not related directly to our operations. It's simply the revaluation of our currencies into Company items.

Okay. Thank you very much.

Rod Lache, Deutsche Bank.

Couple of questions on stationary storage first. One is it seems like utilities move quite a bit slower than other markets. I was wondering when you are thinking about the expectations for the trajectory of growth, if you had the capacity today, how long do you think it would take for you to get to that 15-gigawatt hours?

Also related to this business, it seems like you're using third-parties for most of the distribution installation. If you achieved maybe a 20% gross margin, can you pass along any thoughts on how we should be thinking in terms of SG&A is allocated?

I think it's pretty early days. We are being super speculative at this point.

Maybe one point on the distributors is we're only focusing on that approach for Powerwall, in most cases. So with utilities, we are building those relationships directly.

It's not something that is really effective to go through a distribution channel, and it's something we've also been doing for a number of years. It didn't just start last Thursday.

We've built these relationships partially through vehicle infrastructure and charging questions and back and forth, super-charging as well. There's actually quite a lot of trust built between Tesla and many of the utility companies, which is very helpful to grow that business faster.

Thanks. A clarification on the comment about the fact that a DC/AC inverter is already there for solar. Does the fact that it needs to be bidirectional change the nature of the cost, or not really?

No, not in most cases. Because if you already have solar, typically the energy to charge the battery can come from the solar panel. I think maybe one way to think of it is if the battery pack, the Powerwall with it's internal DC-to-DC converter, it can act much like a solar panel.

It can match voltage with the solar panel array, and it can feed power back into that. The existing inverter doesn't particularly see a big change.

Okay. Got it. On the auto business, do you have any preliminary expectations on the mix of the 70D versus 85? Any thoughts on how that affects margins?

Lastly, a question for Deepak. In the past we were able to look at your supplemental information on leasing where you disclose the value of the leases delivered and the value that you are booking this at on the balance sheet, which is presumably your cost. And it sort of approximated your gross margins, but this quarter when you do the math, the implied margins were something like 44%, not like 29%, 30%. Is there something unusual there?

I will answer the second one. No, there's nothing unusual in those numbers. As the overall -- what you're seeing in the supplementary information is the new cars we've delivered.

What's coming through ultimately is the cumulative impact of all of the leasings that we have done. But to my knowledge, there's nothing else unusual.

Okay. It seems like a pretty high margin, though, when you look at it that way. Anything on the mix, 70D versus 85?

It's still pretty early to be predicting mix because you have to distinguish between what initial demand versus sustained demand. It looks like there maybe going to be comparable number of 70s versus 85s. But this is difficult to -- this is extrapolating on very little information.

I think a key point I should make before this call ends is we really optimizing -- at this point, going forward, we are to be optimizing for the operational efficiency of the Company as opposed to specific quarters. This is -- I feel like what we've done in the past where we've really had to scramble at the end of quarters, and sometimes have not a great customer experience at the end of quarters, it's not really the right thing for the Company.

So we are going to be operating more for steady-state efficiency. And that means the quarterly fluctuations could be a little higher. But in the long-term, it will be better. That's an important consideration.

Also unlike other car companies, Tesla's sales are only recognized when they arrive at the end customer and we've received payment and all the regulatory docs have been processed and all that. Our sales are always true sales. They're not sales to the channel.

In other words, other car companies are -- their sales to the channel. So it's possible for them to make numbers work where they're selling to their channel as opposed to end customers. In our case, it's always end customers.

As we are shipping to Asia and Europe and across the US, it's fairly easy for there to be a plus/minus 5% difference in deliveries, just due to logistics issues. It's not a huge variation, plus/minus 5%. That's kind of like -- if a ship is late or early, it can affect -- it could be 500 cars that are affected. That's 5%, which we've seen in the past.

That's an important thing to bear in mind. The number that's much controllable for us is the production number.

Our production predictions are -- for the vehicles are, at least for the Model S, not so much for a new vehicle but for existing vehicle, our production predictions are a lot more controllable. And as people at this point realize, the demand is different from production. Our cars are almost all ordered in advance.

We really don't see any demand issue. You can see that by delivery times of the cars on our website. Some of these people will interpret a quarterly delivery number as somehow being related to demand. This is not the case.

It's usually related to, we produced a certain number and then we were able to deliver a certain number in the quarter, and there's a whole bunch of cars on ships or on trains or on trucks. And that's -- but because Tesla is different from other car companies, applying the other car companies' template can lead to incorrect conclusions about Tesla.

Is this making sense? Should I elaborate further?

It make sense.

To add to what Elon said, to be more efficient operationally, we are shipping more by train rather than trucks. It's more cost-efficient and creates less damage to the car, gives a better customer experience. We are taking (multiple speakers) actions.

I guess my question is more geared to trying to assess the effect on ASPs and margins as you are changing the mix a little bit. I appreciate that.

Yes. To answer your question on margins, I think there's two things that are having -- the average sales price is going to decline a little bit, but our costs are improving as well. We see that more or less netting out. So that the efficiency improvements offset approximately any average selling price change.

Got it.

And if In the process we're growing the market, which we truly are by hitting the right sweet spot, then we're doing the right thing (multiple speakers). The 70D has a higher gross margin than the 60, for example, and 60 was just not hitting the market, just didn't have the right attributes that our customer liked.

It was a low [day] rate. We brought a product which is far more compelling and much more competitive with other cars.

Great. Thank you.

Colin Rusch, Northland Capital Markets.

As you look at the growth of the organization, the integerity or -- of the culture and the workforce, I know that your employees are extremely motivated. Could you talk about how, as you grow, how that's grown and changing and what you're doing to maintain that integrity as you go forward?

It is tricky as companies grow to maintain a consistent culture. I think we're doing okay as measured relative to most companies.

We do need to get people to think differently and have an expectation of innovation. I think we are doing okay on that front.

I think having a general mix of people that have been here for a very long time and have seen multiple programs and seen the Company when it was even much smaller than today and more scrappy and had to go through even tighter financial difficult launches, things like that, really helps maintain that culture. And we try and pair different managers up with new groups so that that culture sort of infuses throughout.

Yes. Our growth rate -- like this year, personnel-wise, it's, let's say, leveled off. It's still fairly significant, but it's like -- our net personnel growth this year will probably be 20% to 30%, which is, I think, a manageable number for integrating with people that already have the Tesla culture.

It's already a pretty big improvement in productivity, because we're looking at, let's call it at the upper end of that 30% personnel increase, but 100% increase in vehicle volume. I think that's a pretty good indicator of productivity improvement, just like how many people do we have and how many cars are we making?

The second question is really around some of the choices that you made with the battery product and why choosing 10-KW and the size, and 220 pounds is actually pretty hefty for a garage wall. Then also the battery management system I think there's a lot of confusion around where it's located and what the real functionality is, as you look out at interfacing with utilities and the signals that you get from the market in terms of looking at demand charge-offset, demand response and some of the other advanced functionality that are going to be able to be monetized in business models with the product.

Can you clarify where that BMS system is [residing] and who owns that technology? And why you guys made those choices, along with the weight and size choices?

Some of the different revenue streams you are talking about our mixed between Powerpack and Powerwall. The BMS system lives inside the battery pack in both cases.

But with the Powerpack and the more utility-sized installations, there we often will have a site master computer or master controller that controls multiple Powerpacks. That site controller is what then interfaces to the utility, or maybe a commercial customer, to run the schedule charge and discharge that would be appropriate for a given application.

On the Powerwall, it's a bit of a different situation. Still the BMS lives inside the battery pack, but in some cases the inverter, maybe the system that's deciding how to manage energy and the overall house. That can depend on which type of inverter we're using on how that works.

In terms of the size, we really, for Powerwall again, we optimized the size around what was the most common photovoltaic size and also what we felt was kind of the smallest modular increment for backup. It's pretty key to note here that you can install multiple Powerwalls together. Having 10 kilowatt hours doesn't mean that you can't very easily put in 20 or 30 or 40.

In many cases, I think people will to have the backup matching that they want to see. We felt 10 kilowatt hours was the lowest common elements. If you go smaller than that, you start to run into worse economies of scale and more of the system that's not related to storage.

Great. I will have some follow-up offline. Thanks so much, guys.

Hughie, before we go to the next question, I just want to give everybody a status update here. A lot of important things to talk about on this call, but we are 50 minutes into the call and we still have seven callers in queue. Just let you guys think about how we manage time here. Hughie, let's go to the next question, please?

Adam Jonas, Morgan Stanley.

Evening, everybody. First question on the Model S and the release candidates. How many release candidates have you produced?

What have been some of the issues with the recent production? And why the slight -- am I seeing a slight further delay by releasing late 3Q from what I thought was more of an August time horizon for initial deliveries? Anything that might be behind that that you would highlight?

I think we'll pass on answering super-detailed questions about the X ramp. The thing that really matters is not like when the first deliveries of the X occur, but rather when do significant deliveries of the X occur.

For the S we had quite a long ramp from -- we had like six months from the very first deliveries to significant volume. We're trying to compress that to maybe two months or three months at most, cut that in half or more for the X.

We want to make sure we're really delivering a product that has been thoroughly validated in hot and cold weather and through millions of miles of travel and everything. It would be easy for us to do some initial deliveries in August. That would be pretty easy.

But we'd only have -- we have door handle issues, like people are aware of that with the S. We don't want to have falcon-wing door issues with the X. We want to iron everything ironed out and make sure it's good, and then deliver at high volume.

Effectively, we would create like a captive fleet, and iron out the issues with the captive fleet's quite big, several hundred vehicles, basically. Maybe that addresses your release candidate question.

Just make sure that those several hundred vehicles really work well in all circumstances before we start delivering cars en masse. We're going to go from [700] hundred cars to 1000 a week pretty fast.

A second question, in the interest of time, and then a final question. Apple has still, unofficially perhaps, been making inroads in the building of its vehicle engineering and transportation capabilities, according to a lot of sources, and making investments, both in physical and human capital. Are you starting to feel a greater sense of competition with Silicon Valley parties versus your ability to attract and retain key software and automotive engineering talent? If Apple were to get into the electric car business, would you see this as a positive for broader consumer acceptance of electric vehicles?

I certainly hope Apple gets into the car business. That would be great. But no, we are not really seeing significant attrition of engineers to Apple for anything, car or otherwise.

Actually, anyone can figure this out by going on LinkedIn. LinkedIn can produce statistics on what the relative flow of people is from one company to another.

I think it's like something like -- if you look at the trailing 12 months, Intel has recruited 5 times as many as people as Apple as Apple has recruited from Tesla. It's like some fairly high number.

Thanks very much.

Brian Johnson, Barclays.

Just really want to ask about the balance between stationary and automotive. It's not unusual for growth companies to change their focus as they evolve. At least one of the stories around PayPal, it was going to be a cryptographic PDA-focused company until it refocused, perhaps through your help, Elon, on web-based payments.

I'm hearing you talk about stationary storage as being able to ramp faster, perhaps, although you didn't answer this, maybe with a lower CapEx and OpEx ratio than the auto business. Yet at the same time, auto takes a lot of capital, takes a lot of OpEx.

Sergio Marchionne pointed out there is a lot of value-destroying capital spent in the legacy industry. Is it conceivable that instead of savings six 70-30 auto batteries, it could go the other way?

It's possible. We are really super in guesswork territory here. All we know right now is we have demand well in excess of our production ramp.

We know right now, the thing we should work on is trying to increase our production ramp, not trying to increase demand. So that's kind of like the only thing we know for sure right now.

If you look really long term, you could say, what's the total energy -- how many -- what's the total terawatt hours of installation, of factory installation that's needed to go to a fully renewable global economy, taking into account transportation, current electrical needs, full electrification of transport, electrification of all heating and cooling and then the current industrial uses and commercial uses of electricity. To go fully electric, the transport is about 0.5 the size of everything else. That was our global macro calculations.

Ultimately, we think things will get there, could take a long time. But that's where we think things will end up, and that's where we get to the kind of 2 billion Powerpack number. Whether it's made by Tesla or other companies or some combination, that's the number you need to get to, to go fully electric. But how things track between now and then, it's difficult to predict.

In terms of the ROIC, would you be thinking that stationary could be a better ROIC business (inaudible) than, say, then mass market or mass affluent automotive?

It might. It would be nice to get maybe a couple of quarters of experience after making the big announcement to really -- before we -- our degree of uncertainty will diminish quite substantially with each passing month. Certainly a year from now, we'd have broken up a really good idea of it.

It's super speculative at this point. We just know that like, man, there just no way we can meet the demand that we are seeing right now. We've got to scale stationary storage as fast as possible.

You mentioned last time that some CapEx was going to product development in the shareholders letter for Model 3, if Model 3 wasn't in this. Does that mean you might be delaying some Model 3 investments, or just the other things kind of swamp it?

The other things kind of swamp it. We need to make sure -- we're doing quite a bit of advance work on Model 3. This doesn't come out to a lot of CapEx.

Just to clarify, Model 3 spend in the initial stages is more engineering spend.

It's basically design studio.

Exactly.

It's very early engineering prototypes. It's just not very cash intensive.

CapEx comes at a later stage when you have all the designs finalized, you are kicking off tooling. We are in the early stages, as to be expected at this point in time.

Okay. Thanks.

Brian, you might want to ask a follow-up. Does this impact our Model 3 timing? I will throw that out.

Well, go ahead.

We are having to show off the Model 3 in approximately March of next year. Again, don't super hold me to that month. That's our aspiration. Then be in production with the Model 3 in the -- I'd like to say mid-, but probably closer to late 2017 timeframe. Late 2017 is probably more realistic.

In the meantime, the Gigafactory sounds like it's maybe ramping up faster, but perhaps those batteries could go not just to Fremont cars but to Fremont stationary storage? Or to stationary storage, not necessarily touching Fremont?

Yes, it would just be -- actually, just right there, not even go to Fremont, really. It would just be built at the Reno factory or Sparks, [technically] at Sparks. Then to the Nevada factory. And go from -- ship from there to customers.

Starting Q1 next year, that's our plan for all the stationary -- all the Tesla Energy products.

Okay. Thank you.

Ryan Brinkman, JPMorgan.

I have a two-part question on cash. Firstly, on the last call you mentioned reaching free cash flow positive in 4Q this year. I'm just wondering with 1Q behind you how you think you are tracking relative to that goal?

Secondly, if you could perhaps comment on your capitalization and liquidity overall, and the potential for, or desirability of raising any additional capital? Thanks.

We do expect to be free cash flow positive in Q4. That doesn't change. As we go along, clearly we are optimizing for efficiency which results in increased of our finished goods inventory. It makes sense for us to then establish some asset pipelines of credit, which is backed by our finished goods inventory or raw materials. We'll take those actions to make sure we have a solid balance sheet.

Yes. I think it's -- I think things are looking pretty good for Q4. It's hard to predict full quarter exactly because that's the whole quarterly boundary, and where does the exponential ramp of production fall exactly on that start of Q4 boundary? But, I think it's extremely likely that cash flow is really good at the end of Q4.

Okay. That's great to hear.

Yes.

Thank you.

Trip Chowdhry, Global Equity Research.

A quick question, two quick questions. First is regarding the residences which may not have the solar panels and say that residence goes and buys two Powerwalls. Does that customer have to buy two inverters, or one inverter is sufficient for as many Powerwalls the customer buys?

It's basically whatever the capability that inverter is. So typically the inverter would be capable of handling -- one inverter is probably capable of handling, depending on the situation, up to maybe four or five Powerwalls.

The second question I had was regarding the double-back tour you had in 10 cities. And that was (inaudible) indicative of very successfully went with the reservations in autos really skyrocketing.

I think it was only Canada and USA. Wondering if you could expand that tour to other continents?

Yes, absolutely. We just did a big thing in Europe through Germany. There was actually -- I think there's quite a big tour happening from Slovenia, I think, through most of Europe. There's actually quite a lot -- our the sales in Europe lately have been pretty great, really strong.

They did the tour in Germany, too -- and Europe, too. Sorry.

It in fact is been a big hit in Europe.

If I can ask one last question. I was at your Lathrop factory the other day. And there was some interesting construction happening behind it with huge metallic frames being installed. What kind of production could be happening there?

It seems like the factory size has increased by almost 40%. What is happening there? Any parts you can share with us? That's all for me.

We're establishing a big casting foundry and machining facility there. That's intended to support activity in Fremont. That's what's happening in Lathrop, primarily.

Patrick Archambault, Goldman Sachs.

Thanks for squeezing me in. I'll be quick. Just wanted to build on, I think Ron Brinkman's question, on the fourth quarter cash flow guidance. Does that include income from those warehouse facilities to offset leased vehicles, or is that exclusive of that?

In our projection, as Elon said, was the end of the quarter as we ramp up production, we should be free cash flow positive on a pure sense. Frankly, our operations should be looked at without the impact of leasing. And that's a purer measure of operations, which is a lower threshold, I get it. But that's a unique or a distinct decision we've made to be in that business.

(Multiple speakers) Hold on a second, Pat. Deepak, do want to explain about the classification of cash inflows within the statement of cash (multiple speakers)?

If you look at our Q1 cash flows, for example, we received $155 million from either the warehouse line or banking partners related to our leasing business. That cash flow does not show up as an offset in our cash flow from operations. It shows up in our financing activities.

Our cash flow from operations shows up $131 million negative. But if you offset that against the $155 million that we received in the leasing business, we were slightly better than breakeven, purely on an operations basis. So our real cash flow in Q1 was driven by the strong CapEx spend ahead of the Model X launches the way we tend to look at the business internally.

Okay. Got it. Appreciate the clarification.

The other quick question I had is just I was a little surprised by the -- I think you guys had obviously very good gross margins this quarter relative to some of the headwinds that you were facing. I guess I was surprised to see that ex-ZEV they go from 26% to 25%.

With some of the increases in pricing that you are putting through and some of the components that you are buying outside of the US with a stronger dollar now, which where there was a little bit of a lag, I understand, that seemed a bit conservative. I don't know, maybe it ties back into Rob's earlier point about mix. But I guess that was one area that I was a little surprised.

Just to clarify the price increase, that we've announced, it's not going to help us in Q2, given our book of business. It's going to primarily affect us in Q3. Also in April, for most of April, the dollar was pretty strong.

Is just strengthened in the last week or 10 days or so, and we were continuing to deliver cars. That strong dollar impact does rollover sequentially into Q2 for us, despite the lag effect of good news from a strong dollar.

As we said in the letter, we fully expect the mix impact to be offset by cost reductions that we will achieve. It's really related to FX that we see the sequential drop.

The FX was really -- that was a really tricky thing to deal with. We had quite strong European sales. Then the FX impact was quite significant. [

See, if you've got] a car that is 25% gross margin and you see a 20% change in currency, it's like, wow, okay. That's not super easy to make up that difference.

To your point, and maybe this is a broader comment to make, that as we look at our Q1 non-GAAP loss that we have of $45 million, a big chunk, or most of it, could be attributed to the dollar's strength that has happened over the last six months. As we've indicated, our Q1 top line came down by about three percentage points, and then we had the $22 million unrealized losses from currency.

You combine those two and there's an offset in gross margin due to lower cost of parts we buy in foreign currencies. You consider this whole mix, and you can see the foreign currency change over the last six months can explain most of the losses we had in Q1.

Maybe just like that last piece is kind of interesting. How much are you able to take advantage of that natural hedge you just described, right? You are buying components in currencies that have also depreciated, right? Batteries are one obvious thing. I'm sure you are sourcing a lot of stuff in Mexico.

Are you closely optimizing that? Or is that something that can be exercised more?

For better or worse, one can look at it as a pro or a con, depending on the situation, basically about 60% of the car, maybe a little more than 60%, is US and Canada.

We may have -- we certainly have lots of, say, European companies that supply us, but they will supply us from plants in the US or Canada or sometimes Mexico. So that's -- this really isn't a North American-built car, at least not for real.

So, it's like, as Deepak's saying, it's slightly helpful if the dollar strengthens relative to the yen. But then since we have quite strong European sales, this is -- I think it's like 5% of our parts or something come from Europe. It's pretty low.

Right.

If our sales are 30% in Europe, then obviously it's not as much of an offset as [one] would like. I actually thought it would be higher than that, but it's like -- because I'm looking at the names of the suppliers, but not -- what matters is where -- not the name of the supplier, if their headquarter's in Germany, but where is the plant?

Got you. Okay. Thank you for the helpful color there.

All right. I've got to run to another meeting.

We have two more question-askers. If we can quickly go. Hughie?

Brad Erickson, Pacific Crest Securities.

Thanks for taking my question. Just a quick follow-up from the last one on the automotive gross margins. I guess ex-currency, can you quickly talk, maybe a couple examples where you are looking to remove more cost from the cars itself?

I think we can all appreciate there's continued scale advantages over time. If you could talk but other cost saving opportunities we should be thinking about with both the S and the X, and what inning we are in, in terms of achieving those savings? Thank you.

A big factor is labor hours per vehicle. That has steadily improved. It was quite bad in Q4, particularly with the P85D ramp. We've made good progress in Q1. We'll continue to make good progress on that through the rest of the year. So it's labor and direct overhead (multiple speakers).

We're focused across every line item of our COGS. So material cost reductions, freight both inbound. We spoke a little bit about outbound freight efficiencies.

We had a lot of expediting.

Yes.

That big port strike was super unhelpful. That also hurt pretty badly.

It was like the, I think, the worst port strike of the 21st century. We also are soldiering through that. We had to air freight a ton of stuff while everyone else was simultaneously trying to air freight a bunch of stuff.

It cost us several million dollars.

Just to air freight, yes. Cutting back on air freight is really helpful. You had direct labor hours.

We continue with several engineering and commercial actions on material costs. It's a long list, not appropriate to go through. As a Company, we are focused on Model X, but S is also certainly pulling gross margin there as part of our priorities.

Yes.

Got it. That's great. Thank you.

Tyler Frank, Robert W. Baird.

This is Ben. Hey, guys. Very quickly, you mentioned the Gigafactory, and then a lot of talk about stationary storage and increasing of the Gigafactory. Does that change in any way you are looking at partnerships, green partnerships? And then maybe an update on timing there.

And then Elon, you talked about having a fleet of a couple of hundred X to get out on test. And you mentioned Model 3, showing it next March. When should we see it, the Model X in its final version? Thanks, guys.

Well, no particular change to the partnership model for the Gigafactory. I think, as Elon said, it's early days on trying to make some of those changes in direction. Largely, it's just a bigger scope of opportunity for a lot of the people we are already working with.

It is continuing to go well. We are actually beginning to hire operational people at the Gigafactory, and beginning to step up on that ahead of starting to train and then ramp up production of the stationary products early next year.

Anything else, Ben?

Just on the Model X, when we can see that?

Elon, you want to talk about Model X? Have had a number of questions from investors, and Ben's question here, about when you will be able to configure Model X online, when we'll show that to the public, how that actually deploy?

Yes. Probably three months from now -- no, (inaudible) I started (inaudible) in May. So it's -- this is -- probably July is when will ask people to do the configuration. It will be very similar to the S. It's not going to be super -- no big surprises there. Very similar to the S.

We certainly have a lot of customers who've been waiting a long time (technical difficulties) To get the car.

We look forward to that.

This is really a great car. Because it has a low center of mass, it handles like a sports car even though it's an SUV.

The battery pack's in the (inaudible) pan. It's got incredible acceleration because there is basically a performance version of it, like there is for the P85D. The performance is surreal. It's like nothing else is comparable.

Great. Thank you, everyone, for joining us this afternoon. Everybody have a great night. Hughie, thanks for your help, and we talk to you all next quarter. Bye-bye.

Pleasure, sir. Ladies and gentlemen, this will conclude today's conference. Thank you for your participation, and have a wonderful day.