特斯拉 (TSLA) 2017 Q2 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Tesla Second Quarter 2017 Financial Results Q&A Conference Call.

(Operator Instructions) As a reminder, this conference is being recorded.

I would now like to introduce your host for today's conference, Mr. Jeff Evanson.

Mr. Evanson, you may now begin.

Thank you, Sharie, and good afternoon, everyone.

Welcome to Tesla's second quarter 2017 Q&A webcast.

I'm joined today by Elon Musk; J.B. Straubel; Deepak Ahuja; and Jon McNeill.

Our Q2 results were announced 80 minutes ago in the update letter we published at the same link as this webcast.

During this call, we will discuss our business outlook and make forward-looking statements.

These comments 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 filings with the SEC.

(Operator Instructions) But before we jump into the Q&A, Elon has some opening remarks.

Elon?

(technical difficulty)

Yes, apparently, we're having trouble hearing Elon.

Try a different microphone.

Hear me?

(technical difficulty)

We're getting messages from dealers that they cannot hear Elon.

I can hear you, Mr. Evanson, so if he can speak from your microphone, you're coming in loud and clear.

All right, we'll try that.

(technical difficulty)

Sharie, are we in the call now?

Yes, your line is open.

All right.

Thank you.

All right.

We apologize, everyone.

(Operator Instructions) Elon, over to you.

All right.

Thank you.

My apologies.

We actually tried a new audio system with a bunch of individual mics.

That seems to have malfunctioned, so we went back to our standard conference call object.

Anyway, just want to confirm, people can hear what I am saying.

Okay, great.

So first of all, I want to say that Friday night was an amazing time for Tesla.

It was the -- one of the most important days in the history of the company.

It's something we've been striving for, for 14 years.

It's the car that we've -- the Model 3 -- hanging with us [30] production Model 3s was just an incredible milestone in the company's history.

We wanted to make a great, affordable electric car, which is the fundamental thing that is missing.

We wanted to make that from day one.

And if we could only have done it sooner, we would have.

And I'm glad that this day has come.

What we have ahead of us, of course, is an incredibly difficult production ramp.

Nonetheless, I think we've got a great team.

And I'm very confident that we'll be able to reach a production rate of 10,000 vehicles per week towards the end of next year.

And we remain, we believe, on track to achieve a 5,000-unit week by the end of this year.

So I would certainly urge people to not get too caught up in what exactly falls within the exact calendar boundaries of a quarter, one quarter over the next, because when you have an exponentially growing production ramp, slight changes of a few weeks here or there can appear to have dramatic changes, but that is simply because of the arbitrary nature of the -- of when a quarter ends.

So -- but what people should absolutely have 0 concern about, 0, is that Tesla will achieve a 10,000-unit production week by the end of next year.

So if you can sort of say where we came from, the Roadster, where we were making only 600 units a week, where the nonpowertrain portion of the car was made by Lotus.

And we did the powertrain and the final assembly of the car, and then we went from that to 20,000 units a year of the Model S, a far more complex car where we did the whole thing.

And then we -- with Model 3, even more vertically integrated.

I think people should really not have any concerns that we will reach that outcome from a production rate.

We're also very confident about costs.

We feel we gained a lot of experience.

We certainly aspire to learn from the mistakes of the past.

And I think we largely have.

Deepak will go into some of our margin expectations there.

And unlike, say for example, the Model X, where the mistake that we made, I obviously take primary responsibility here, was having far too much advanced technology in version 1 of a product, it's an -- Model X is an incredible car, but it was overreaching for the first generation of a product.

But in the case of the Model 3, we've strived hard to simplify and make sure that it has everything that's necessary to be a fantastic car.

If you see the reviews, the reviews are -- you really could -- one could not ask for better reviews.

And I'll just sort of give you one little anecdote, which was -- which I found quite surprising is that when we were giving test drives to -- or tests -- the journalists were doing -- were driving the car and doing test drives.

About 80% of the journalists said that they would buy the car themselves.

Most of the remaining 20% said probably.

This is crazy.

I've never seen anything like it.

So this is a very good sign.

It should also be noted that one of our big concerns was that Model S, particularly, and Model X demand would suffer with the introduction of the 3. In fact, this has turned out to be the opposite situation.

Model S and X demand increased with the release of 3.

Jon, would you like to just elaborate on that?

There was -- we did express this as a concern, and it was a big concern, but it has turned out to be a pleasant surprise.

Yes, I think that's right.

Not only, as Elon said, we expressed it as a concern.

We had positive comps both year-over-year and quarter-over-quarter and in orders in the second quarter.

But since then, orders have accelerated in July as we noted in our shareholder letter.

And it accelerated further since the handover event on Friday for the Model 3.

So it clearly shows that S and X as our flagship products have a strong position in the market and strong demand.

And that's super encouraging that we've got those -- a strong product lineup with 3 cars that are proving to be very popular in their individual segments.

Yes.

In fact, I don't know -- I think we mentioned some of this in the earnings letter, but just some of the key stats on July orders for S and X were...

Yes.

July orders were 15% higher than our Q2 average weekly order rate, so we've accelerated off of Q2 into July.

And as we noted in the shareholder letter, deliveries grew by 53% compared to the Q2 '16 in a flat luxury vehicle market.

So we're gaining share in a flat to market down -- to down market, and the order rate's accelerated.

Yes.

So July was one of our best months ever.

Yes.

Again, contrary to our expectations, I want to emphasize.

Of course, we know who knows if this will continue, but all indications are that it will, so that's very exciting.

As a side note, we're also making great progress on our internal Autopilot software.

It's getting better and better.

I'm really, really excited.

I test drive the latest development release as soon as it comes out, and I'm like, this is really getting to be something special.

Because, yes, it's really -- and I think it's going to accelerate from here.

And the talent that we're seeing drawing on the technical side for Autopilot is really world class.

I don't think there's -- it's unmatched anywhere, I would say.

So let's see then.

Model 3 orders are -- net-net orders, there's not that many cancellations, there are about 1,800 a day.

It's important to emphasize, there -- you can't see the car unless you want to look at pictures online, you can't test drive the car.

You have to put down a $1,000 deposit.

We're not promoting the car.

We're not promoting the car.

If you go to our stores, we don't even want to talk about it really and -- because we want to talk about the thing that we can supply.

If somebody orders a Model 3 now, it's probably late next year before they get it.

We want to give people the car where it's like maybe a 1- or 2-month wait for an S or an X. And I think the point that we're trying to make that the S is still a superior sedan seems to come through -- and it is true.

Things get a little confusing because of the nomenclature of being Model 3, but Model S and X, which is I guess sort of my fault being too quick -- too clever for my own good there, because it was supposed to be the Model E. As you can tell, I have a wonderful sense of humor, and -- but then people mistook that for generation 3. But in fact, if you look at, say -- what we're really on right now, I would say is approximately generation 4. But we're on generation 4 of S, X and 3 at the risk of really confusing matters.

Model 3 is generation 4 and so -- but so are S and X. We evolve the technology all at the same time.

So overall looking really good.

And then Solar Roof.

We have installed and working the Solar Roof tiles.

I have it on my house.

J.B. has it on his house.

We have -- I think we included some of the pictures in the earnings letter.

I want to be -- I want to emphasize those are un -- the roof, there's no Photoshop-ing on the roof.

That is actually how it looks.

And it wasn't taken by some amaze -- it was like take some pics with your phone and send them over.

That's what we're talking about here, not some special lighting conditions, pro photographer art situation.

And this is version 1. I think this roof's going to look really knockout as we just keep iterating.

Now it is a very challenging technical task to get this right, get the costs good, streamline the installation process, ramp up the production.

Again, this is sort of -- it follows some of the S-curve to vehicles where it starts up very slow -- it starts up very slow, but then it grows exponentially.

Also, our conventional solar is doing quite well and generating significant positive cash flow, standard flat panel stuff, which I think is solar the right solution for any product to -- for any kind of flat roof situation, which is most commercial installations on a lot of houses or some part of the roof where it's just really not visible and, therefore, does not have any aesthetic -- it doesn't really matter from an aesthetic standpoint.

And then batteries are also making great, great progress in the battery front.

I'm hoping to do something around the International Astronautical Congress, which is in Adelaide this year, but not promising anything.

But we're aspirationally going to have a very substantial portion of the battery pack already done in about 8 weeks, which is hard because this -- we have all the shipping and logistic challenges of getting things across the Pacific.

I'm not promising anything.

It's an aspirational goal.

Team's working super hard to make it happen.

But I'm excited by the prospect, and I feel, of course, optimistic that, that will take place.

So yes, really, I think -- and we're really proud of the Tesla team for getting to this point.

And we want to thank the whole Tesla team.

We already have 33,000 people at this point who are working hard to achieve some very difficult things.

And I can be proud to work with such a great team.

So let's maybe go to -- anything else you want to add, guys?

All right, let's go to questions.

All right.

Sharie, let's open it up to Q&A.

(Operator Instructions)

Yes, not the 1 question with 8 nested questions.

Correct.

No nesting.

One question, Part A through H.

(Operator Instructions) Our first question comes from James Albertine with Consumer Edge Research.

Congratulations on the first 30 deliveries last week.

It was a great event.

Wanted to ask, if I may, my one question on capital expenditures.

Wanted to get an idea.

What comes next with respect to some of your spending on the Model 3?

And then, I guess, if I can nest one in related to Gigafactory.

It's one question on CapEx but really want to understand what the big next steps are in 3Q and 4Q as we start to kind of build out our models and figure it out from there.

Certainly.

I mean, I do want to emphasize, like, the -- a lot of this is actually very hard for us to know.

When we make mistakes, it's because we're stupid not because we're trying to mislead anyone.

I just want to emphasize, I, we, sort of we aspire to be less dumb over time.

So if I knew it, I would tell you.

It's not like I've got this like secret hand of cards that I'm holding close to my vest, and I'm not telling you.

It's just fundamentally impossible to predict the exponential part of the manufacturing S-curve.

It's crazy hard.

And S-curve is a simplification because it's really running through a series of constraints that, if you -- and so it's like a really jagged sort of upward growth, and it'll plateau, and then it'll grow rapidly, and it'll plateau again.

And then sometimes it'll go backwards because something broke.

When I said manufacturing hell on -- and supply chain hell on Friday, I meant it.

I mean -- but we know this.

Signed up for it, not blaming hell because we bought the ticket.

So I -- but I think at a high level, I don't think we should expect any significant negative surprises.

There will be -- as you look at the case, there tends to be some cost growth in CapEx for unexpected things.

We've got to expedite this.

You've got to fix that.

Or this supplier doesn't work out.

Or this machine we bought doesn't work out.

And you've got to be all hands on deck 24/7 to fix it or replace it.

But I don't expect any significant -- I think that it's relatively contained.

Deepak, do you want to...

Yes, I mean, I think maybe the other way, James, to answer your question is...

You may need to talk close to this.

Yes, I think you're asking where we're spending the money.

I think it's in the completion of the Model 3. We are paying off on the equipment.

And also we're just continuing with the construction of Gigafactory to continue to scale that, and so that's where majority of our capital...

Overwhelmingly.

Let's be fair.

Overwhelmingly is Model 3. Obviously, there are expenditures associated with Solar Roof, and with our Buffalo factory, we're trying to keep those relatively light for the next few months.

Yes.

And the marketing and sales are growing our infrastructure there and our Supercharging networks, so those are the other small pieces.

Yes.

But just on Buffalo, if I may, I really want to emphasize we expect the Buffalo Gigafactory to be a powerhouse of solar panel and solar glass tile output.

It is going to be a kickass facility.

We've made that commitment to the state of New York.

We are going to keep that commitment.

So -- and then we're also thinking hard about where do we put Gigafactories 3, 4, 5 and 6. We expect to keep the majority of our production in the U.S. and -- but, it's obviously going to make sense to establish a Gigafactory in China and Europe, just to serve the markets there because it's not too -- (inaudible) in California and truck them halfway around the world, particularly when you're trying to make things as affordable as possible, that really hurt.

We really want to get -- make our cars as affordable as possible, and so that does require some amount of local market production, particularly for the mass market vehicles in order to make it as accessible as possible.

So we think hard about that.

I think we'll have some announcements on at least a few of those locations before the end of the year.

But we don't expect to spend significant money on them.

Just it's identifying the location, doing the long lead-time stuff, the permits, the planning.

This doesn't cost a lot of money.

It's only when you really start moving dirt and putting up concrete and steel and buying equipment that the big money starts to be required.

Yes.

So, anything you want to add on that?

It's good.

CapEx on Model S and X is pretty -- it's like really not -- it's minor compared to -- yes, yes, yes.

There is continued improvement, of course, to keep pace with Model 3 so that all of our products are at the same level of technology, but it's small potatoes compared to the Model 3.

And you're continuing to achieve cost reductions in S and X, so there's a bit of investment but (inaudible).

Yes, exactly.

Absolutely.

And in cases where we see cost reductions in S and X, there are cases where we want to pass along some of those cost reductions to customers.

So yes, so overall feeling really -- this is like maybe the best I've ever felt about Tesla, to be frank.

It's like really -- I mean, last week stressed the hell out of me, but I really think that this is -- this is probably the best I've ever felt about the company.

And -- oh and one thing I wanted to correct.

I think in a prior call, publicly I had said that Model Y or our compact SUV, which is called Model Y, may or may not be -- would be totally architecture.

I've gone -- I've decided -- well, upon the counsel of my executive team, thank you, thank you, thanks, guys, who reeled me back from the Cliffs of Insanity, much appreciated, the Model Y will, in fact, be using a substantial carryover from Model 3 in order to bring it to market faster.

Yes.

So that will really accelerate our ability to get Model Y to market faster for -- because it's fundamentally to -- you have people who prefer a sedan, people who prefer an SUV and, in fact, the SUV market is larger which is the biggest single product I believe in the world.

So which I thank my executive team for stopping me from being a fool.

And yes.

The Model Y, what the hell, will be -- have relatively low technical and production reschedule result.

I still think we want to do the crazy thing in the future, but we will punt that to after the model -- until after the compact SUV.

Anything else that you think (inaudible).

So yes, we'll try to leave as much time for questions as we can.

We have a lot of operational issues to get back to but like to make work on that manufacturing ramp, and I'm always incredibly grateful for anyone who has -- who is an investor in Tesla.

You put your faith in us.

We will do whatever is necessary to reward that faith.

Our next question comes from Rod Lache with Deutsche Bank.

I was going to ask you which is harder, AI or AV.

But I think at this point, we may not know the answer.

Well, as you know, I'm terrified of AI.

I've read that.

Yes, you may have heard that in a few places, and it's just something we -- anyway, I definitely don't want to derail the conversation on that front.

It's just something that I think anything that represents -- that is a risk to the public deserves at least insight from the government because one of the mandates of the government is the public well-being, and that's -- insight is different from oversight.

So it's just -- at least the government can gain insight to understand what's going on and then decide what rules are appropriate to ensure public safety.

That is what I'm advocating for.

I'm not advocating for that we stop development of AI or any of the sort of straw man hyperbole, things that have been written, I do think there are great benefits to AI.

We just need to make sure that they are indeed benefits and we don't do something really dumb.

Okay.

Well, I hope that doesn't count against my Tesla questions.

But the 2 things I was going to ask you were you mentioned in the letter this confidence in getting the 25% margin on Model 3. Could you just mention what the level of production is that you feel you need to get to in order to get there?

What run rate?

And secondly, there have been a fair number of battery announcements, solid-state battery technology, Toyota and a few others.

Can you -- what's your general assessment.

Are we getting close to some kind of breakthroughs here or just what's your thoughts?

Oh, God.

Okay, here's my opinion of the battery breakthrough of the week, battery breakthrough du jour.

When somebody has like some great claim that they've got this awesome battery, you know what?

Send us a sample.

Or if you don't trust us, send it to an independent lab where the parameters can be verified.

Otherwise, STF.

Yes.

So everything works on PowerPoint.

If you like, I'll give you a PowerPoint presentation about teleportation to the Andromeda Galaxy.

That doesn't mean it works.

So Tesla is the biggest buyer of lithium-ion batteries on earth.

You know who people come to first when they've got a lithium-ion battery?

Us.

Because we're their biggest customer.

I would love it if we could have some breakthrough.

It'd be awesome.

I think there are some interesting things on the horizon, but then the time it takes from something working in the lab to working at moderate production levels, to working at higher production levels, to optimizing the cost is several years.

So it's not like it suddenly pops out of nowhere.

J.B., do you want to add to that?

I totally agree with the sort of cautious skepticism on all these announcements.

And just more specifically on the solid-state batteries, Rod.

I mean, we do -- we've talked to a number of different groups that are researching this.

We actually have tested a number of those different prototype -- very early prototype single cells.

But it's -- we don't yet see anything that changes our strategy, and we don't see anything there that's...

However, we'd love it if it did.

Please.

Can someone please come up with a battery breakthrough.

We'd love it.

We would be the first ones to want to implement it.

Yes.

Totally.

I mean, there are some breakthroughs that I think are achievable.

I -- they're confidential, so I can't talk about them on this call.

But there's one particular avenue that I am confident could be made to work that would be not -- from a -- the most significant one -- breakthrough in a while.

But again, you've got to make it work in the lab.

It doesn't yet work in the lab.

It's promising in the lab.

You go from the lab to small production, then you go to large production, then you get to cost optimization.

These are several years, okay.

I wish it were shorter, but that's the way it goes.

So don't worry about that.

And yes, so was there...

The 25% Model 3 gross margin target.

When will we get there?

Yes.

So J.B. might want to elaborate on this, but I feel like the point at which we are at steady-state 5,000 units a week for Model 3 is about when we've reached the 25% gross margin level.

So it wouldn't be right when we get to 5,000 because initially when you get to 5,000 a week, there's still a lot of overtime.

You're still expediting parts from all around the world.

So you've a lot of expedite fees.

You've got a lot of overtime, and so it takes probably from the point at which you get to the 5,000 a week, it's probably another 3 or 4 months before you hit the 25% gross margin.

Would you agree, Deepak?

I agree.

You're -- yes, I was just going to be more cautious.

But I think...

It's something like that.

It's certainly -- you need to reach a production level and then optimize at that production level.

Yes, I mean, I think ultimately it's a variety of factors, including material cost and cell and the efficiencies we achieve at the Gigafactory and ourselves.

And I'm -- we are very confident we will achieve the 25% target, firstly on Model 3.

For sure next year.

100%.

It's a question of actually when 100%...

Again, I say 100% probability of achieving that at some point next year.

Yes.

And I feel really good about it because the bill of material that we have is so well defined and so clear in our premiums that we have on prototypes is (inaudible).

Another way of saying we're significantly less dumb this time, we think.

Yes.

So and the labor hours required are significantly lower, there will be a structured manufacturing.

It's designed for manufacturing.

Exactly.

And so all of those give me much more confidence in this target.

And exactly when we'll achieve, I think we'll give you more clarity over time.

Yes.

And I'd like to give some credit to our suppliers here.

With Roadster, and certainly with Model S and to a slightly less degree with Model X, yes, we often could not get the top suppliers or -- and we certainly couldn't get the A-Team at the top suppliers.

What's great about the Model 3 is we have the A supplier and we have the A-Team at the A supplier.

I can't tell you how important this is.

It makes a massive difference.

So just a thank you to all the suppliers that worked so hard to get us to this point.

They deserve a lot of credit for any success that we have.

Yes.

Our next question comes from Ryan Brinkman with JP Morgan.

In just thinking about your liquidity position, while you're operating with more cash than you historically have, $3 billion, I see you're also guiding to $2 billion CapEx in the back half and you previously said $1 billion of gross cash is as low as you're comfortable operating at.

So you guide to positive cash from operations the back half, presumably on Model 3 ramp in 4Q.

But if it's only a little positive, then I guess you would be close to your targeted cash level.

So the question is can you help us size up how positive do you expect the cash from operations to be in the back half?

And if that level of cash from operations plus whatever remains available to draw on your asset-backed line, if that's sufficient cushion for you relative to your $1 billion target or whether it might make sense to do another equity raise?

Yes.

Deepak, do you want to...

Yes, sure, sure, Ryan.

So we expect our operating cash flows to be significantly better in the second half compared to the first half.

At their highest level, scaling generates cash.

Yes, it certainly does.

And it's a better situation than S and X. And our cash conversion cycle, particularly for the next full quarter is going to be really great while we are shipping Model 3s in North America.

And in addition...

And one thing, perhaps you were about to get to it is -- is that the -- with Model 3, with our suppliers, we've been able to get -- negotiate much better terms, payment terms.

The payment terms are significantly longer.

So I think we're close to, is that 60?

Close to 60, correct.

Close to 60 days payment terms with our suppliers.

The -- and we are also able to make the car a lot faster.

So obviously, the nirvana is that we can make the car and get paid for the car before we have to pay our suppliers, which then the faster you grow, the faster your cash position grows.

Obviously that's like the promised land right there.

And that's how it's -- that certainly -- that's how it's -- what we've aimed for.

And I think we will achieve that, maybe not immediately, but pretty quickly.

Now that said, there may be some wisdom in having a cash cushion for unexpected events.

You just never know if there's going to be some significant force majeure events in the world.

It could be an earthquake in California, for example.

And -- but we're not, at this point, considering an equity raise.

We are thinking about debt, but we're not thinking about an equity raise.

Okay.

That's very helpful.

And then just the follow-up is about the $1 billion of desired minimum gross cash.

Does that go up when the Model 3 launches because you're a bigger company?

Or does it go down because of what you just said about the ability to generate cash and working capital while production is ramping?

In the long run, it will go up as our balance sheet grows.

And just to finish off on your question, we also have liquidity through the lines of credit, our ABL line is -- we've just grown it to $1.9 billion.

We have untapped $800 million there.

Of course, how much we can tap there depends on our borrowing base but that's a source of liquidity.

And then for our solar leased assets, we have $700 million of funding which is untapped on our tax equity funds and aggregation debt.

So we have significant amounts of liquidity to be able to fund those lines too.

Yes.

One side point that's perhaps worth noting is that some of these people look at our sort of finished goods inventory and compare that to other car companies, but they compare it in the wrong way.

Because Tesla does direct distribution, we are the dealers.

You really -- to accurately compare Tesla to other car companies, you must include the finished goods inventory, not just at the car companies, but at the dealers.

And typically, the -- that combined time of finished goods from manufacturer all the way through to dealer to an end customer is, I believe, on the order of 90 days.

So it may be 70 to 90 days.

For other OEMs.

For the other OEMs, yes.

And for us, that same metric would be approximately 30.

So this is a -- in other words, at a systemic level, we're substantially more efficient than other carmakers when you consider the system as a whole.

Our next question comes from Adam Jonas with Morgan Stanley.

Just a couple quick ones.

First on safety.

Elon, you're putting a liquid cooled supercomputer in all of your cars.

You're obviously ramping up more and more of those.

Whether the system is being activated or not, it's collecting data, learning, you're getting better every mile.

I imagine you're in a position to kind of share that data with the public or the regulators or Congress, whoever, where it matters.

I haven't seen an announcement since the 40% reduction in accidents from NHTSA back in January.

When could we be in a position to hear some more on this?

It's true that there is an enormous amount of sort of visual data being gathered.

It's actually quite a challenge to process that data and then train against that data and have the vehicle learn effectively from the data, because it's just a vast quantity of data.

I do want to emphasize that this is disaggregated from the specific vehicle.

So we're always on the side of the owner of the car and do whatever is possible, within the bounds of the law, to protect privacy.

So -- but I don't have a good answer for you.

If I could answer you right now, I spend a lot of my week working on Autopilot with the Autopilot team, right down in the trenches of the individual details of how we can improve this or that or enhance the neural net, enhance vision, advance -- improved control.

And I think the release that should go out soon is, I think people were really pleased with it.

And that is going to -- (inaudible) get actually better from there.

Yes, obviously, over time an autonomous vehicle is going to be far, far safer than a person.

Yes, it's just -- it's really hard for a person to compete.

I mean the car has 8 cameras looking 360 degrees all the time.

It's got onboard radar, it's got 12 high-precision ultrasonic sonars.

It's got inertial measurement units, high-accuracy GPS.

And over 10 tera ops of computing capability that never sleeps.

Okay.

And just a follow-up then, on space but not on Mars but more Earth, Earth (inaudible).

Oh God.

Come on, let's not get -- space.

Come on.

No, it's actually -- it's really relevant.

I was just curious is there anything that SpaceX is doing that -- or enabling that could be advantageous to Tesla's mission to accelerate sustainable transport?

There's a recent anecdote actually that Jon just shared with me.

Jon, maybe you...

Yes, there's some really great collaboration continuously between the SpaceX teams on materials and other challenges.

And we had a challenge in service over the past -- just over the past weeks.

Yes, this is -- yes, in fact, just talked about this today.

Yes.

Where we needed to determine the porosity of an object deep within our structure and that's something that SpaceX...

Aluminum casting.

Aluminum casting.

It's something that SpaceX knows how to do.

Our team reached out to the SpaceX team.

The SpaceX team helped us to solve that with some ultrasound sensors that we could quickly isolate where the issue was and take corrective action.

And that...

It saved us 8 hours of work per car.

Per car.

That was -- that could potentially experience this issue.

And that's just 1 example of a lot of examples of how the SpaceX team and the Tesla team collaborate and we get help from them continually on material issues and other issues like that.

Yes.

There's cross-fertilization of knowledge from the rocket and spacecraft history to auto back and forth is -- I think has really been quite valuable.

It's something that's been very valuable for me in thinking about how do we make mass optimized vehicles, because in space, mass optimization is extremely important.

On the space side, it's helpful because it -- what really goes into high-volume manufacturing of something that has to be extremely reliable.

So it's been good.

And of course, the companies don't compete in any way, so it's been quite helpful, actually.

Our next question comes from Colin Langan with UBS.

How do you come up with your estimate for the number of Superchargers and dealers that you need?

Because you're doubling the number of the base, the number of Superchargers, according to the release.

But the 3 is going to have a multiple higher in terms of demand.

So I mean how do you -- yes, how do you frame that and gauge that?

Yes.

First of all, actually I should clarify that the number of Superchargers will, in fact, triple between now and the end of next year.

And that's -- we're confident that, that will address the Supercharging needs of S, X and 3. So we're trying to stay ahead of it.

There are occasional places where -- that are tricky to find a location, like Malibu's really difficult.

There are a few places.

But the -- we're staying ahead of that.

I think it's going to be good.

We should see some immediate relief, even for S and X customers on some of the key Supercharger locations.

We'll also be experimenting with our first sort of, I don't recall, mega Supercharger location, like a really big Supercharger location with a bunch of amenities.

So we're going to unveil the first of those relatively soon.

I think we'll get a sense for just sort of how cool it can be to have a great place to -- if you've been driving for 3, 4 hours, stop, have great restrooms, great food, amenities, hang out for half an hour and then be on your way.

And if I could just follow up with a related question...

Well, maybe just one other point on that and how this can scale pretty efficiently.

We have Superchargers that serve 2 major separate needs.

There's long-distance route enabling between cities and then there's also within the cities.

And while there are definitely some congestion issues, which we're expanding out of very quickly in the cities, for the most part, the Superchargers that are in between cities have a lot of extra capacity.

And we've put those stations in place to serve travel between the cities, but they can absorb a lot more cars.

So even if we double fleet size, it doesn't mean that we need to double the entire Supercharge network.

We have to address the few urban sites that are currently in high use, but that can be done much more efficiently with less CapEx.

So that's kind of what you're seeing.

And what about, just as a follow-up, the charge time?

I know Porsche has said that they could charge in 15 minutes.

Do you think that's possible in the future?

And is the charge time on the 3 the same as the S?

I wasn't sure with some of the release.

It's about the same.

Yes, it's comparable to the high-end S. Like your -- the recharge rate of how many miles per hour you recharge is sort of your function of the battery pack size.

So like a 100-kilowatt hour pack -- because charge rate is a function of percentage of packs.

So think of 3 and the high-end S as being similar in charge rates.

And over time, we want to keep moving that rate up.

But they're similar.

One thing I want to correct from Friday, I don't think it really has much materiality, but I did misspeak at the generalist review on Friday.

I had said that there were 500,000 net reservations.

Well, I did also say that I wasn't sure because I don't follow this number and that this was just a guess.

And so we did check for the -- to get some precision on this.

To be more accurate, there are -- there have been 518,000 gross reservations for 3 and then -- and we have 455,000 net reservations.

But those cancellations occurred over the course of more than a year.

The net gain over the last -- net gain since Friday, net of cancellations, has been over 1,800 per day.

But I just didn't want to leave people with the wrong impression.

I think this is like inconsequential because with a small amount of effort, we could easily drive the Model 3 reservation number to something much higher, but there's no point.

It's like if you're a restaurant and you're serving hamburgers and there's like a 1.5 hour wait for the hamburger, do you really want to encourage more people to come order hamburgers?

That doesn't make sense.

So I think it's neither here nor there, but I wanted to just make sure that there was not a misunderstanding.

And maybe just one quick point on your very fast charge time comment or question.

We've actually tested cells and even full battery packs that can do something like a 15-minute recharge.

But to-date, the trade-offs to achieve that, we don't feel are the right ones for the customer overall.

You end up sacrificing on overall cost per kilowatt hour and also sacrificing on energy density in the product.

And for something that's used not every single day, not every single charge, we feel that we've kind of hit the sweet spot in terms of the value to the customer and the best product.

And that's kind of what's guided our philosophy.

But obviously there's ongoing work to reduce those trade-offs and make it better still.

But yes, we've been pretty...

Yes, but particularly if somebody buys like the 310-mile range Model 3, the amount -- let me tell you, the amount of times you'll have range anxiety is 0. You won't even think about it.

The next question comes from Toni Sacconaghi from Bernstein.

I have 1 question and 1 follow-up as well, please.

In terms of the Model 3 at the delivery event, 20 of the units were for engineering validation and the first several thousand, it appears, are going to be going to employees prior to going to the general public.

So I guess the question is what are you hoping to learn or what might you learn from these engineering validation units that have come out from your employees?

And then realistically, just what's in the (inaudible) going forward?

Well, they're not engineering validation units.

They're fully certified, fully DOT approved, EPA approved production cars.

These are not prototypes in any way.

They're not validation anything.

They are full production cars.

The reason they are initially going to employees and, in some cases investors or anyone who's been a long-time investor, is that for the first several thousand vehicles, there are problems that crop up that are rare.

On a percentage basis, they might be like 0.1% likely to occur, but then there are a whole bunch of these things that only show up 1 in 1,000 cases.

And it's good to iron out these things with a tight internal feedback loop than to do it with customers.

It also, to some degree, is a reward for those who work on the design, development, integration of the vehicle.

Yes, yes.

I mean, it's important to note, too, that all those people paid full price for the car.

Yes, full price.

There was no discount internally at all.

Exactly.

Right.

No, I mean, the root of the question is if you realistically uncover something, and even if it's 1 in 1,000, what flexibility do you think you have to be actually be able to rectify that concern in a way that won't impact your ramp?

And if this is being done arguably later in a process than a traditional OEM that's not trying to ramp necessarily as aggressively as you need to, I'm just -- again, just trying to understand realistically what can be done and what kinds of things -- like perhaps you can give an example of what you might uncover and what the rectification might be.

Yes, these are not showstoppers.

These are -- what we're talking about here are inconsistencies in the production process or in the quality control from a supplier.

So they're relatively easy to correct.

They're just -- there tends to be quite a large number of them.

But again, only rarely occurring.

It usually involves like a tolerance stack up or some combination of factors that we didn't anticipate.

But they're almost always very easy to correct.

But there are just a bunch of them and it's a lot of work.

Yes.

And these may be software issues as well, not necessarily supplier or hardware issues.

It could...

Well, and software-hardware interaction as well.

Yes.

And I think the main benefit is that we can learn about them faster and therefore, we can fix them faster.

That simply is the benefit.

Right.

Exactly.

The people driving them and the people -- are the same ones who have to fix the problem.

That's a great feedback loop.

And what we're doing is above and beyond others.

We have done a lot of testing like the other OEMs.

So this is just helping us get above and beyond by selling it to our employees and getting feedback from them.

And investors and -- yes.

Our next question is from David Tamberrino with Goldman Sachs.

Thanks for updating with the net reservations number.

I actually want to follow along those lines on your order rates.

You've given us additional color based off of 2Q trends and average weekly orders.

Can you share a little bit more maybe what the 1Q and 2Q order rate trends look like for the Model S and the X?

I don't think that those numbers would be helpful for predicting things in the future.

And like once you get into the granularity, people read things into numbers that really don't have a lot of relevance.

There's for sure seasonality in vehicle orders.

Fewer people order cars in the dead of winter than order them in spring or summer, there's just like other retailers really.

So I'm -- and then we do batch cars so that we'll make -- typically, at the beginning of the quarter, we'll make cars for Europe and Asia, and then we'll make cars for the East Coast of North America, then the West Coast of North America, and that's generally the sequence within a quarter.

And then you'll see someone write something where they think they've uncovered some "gotcha" where there were very few Teslas registered in a given country in a particular month.

It's like yes, because none of them have arrived.

This is meaningless extrapolation.

Okay.

Then I guess my follow-up question would just be on your 3Q gross margin guidance of a dip below 20%.

How far below -- to phrase correctly, how dependent upon production and hitting an S-curve or ramping up do you think that below 20% is?

Could it be a couple hundred basis points below 20% or is it just think you're going to be around that area based on what the curve that you've laid out so far is going to look like?

Yes.

This is just because of Model 3 is fundamentally negative gross margin in the very beginning, because you've got a gigantic machine producing -- that's meant for 5,000 vehicles a week and it's producing a few hundred vehicles a week.

Exactly.

That's the sole explanation.

It's a denominator problem.

Yes, yes.

Nothing fundamentally wrong with the...

Nothing -- yes, it's a temporary situation and it's a dent which corrects itself.

This is true for anything.

If like -- if you had like a soap factory, let me tell you, your first bar of soap would be like millions of dollars, okay.

But then you get into volume production and then it's like $2, okay.

So true for any manufacturing situation.

Our next question is from Brian Johnson with Barclays.

Yes, just want to ask about a couple of questions around the pace of spend in the second half (inaudible).

Please speak up.

You're a bit soft.

Yes.

Yes, the pace of OpEx through second half, you guided was flat.

With the Model 3 going out, are you basically saying you have the operating infrastructure to handle that, but then does that ramp in 2018?

And similarly for CapEx to go from 30 Model 3s up to the exit rate, you're talking $2 billion.

How do we think about, a, that run rate and given how do we tie it to your exponential ramp?

And b, what does that imply for CapEx going into 2018?

Yes, so your first question was operating expenses?

And do we have enough infrastructure in place to service the Model 3. And I think we're finally actually leveraging our own infrastructure and that's helping us.

So I'll give you an example of that.

In service, as we've talked about, we discovered that 80% of the cars that we repair don't require a lift.

And so we're deploying a mobile service strategy to take 80% of the cars and fix them where it's convenient to the customer.

Not at our location, at their location.

Make it invisible to them.

Yes.

Exactly.

Like the nice thing is like you -- like the ideal service is it's invisible.

You don't even notice it and when it's done, you love it.

And so what we're talking about here with the mobile service trucks, as Jon was saying, really most of the time we don't really need a lift, is that your car could be parked, it could be at your office parking lot or at your house.

But let's say it's like it's at work.

The -- Tesla will come there, fix your car and by the time you need to leave for work, it's done.

That's right.

And so what that does for us is it takes 80% of the volume out of our existing footprint.

And allows us to leverage that footprint to grow with Model 3. And we'd give you similar examples in stores.

So that's why we've guided to the OpEx that we've guided for the second half.

We feel like we've got leverage there and we've got plans in place to further lever that in 2018.

I think it's been really great for customers.

I mean, this is what you want.

Totally.

I mean, you don't want to bring your car into a service center.

You just want your car to be magically fixed in the parking lot and that's what we're going to do.

Yes.

So we're going to provide really great consumer happiness at increasing OpEx levels of leverage.

And second question was similarly around CapEx leverage.

So you put in $2 billion second half to get to that exit rate.

One, is that affected by the timing of Elon's production ramp?

And then two, given the ramp from 5,000 to 10,000 in 2018, what's your preliminary view of CapEx for 2018?

So I want to put a pin on 2018.

We'll talk about that when we get to that time.

But in 2017, our CapEx expense is a continuum.

There are no -- there are long lead items of different kinds like the Gigafactory.

And for the Model 3 and the equipment we are buying, our CapEx spend is at historical highs.

We're spending $100 million a week.

So a week or 2 here or there is a couple of hundred million.

So what we are spending now is the completion of all of our Model 3 equipment and tooling as our -- that gets signed off and it's taking us to 5,000 and beyond.

So I can't necessarily break it out for you but it just allows us to hit our operating plan that we have at a high level.

I want to do a quick time check here.

So some of us have some other things scheduled in 20 minutes.

So we've gotten some good, thorough answers here.

So we'll probably take a few more questions and then wrap it up.

Yes, a few more questions and like yes.

And then wrap it up, yes.

Our next question is from Colin Rusch with Oppenheimer.

Can you talk a little bit about the conversion rate of customers coming into stores and actually ordering cars?

And then similar question on what's happening with solar and energy storage in terms of how many customer impressions you've got and the conversion rate into actual sales?

The conversion rates have continued to improve.

Quarter-over-quarter and month-after-month, our conversion rates get stronger.

And we don't, obviously, publicize specifics on those but they're improving with every week and every month.

On the solar side, one of the interesting things that we're seeing is we put a solar display in -- an energy display into our stores in North America.

And we've got energy experts that are on staff, and what we're finding is it's a really natural transition in conversation from somebody is buying a car and talking about where they're going to charge the car to then where their energy comes from.

Yes, totally.

It's just -- I mean, we talked about the importance of integrating energy, production, storage and electric vehicle transport, and what we said is coming true.

It's really working well together.

And we're actually able to leverage our existing stores to generate even more sales per square foot.

I'm not sure what our -- I think our sales per square foot are so high, it's...

Our sales per square foot are so high, it actually moves the total (inaudible).

You need a telescope to see who's in second place.

Yes, exactly.

Yes, and it actually moves the overall...

It's like stupidly high.

Square footage -- sales per square foot numbers for some of the larger more (inaudible).

We've really [stepped] with more stores.

I mean, like there's no point in (inaudible) all 3, because there's like people just -- that's no point there.

But (inaudible) like eventually.

But the -- and I think like the new integrated app with -- where you can see the status of your car, your Powerwall and your solar, and see at any given time of the day, how much energy is coming from the sun, how much is coming from the Powerwall, what your house is consuming.

You can also -- it tells you when the Powerwall saved you from a utility interruption.

A lot of people don't realize that like there are many small utility interruptions in a given month, and that's why your -- you see the blinking 12 on your microwave oven or whatever, kids maybe, or your computer suddenly went dark and -- or you can even get data corruption and that kind of thing.

Or your food went bad mysteriously.

The Powerwall saves you from all of that.

And I think particularly important in cases where there's like a natural disaster, which could be floods, hurricanes, ice storms, earthquakes, fires, anything that disrupts the utility system.

But having uninterruptible power supply in the form of Powerwall gives you security in those situations.

And it's kind of like insurance.

Like you only really want it when you really want it.

And I think people are going to -- a lot of that, "I just saw the app for the first time today.

I'm using it myself" and it's like wow, this is great.

Our next question comes from Martin Viecha with Redburn.

I have just 2 very quick questions.

The first one is on the battery production for S and X. Is there any plan to move it to the Gigafactory?

For pack production?

Right.

(inaudible) production.

We do not -- in the short term, we will not be moving it.

Depending -- so sometime next year, we may move it sometime next year in order to make space for additional production volume of Model 3. That's one of the things under consideration.

But in the short term, we're keeping it here in Fremont.

But it is going to be tricky to squeeze in all of the space for increased Model 3 production, particularly if that run rate goes above 10,000 units a week, then we're going to have to move more stuff out.

Yes, and it may just be worth a reminder also that the cells for S and X are actually still 18650, and those are coming from a different production pathway in Japan.

Very similar technology.

Almost the same technology.

Yes, but it's similar to the -- yes, the internals are essentially the same for both cars.

Yes, but it's -- yes, but different supply chain, different set of geography.

And then the follow-up question is on the Model Y. I think you just mentioned an hour ago that it's going to be made probably on the same platform or very similar platform as a Model 3.

Yes, we're really going to have to...

Is it still going to have...

We're going to aim for maximum carryover.

Okay.

I mean, just one thing to clarify.

Is it still going to be the 100 meters of cables which you'd touched upon last time or actually, that's going to be the next generation of vehicles?

No, that's one of the things that we would include.

We would aim to switch out the wiring models for (inaudible) the wiring models for a redundant flex circuit that's more in the order of 100 meters or so.

But -- and then we'd -- obviously, we'd aim to do that both for the Y, if it's called the Y, and the 3 as well.

Our next question is from Alex Potter with Piper Jaffray.

Just one from me.

I was wondering the degree to which Tesla would eventually consider maybe charging more for, I guess, what you would call nontraditional product offerings.

Things like software, over-the-air updates, but also shared mobility, Supercharging, aftermarket.

I know in the past you've talked about running a lot of those businesses just to break even.

But I guess, maybe just wondering the circumstances under which you would consider trying to earn a margin on some of these businesses versus situations where you'd prefer to just give them away?

Well, I think the -- we have a major element of that which is the Autopilot.

That's a software -- that's basically uploading software to the car.

It could -- every car made since October last year is capable of autonomy, we believe.

And so it's really just a question of uploading the software for autonomy.

There will be some other things, I think in the future.

But our focus is on the Model 3 ramp and we don't want to get too distracted.

Okay, maybe one more question.

Our final question comes from Rob Cihra from Guggenheim Partners.

Just going back to Autopilot development.

Obviously, not talking personal details or anything, but you've had some personnel changes in the quarter.

And just wondering if those reflected any kind of change of strategy or scope, or if it was just kind of a personal thing?

And I guess just related, are you still hoping to be able to do the autonomous drive, LA to New York, by the end of this year?

Yes, I mean I won't comment too much on like individual personnel changes.

Look, Tesla's a 33,000-person company.

If you actually look at our executive tenure at Tesla, it's extremely good, it's above average.

I think we're at least maybe a year or 2 above average in terms of the executive tenure here.

Every now and again, something doesn't work out for one reason or another.

In the case of Autopilot, it's very centrally about vision and image recognition, neural nets, effectively narrow AI and so that's the focus from a recruiting standpoint.

And I think we've really got -- I think we've got the best team in the world by a long shot on that front and we are growing it rapidly with world-class talent.

And then the coast-coast drive, autonomous drive by the end of the year, I believe we are still on track for that.

It is certainly possible that I may have egg on my face on that front.

But if it is not at the end of the year, it'll be very close.

Great.

Thank you, everybody, for joining us today.

Thank you for, Sharie, for your help, and wish everybody a great day.

Bye-bye.

Thank you.

Ladies and gentlemen, thank you for participating in today's conference.

This does conclude the program.

You may all disconnect and have a wonderful day.