The one where Tesla loses money and decides to make even more cars.

The biggest news in Tesla's Q1 2016 earnings report wasn't that the company is still losing money despite producing more cars than ever — it was that they're rapidly accelerating their production rate. Here's CEO Elon Musk, in his own words.

Musk's opening remarks

Elon Musk, Tesla CEO: I think the most important point here that we want to make is that we're advancing the Model 3 build plan substantially and the overall volume plan, with Tesla aiming to get to the half-million unit-per-year run rate in 2018 instead of 2020. This is based off of tremendous amount [of reservations] received for the Model 3, which I think is actually a fraction of the ultimate demand when people fully understand what the car's capable of and are able to do a test drive. So, this is probably the biggest change strategically.

Also, Tesla is going to be hell-bent on becoming the best manufacturer on Earth. Thus far I think we've done a good job on design and technology of our products. The Model S and X, I think, are generally regarded by critical judges as technologically the most advanced cars in the world, and so we've done well in that respect. The key thing we need to achieve in the future is to also be the leader in manufacturing.

We take manufacturing very seriously at Tesla. It's a thing that we need to obviously solve if we're going to scale and scale rapidly and make the cars more affordable. So I really want to send the message out there to the best manufacturing people in the world: we want you to come join our company. And that is going to be the primary focus of Tesla — how do we get super good at making large, complex objects?

So that's the most salient point. It's easy to get wrapped up in a bunch of sort of short-term issues, but I think in terms of what matters to the future that's the most significant thing.

Overall, on the short-term stuff, the quarter-on-quarter stuff has improved quite significantly. Model X production increased by a factor of five from Q4 to Q1, and we continue to make huge strides in volume and quality of the vehicle. I'm personally spending an enormous amount of time on the production line. My desk is at the end of the production line; I have a sleeping bag in a conference room adjacent to the production line which I use quite frequently. The whole team is super focused on achieving rate and quality at the target cost. I feel very confident in us achieving that goal.

With the increase in ramp we do feel comfortable affirming the 80-90,000 deliveries this year. The rate of improvement with each passing day is very significant.

I'd like to thank Greg Reichow, who was our head of production, for his tremendous contribution over the last five years. Contrary to some media reports, Greg is still at Tesla, he's still with the company, and he's helping with the transition to some new leadership. We have some exciting announcements coming in possibly the next few weeks about additions to the Tesla management team on the production side. I'm feeling really, really excited about where things are heading in that direction.

With that, let's go to questions.

On Model 3 obstacles

James Albertine, Stifel Nicolaus: Great, thank you so much for the question, and Elon, thank you for that introduction. There's no doubt you have an incredible undertaking in front of you. There's no doubt you have an incredible undertaking in front of you. Can you help us understand some of the key obstacles and how you should consider those obstacles between now and your anticipated launch of the Model 3 in late 2017. Whether it's a P&L adjustment that we need to make along the way — can you help us sort of choreograph how that's going to take place?

Elon Musk: Sure. With the Model 3, as I mentioned on the last call, we're really trying to take a lot of lessons learned from Model X. We put a lot of bells and whistles on Model X and a lot of advanced technology that weren't necessary for version one of the vehicle. With Model 3 we're being incredibly rigorous about ensuring that we don't have anything that isn't really necessary to make a great, compelling version one of the car. We also have a much tighter feedback loop between design engineering, manufacturing engineering, and production. So no element of Model 3 can be approved unless manufacturing has said that his is easy to manufacture and that the risks associated with manufacturing it [are] low.

There are many ways to skin a cat, and it's remarkable how you can achieve the same objective with a hugely varying degree of difficulty. You can take the analogy of it, say, you wanted to kill a fly. You can kill a fly with a thermonuclear weapon, with a MOAB, with a cruise missile, with a machine gun, or with a fly swatter. The end result is the same, but the difficulty is considerably more significant from one to the other and the collateral damage is considerably more significant.

Having production be really fundamental to the design of Model 3 is very important, and to make sure that we're not adding extraneous features to the 3 that aren't necessary to achieve production volume is also extremely important.

At the risk of this being misinterpreted, and probably there will be some number of articles that do, I think it's worth explaining sort of how manufacturing a complex object with several thousand unique components actually works. In order to achieve volume production of a new car with several thousand unique items, you actually have to set a target date — internally and with suppliers — that is quite aggressive. That is a date that has to be taken seriously. The date, and I'm sure this will leak — it's hard to keep a secret, really — the date we are setting with suppliers to get to volume production capability with the Model 3 is July 1st next year.

Now will we actually be able to achieve volume production on July 1st next year? Of course not. The reason is that even if 99% of the internally-produced and supplier items are available on July 1st, we still cannot produce the car, because you cannot produce a car that is missing 1% of its components. Nonetheless, we need to both internally and with suppliers take that date seriously and there need to be some penalties for anyone internally or externally who does not meet that timeframe.

This has to be the case, because there's no way that you have several thousand components all of whom make it on a particular date. The reality is that the volume production will be some number of dates later as we solve the supply chain and internal production issues. But is a bit of a confusing thing and it does create some churn, because people, like "Well, what's the real date?" You have to take the July 1st date seriously in order for some date a number of months later to actually be the real date. That's actually how it has to work. In order for us to be competent of achieving volume production of Model 3 by late 2017 we actually have to set a date of mid-2017 and really hold people's feet to the fire, internally and externally, to achieve an actual volume production date of late 2016.

As a rough guess, I would say we aim to produce 100-200,000 Model 3s in the second half of next year. That's my expectation right now. What I would say to anyone that is thinking about ordering a Model 3: now is a good time to actually place your reservation, place your order, because you don't have to worry about placing your order and receiving it five years from. If you place your order now, there's a high probability you will receive your car in 2018. I'd really recommend that anybody who wants to receive their car in 2018 place their order very soon.

James Albertine, Stifel Nicolaus: Elon, thank you, if I may, as a follow up: can you give us some reference, you said fourth quarter of 2017, so six months at its most generous calculation, how did that compare with the volume production agreed date for the Model X, just as an example, and how does this flow with your cash needs as you've articulated, it seems you've walked back a little bit from the prior quarter's discussion around cashflow positive and no need for capital markets raise. If you could just articulate how the two fit together, that'd be helpful, thanks.

Elon Musk: It's always tempting for people to read my an analogy instead of those principles, and that would be the mistake of assuming that anything to do with the X production has bearing on Model 3. They are very different programs with completely different approaches. I would not try to extrapolate from that, any more than it would have made sense to extrapolate from the Roadster, when we were making 600 cars a year to 20,000 cars a year with the Model S.

With the Roadster's case we went from making 600 cars a year in 2010, where Lotus made the body and chassis, we made the powertrain and we did final assembly. It was a far simpler car than Model S. We told people we were going to get to get to a run rate of 20,000 cars a year with Model S, despite it being a vastly more complicated car and a car where we make the whole car, not just the powertrain. If you were to extrapolate from the Roadster experience, you would be completely wrong about the Model S outcome — and many people were. That's why I would say X is not relevant.

As far as the increased capital rates, obviously, if you double your planned volume you can't expect the capital to stay the same. I think our capital efficiency will improve on a per-car basis, but obviously it can't stay the same.

On Model 3 confidence

Colin Langan, UBS: Oh, great, thanks for taking my questions. Just to follow up, you've had issues with the X, there's been management changes, what gives you the confidence of 500,000 — a pretty amazing jump in the next year. What gives that conviction that that's going to be possible by 2017?

Elon Musk: You mean by 2018?

Colin Langan, UBS: Yes, 2018.

Elon Musk: First of all, I think we've got an excellent team at Tesla in production. We're adding world-class aces in production with each passing week. It's fair to say it's probably the most compelling product program in the world with the Model 3. I'm not sure what would be more compelling — there's a good argument that Model 3 is the most compelling program on Earth from a manufacturing standpoint. Our ability to recruit top manufacturing talent to the most compelling product on Earth is very strong. We find the response to be extremely good when we call people up. Based on the rate we're adding, we'll pass manufacturing expertise, and some of the things I know we're going to announce in the future, I feel highly confident that Model 3 is going to be well executed as a program.

Jason Wheeler, Tesla CFO: The design of the vehicle lends itself to high volume production very efficiently.

Elon Musk: Designed for manufacturing!

Jason Wheeler: And that's something that we're doing even today, those designs are firming up. This is something happening far, far ahead of time. The second point would be the quailty and motivation of the suppliers involved in the program.

Elon Musk: Yeah, massively increased. Every supplier wants to be in this program.

On battery costs

Colin Langan, UBS: If I could ask a follow-up. Obviously, cost is going to be an important factor when 3 launches. I think you've indicated that your body costs per pack are now under $190 per kilowatt hour. How do you think that compares to the industry? Where do you think it will be by the time the Model 3 is launching?

Elon Musk: We try not to comment on individual component costs; that stuff's fairly proprietary, it's kind of like giving away our playbook. I think it's pretty obvious that we will exceed anyone else in scale of economies with the Gigafactory. We're very confident in Panasonic's ability to execute on that front. I just don't know anyone who in terms of intrisic costs is going to be close to what the Gigafactory can produce on a cost-per-kilowatt-hour basis.

Colin Langan, UBS: When we think the 190, how much like a figure of decline below the Gigafactory, like a 30% bump line?

Elon Musk: Yeah, next question.

On increased production rate spending

Colin Rusch, Oppenheimer: Thanks so much. As you look at this accelerated plan for production, what can we expect on op-ex spending to support all of those cars coming out a lot faster than you previously expected.

Jason Wheeler: We updated our guidance on op-ex for the year a little bit in the letter. We talked about 20% last year and moving that range to 20-25% for 2016, so there's obviously going to need to be more op-ex. However, at the same time, you see how we improved quarter-over-quarter in terms of op-ex; we were down 12 million from Q4, down 3%. So there's a renewed focus in the halls here at Tesla on making sure that we're managing costs extremely effectively, and all of our employees get that and are contributing to that.

Elon Musk: Our sort of operating leverage means there are fixed costs relative to our variable costs are going to improve dramatically as you get volume up.

Jason Wheeler: We talked a little bit about this on the call last quarter. The potential for operating leverage is massive with production scaling.

On a CEO whose desk is on the factory floor

Colin Rusch, Oppenheimer: Elon, what do you need to see to move your desk out of the factory? It's kind of a dramatic thing to talk about, having your sleeping bag there, so obviously there are some things you were concerned about. What are you going to want to see to go back to a different location?

Elon Musk: My desk has frequently been in the factory, so this is not some new thing. On the Model S my desk was also in the middle of the factory, at the start of the body line, for a year. I move my desk around to wherever the most important place is for the company, and then I'll maintain a desk there over time to come in and check on things. I suspect, probably by the end of this quarter, most of my time will not be spent on the factory floor.

On capital costs

Pat Archambault, Goldman Sachs: Thank you, good afternoon. Getting back to the capital requirements for the expanded Model 3 production. Appreciate the guidance that you've provided for this year, from a cap-ex perspective that's helpful. Maybe this is a question for Jason: can you share with us, maybe, what a total capital cost estimate might look like for the Model 3 program now that you've got a handle on what your volume's going to be or what you want to produce too?

Jason Wheeler: A couple things there. So, one, we've provided some breadcrumbs. We updated our cap-ex guidance, we guided at $1.5 billion last quarter, and we think it will be 50% higher than that for 2016, into 2017 — we're not going to talk about that now.

The other thing to pay attention to is our cap-ex for this quarter was $216 million, which was a 47% decrease over Q4. A little bit of that is what we talked about last quarter, where a lot of the big investments for Model X had already been made. But also, we're just really focusing on capital efficiency and making sure that we're investing in the highest and best uses of cash. I think those principles are what's going to guide the Model 3 program.

Pat Archambault, Goldman Sachs: That's a good starting point to work with for us, we appreciate the update for this year. Maybe the way to take the question is just to kind of understand when the peak spending period are going to be. If you're launching through the middle of next year, is it kind of a good idea to maybe extend the amount of capital you see spending in the balance of three quarters through the second half of next year? Clearly, with the launch, that tapers off, is that a right way to think about it? And then, second to that, I'd probably ask the same question, just on the R&D front: when do those costs spike in the timeframe of that program?

Jason Wheeler: I think you're thinking about it the right way, the way you've laid it out. You can kind of use Model X and Model S and the ramp of capital for those programs as a way to think about Model 3. On the R&D piece of it, that is a big driver behind our updating of our range to 20 to 25% op-ex in 2016, so we'll start to see a little bit of that in the second half of this year, and then certainly some more into the first half of 2017.

On in-sourcing vs. out-sourcing

Pat Archambault, Goldman Sachs: Got it. If I could squeeze in one last one, just on the sourcing: is this changing your strategy of working with suppliers? You've done a lot in house for all of your products so far, but obviously this is a very different kind of volume number you're talking about. So are you thinking about changing the level of vertical integration and how does that work into the ongoing capital requirements for this program?

Elon Musk: No, I think we're actually going to increase the amount of vertical integration we have. I think it's very important for us to have the ability to produce almost any part on the car at will, because it alleviates risk with suppliers where, going back, if 2% of suppliers aren't ready, we can't make the car. Having the ability internally to adapt and make those 2% of parts internally really massively reduces the risks associated with the production ramp. That is a very important thing.

Once we get to steady state and we talk to a supplier and they can do a very efficient job of making that part, we have to problem transition it from in-sourced to out-sourced. Our goal is not to in-source for the sake of in-sourcing, but rather to in-source if we think that it has a meaningful improvement on schedule or cost or quality.

One of the challenges we face is that for a lot of supply chains they're impedance-matched to the timeframe of the big OEMS, and Tesla just moves a lot faster than the big OEMS. And so, if they're impedance-matched to a typical sort of 6-year development cycle and we're on a 2- or 3-year development cycle, it just doesn't connect properly. Some suppliers can handle that, and some can't.

Pat Archambault, Goldman Sachs: That certainly seems to limit some of the people you could work with. Thanks for the clarification.

On timelines and personnel

Brian Johnson, Barkley: Good evening, afternoon. I just want to talk a little bit about some of the milestones that you see in terms of this accelerated development in launch and scale up of the Model 3. First, it looked like in the proxy that the alpha prototype was completed as of when it was filed a few weeks ago. So, a few questions. One: when do you expect the beta prototype to be achieved? When do you think you'll have firm specs for both your internal parts operations and for your external suppliers? And in terms of the capital, when would you see raising capital, if at all, to meet this? And, finally, given the volume of trade-off decisions you're talking about making between manufacturing, design, engineering, do you see any role for a COO type similar to what you have a SpaceX to accomplish this timeline?

Elon Musk: That's like 17 questions in one.

Brian Johnson, Barkley: You can send us the project plan.

Elon Musk: From an engineering standpoint we're already almost complete with the design of Model 3. In fact, the prototype that was driving at the Model 3 event at the end of March was actually using the production drivetrain. So we feel pretty good about engineering completion of the last items, probably within the next 6-8 weeks or about, completely final release for tooling at the end of June. That leaves roughly nine months for the tools to be manufactured, which I think is an achievable timeframe — for some suppliers that's an achievable timeframe, you can have a human baby in nine months, you can pretty much make a tool in nine months. That's our expectation.

We want to have parts of production tooling starting in April next year so we've got three months of validation for a normal start of volume production in July. Again, it's a nominal start, and it's a date that we internally need to take seriously and that suppliers need to take seriously, but it is one where inevitably there will be some small number of items that cause slippage such that the actual date of reaching volume production is some months after that. This is simply the nature of things, it is unavoidable. If you could tell me what those parts would be, we would be able to take action now; it's easier to see what these things are in hindsight, not in advance. And sometimes they're the things you don't expect to be a problem.

And I don't want to comment too specifically on senior executive hires.

Brian Johnson, Barkley: Does this imply similar accelerated schedule for the Gigafactory, which always seemed tied to the 2020 half-million-unit goal.

Elon Musk: It does.

Jason Wheeler: As we've discussed previously, this is a small part of why we accelerated some of our plans there. We're still on track to have first cell production starting at the end of this year so that we'll be able to ramp up to match the Model 3 schedule as well.

Elon Musk: I want to emphasize some comments that I made earlier during this call. Tesla is really hell-bent on being the world's best at manufacturing. Like, this is a big deal. And I think it's the right thing to do, because what we're trying do is get as many electric cars on the road as possible, and what's the limiting factor? Well, it's production. How can we scale, and scale efficiently? We need to figure out how to be the world's best manufacturer and that's what we're going to be hell-bent on doing.

Brian Johnson, Barkley: Okay, thanks.

On the investor advantage of Autopilot

Adam Jonas, Morgan Stanley: Hey, Elon. So on our math your combined fleet of Model S and X are driving more than three million miles a day so in just one day you cars do about 2x the distance that Google has done in the entire history of their self-driving car project. Now while your cars aren't exactly sensor encrusted Christmas trees with tens of thousands of dollars of equipment like a retrofitted Google Car, it's still a lot of miles. I was just wondering if you could explain to the investment community what advantage this gives Tesla in the race vs. animal transport and accident free driving in some commercial or financial terms. Thanks. Or even engineering terms.

Elon Musk: Well I think you've pretty much asked the question and answered it. I mean data is everything, really, when you're trying to solve the autonomous transport problem and having millions of miles per day of data accumulating and then as the fleet grows that grows proportionate to the fleet is incredibly helpful.

I mean, I think really, particularly as you go to, say, you know, kind of long term, kind of fully autonomous driving, which I think is going to — that's going to require quite a lot of regulatory oversight and I think in order for regulators to feel comfortable approving that, they're going to want to see large amounts of data. Maybe billions of miles, showing that the car is unequivocally safe in autonomous mode compared to manual mode in a wide range of circumstances in countries all over the world with different rules of the road and ways of behavior. And, yeah, it'll have to be something statistically significant like billions of miles.

On pushing Autopilot through Congress

Adam Jonas, Morgan Stanley: Okay, well that actually, Elon, leads to my follow up which is, you know, once high volumes of statistical data for your autonomous miles are collected and analyzed, I can't help get out of my mind — I have this image of you and some CEOs of other auto companies and CEOs of other software and tech hardware firms testifying in Congress about the urgent need to replace these dangerous, purely human driven cars on the road with available, affordable, and proven even L2, L3 technology or semi-autonomous that's ready for introduction to dramatically improve the epidemic of traffic fatalities as like a national public health and safety priority.

Am I crazy, Elon, about kind of that type of… that role for people in your position to play? You know, armed with the data empirically? If I'm not crazy, then how soon do you think it will take for tech firms like you to have a sufficient quantity and quality of data to make such a case? Thanks.

Elon Musk: Tesla will argue for autonomous driving but we're not going to argue against manual driving. We believe people should have the freedom to choose to do what they want to do. And yes, sometimes those things are dangerous, but you know, freedom is important. And, if people want to drive, even if it's dangerous, they should be allowed to drive in my view. But then the autonomous safety system should be in there so that even if you're in manual mode the car will still aid you in avoiding an accident.

Adam Jonas, Morgan Stanley: Okay, thanks.

On Gigafactory capacity

Joe Spak, RBC Capital Markets: Thanks, good afternoon everyone. Also wanted to focus on the adjusting the Gigafactory plans. I believe originally you indicated about 15 gigawatt-hours per year earmarked for energy, and with Model 3 demand clearly robust and likely more robust than you originally planned, I'm wondering if that moves some of the Tesla Energy ambitions to the back burner? Does it accelerate, you know, the need for a second Gigafactory or maybe you found a way to squeeze more out of the existing one?

JB Straubel, Tesla CTO: I think the simplest answer is that we have a lot more capacity at that site than the initial 35 and 15 gigawatt hours that we discussed. That's part of why we've so aggressively made sure that we have extra land and extra space around the site so that we can continue to expand. And we won't need to rob from Tesla Energy plans in order to meet the Tesla 3 schedule. We definitely have a way to solve both.

Joe Spak, RBC Capital Markets: And are you willing to provide an update to those initial targets?

Elon Musk: Not yet. Maybe in one or two earnings calls from now I think we'll be able to provide… shed more light on that. But, yes, as JB was saying, we're going to make sure Tesla Energy isn't constrained by vehicle needs. And I think we're, yeah… the growth rate of Tesla Energy is on a tangent basis is going to be far greater than the growth rate of cars.

Joe Spak, RBC Capital Markets: Thank you.

On meeting deadlines

Ryan Brinkman, JP Morgan: Great, thanks for taking my question. We can all now see that, with the Model 3 pre-orders, that you are entirely correct that there's tons of demand for the car just like you've been saying all along. So, I think about a month ago when you started tweeting those pre-orders, the investor and parts-supplier confidence in your ability to ramp to half a million units rightfully skyrocketed.

With that said, from a supply perspective you have sometimes had difficulty in achieving delivery targets because of issues and smoothly increasing capacity in assembly and you've shown a strong preference in emphasizing quality over quantity. So, is there anything that's changed on the supply side of the equation that should also be confidence instilling? Maybe, I don't know, lessons learned from the launch of X or some other factor that should give confidence in your ability to be at a 200-400 thousand annual run-rate of Model 3 production approximately 14 months from now?

Elon Musk: Yeah. Again, I want to emphasize that the July 1 date is not a date that will be met. It's an impossible date. However, it is a date we need to hold ourselves to internally and need to hold suppliers to. But it's an impossible date because there are 6-7 thousand unique components in the Model 3 and that would assume that all of them arrive on time. And just like, you know, with college term papers, there are always late term papers. You still have to have deadline, and it has to be real, and one with consequences if the deadline is not met but it absolutely will not… the probability of it occurring is incredibly low of actually achieving July 1 but nonetheless it is a date we have to take seriously. I explain that with some risk of it being misinterpreted but hopefully we'll, you know, appreciate how it needs to work and kind of has to work that way. There's no other way to do it.

And the things that help us get there are designing Model 3 for manufacturing, engineering, manufacturing engineering, and production all in a very close loop. And making sure we design the car to be easy to make. That we iterate with suppliers and ask them if we've given them a design that's easy to make or hard to make, or how do we reduce risk and improve it and make it easier to build.

This is really fundamentally different from S and X. The S was the first car we really designed ourselves. And it was all about just trying to make a car work in the first place. The X was basically built off of the S platform but then even more complicated. So, unfortunately, even harder to make.

The Model 3 will be the first car Tesla is creating that's designed to be easy to make. This is really a fundamental difference. And then I mentioned also increasing the scope of our in-house abilities so that if there's supplier that isn't able to deliver on time, we can scramble fast and produce that component in-house.

Ryan Brinkman, JP Morgan: Okay, that's helpful. Thank you.

On cash sources and supplier costs

John Murphy, Bank of America: Good afternoon. Just a first question on the capital needs. I mean, it looks like there's a little over 400 million left on the ABL, and given the pre-orders, or the reservations, for the Model 3 it looks like you'll have another 400 million flowing in in the second quarter so, just curious, as you look at that kind of cash potential or liquidity and potential inflow do you really think you need to do a capital raise this year or could you get by with those sources of cash?

Elon Musk: Well, I don't think we want to rely too much on customer reservation money as opposed to capital. Maybe there's a buffer or something but not as a primary source of capital. So, yeah, I mean, I think it's going to make sense for us to raise some amount of money. So a combination of equity and debt, and make sure the company has a good buffer of cash on hand. I think it's important for de-risking the company.

Jason Wheeler: The only thing I'd add to that is we did draw 430 million on the ABL this quarter. A lot of that is that was we had a large amount of cash in transit at the end of the quarter and deliveries were back-end loaded. And as those cars were delivered in late April we were able to pay a significant portion of that back.

Elon Musk: Yeah I think most people are familiar with the Asset Back Line but it's important to say like, why is that different from general debt? Unlike other automotive companies, Tesla doesn't ship to dealers, we ship to customers. We build the cars to order, the car is complete, and it's going to a known customer. So really, the only risk associated with that is like if the ship sinks or something. Or, you know, the truck that's carrying the car crashes. The ABL —the Asset Back Line is basically finished good in transit to known customers. It's not like general corporate debt. It's, I think, more appropriately thought of as a slight increase in cost of goods sold.

John Murphy: Okay, that's helpful. And then if I could just ask one follow up from another question? As you look at the ramp with suppliers, is there any recourse to suppliers that don't meet that sort of start of production next year or any point of the production schedule, or is it really you just cancel the business and move on to a different supplier?

Elon Musk: Yeah, so, we'll be asking for firm commitments from suppliers to meet that timeframe. The… and I'm meeting personally with the team from that supplier that's going to who is going to execute on the task. So that I have not just the commitment of the CEO or general manager but the actual team that will execute on the product. And we want to confirm that we feel confident in the actual team and basically what we're asking for the A team from the A supplier and a commitment from that A team that they intend to work harder than they ever have on any other program. If they're willing to do that then we'll work together. Otherwise, not.

John Murphy: And recourse if they miss targets?

Elon Musk: Yes, so, along the way we'll be assessing progress and our confidence level that suppliers will meet the July 1 target. If it looks like they will not, we'll have a conversation with them. If our confidence level drops below a certain level then they will not be a supplier to Tesla.

John Murphy: Great, thank you.

On franchise dealership laws

Rod Lash, Deutsche Bank: Hi everybody. A couple questions. One, distribution and franchise laws in the U.S. have always seemed like an issue that would need to be dealt with at some point. Does this trajectory force the issue or is this something that you can accommodate even with the distribution constraints?

Elon Musk: It's worth emphasizing that the whole dealership thing only applies in the U.S. We don't encounter the issue anywhere else in the world. And what's happening is that dealers are using vestigial legislation that was originally put in for a just purpose—which is to protect them from predatory practices from the franchisor—and have been using it for an unjust purpose which is to prevent direct distribution. We believe that in the long term justice will prevail.

Rod Lash: Okay, but is there a few that you can actually achieve this even under the constraints that exist today or is it something you need to address to achieve this plan?

Elon Musk: We believe that it's not a constraint on our ability to achieve the plan.

On factory expansion

Rod Lash: Okay. Second question is, I'm assuming that concurrently with this plan there's a kind of longer term plan for growth and that there's going to be a Fremont #2 and I think you alluded to further expansion of Gigafactory can you just give us the sense of what you're aspiring to in terms of the trajectory by the end of the decade as you've done before and Jason, I know you didn't want to get into details on project spending but it would be helpful just to pass along some thoughts on what needs to go into the company in terms of investment in order to get that sort of thing out. Is it reasonable to assume that new level of spending that we're seeing now that we should assume is being a sustained level going forward.

Jason Wheeler: Sure, on that, so, yeah, I don't want to go into the details on what we think the total capital cost will be for the Model 3 program but certainly as we continue to ramp there are going to be more capital costs for the company, that's just a fact. Ideally I'd like to fund as much of that as possible with cashflow from operations. So that is really the focus that we have in the short term.

Elon Musk: It's fair to say that probably there's a 2020 target for volume is closer to, you know, maybe a million vehicles in 2020 or something like that.

Rod Lash: Okay, great, thank you.

On first-time Tesla buyers

Charlie Anderson, Dougherty: Yeah thanks for taking my question. I have just a two parter for Model 3 reservation holders. I imagine this was their first interaction with Tesla and maybe the first time they went to a store. And I wonder, as you've looked at that base, if there's any potential to upsell to an S or X in the interim as they wait for their cars? Any programs planned to address that?

And secondarily, I was curious if you had any color on sort of the geographic split of the reservation holders? Thanks.

John McNeill, Tesla President of Global Sales & Service: In terms of the, your first question on whether or not the reservation holders, if this was their first interaction with Tesla, just about 93% of the reservation holders, this is their first interaction with Tesla. So, it is a super, super majority of new client base or customer base for Tesla and it's exciting. It was exciting when we walked the lines for… there were people waiting in line at the stores and they were excited to become part of the Tesla community and family.

And the demographics of the owners, we're not going to say much about that, but they are a bit different as you can imagine from the Model S and Model X owners today. And it presents an exciting new market for Tesla as well.

And it should be noted that these folks are not interested in just only Tesla Motors but also Tesla Energy. Because the price point of the Tesla Power Wall is an accessible price point for many of these folks. So, they're expressing interest in both.

In terms of their S and X as a bridge to Model 3 we are talking through and thinking through that because, as Elon mentioned earlier, the quickest path to receiving a Model 3 is being a Tesla owner. We've agreed that Tesla owners are receiving priority in terms of production and so you can run the math I just mentioned. If 93% are new to Tesla then 7% are Tesla owners and the fastest way to get a production vehicle, even in 2017, is through Tesla ownership. And so, we're finding there's a good conversion rate of folks coming in to Tesla to drive an S or X are Model 3 reservation holders. And are motivated to be Tesla owners now so they can receive their Model 3 earlier.

Elon Musk: Yeah, actually, you know, an important point worth mentioning is like we were fairly worried about what would happen with the Model 3 announcement. Would it cause like some big drop in Model S sales? It seems to have had the opposite effect. It seems as though S demand has increased. I mean, it has.

John McNeill: It has increased.

Elon Musk: It has increased. Yeah.

John McNeill: Yeah, I think you saw the S demand in the quarter is up 45% year-over-year. And that demand continues.

Charlie Anderson: Thanks so much.

On new factories

Emmanuel Rosner, CLSA: Hi, good afternoon. I wanted to ask you guys about any early thoughts on need for, you know, manufacturing expansion. Obviously, if you're already doing 500 thousand units by 2018, I think that's the original capacity of the Fremont plant so do you need to start thinking about, you know, an additional plant and in that context any thoughts on global expansion you were mentioning obviously a very strong increase in Model S orders in Asia, for example? Anything you could share with us at this point.

Elon Musk: Our plans for, you know, international expansion and establishment of new plants is sort of speculative. We haven't made any firm decision. But I mean some of things are just sort of common sense that manufacturing cars in California and then shipping them all around the world isn't a very efficient thing to do particularly as you go towards more affordable vehicles. So, at some point it's going to make sense to have a plant in Europe and a plant in China and probably a plant in other parts of the world. So that's a kind of natural thing to expect us to do. LIke, it wouldn't make sense to ship cars from California to Europe or California to Asia in this volume. It's just not an efficient way to go. And, particularly, as we saturate on Freemont volume in terms of satisfying demand in North America I think we'll just to satisfy demand in North America for our future product lineup we're going to need more than one plant in North America. Just to satisfy North America demand.

Emmanuel Rosner: Right, so when we think about these extra capital needs that you're sort of alluding to in addition to just the obviously just the cost of the Model 3 in the development are you also contemplating as part of that to raise the money an extra factory to the extent that just beyond 2018 you will need some extra capacity.

Elon Musk: I don't think we'll be raising money for new factories before we're at volume production of the Model 3, then as Jason was saying earlier, we'll try to find as much of it as possible from operating cash flow.

Emmanuel Rosner: Got it, thank you.

On the naysayers

Ben Kallo, Robert Baird: Hey, thanks a lot. I have eighteen questions. The first one I have—Model X production, where are we at right there because you've had all these consumer reports issues I think is a little back dated but can you talk to us about the state production of that? Number two, on the Gigafactory and the battery size of the Model 3, I think everyone is dividing by 80 kilowatt hours or 75 kilowatt hours for the number of cars and how do we think about the Model 3 battery size and what the Gigafactory can support. And then the third question is why is Bob Lutz and Jim Chanos they keep on saying such negative things about you guys, what do you have to do to get the dissenters to actually believe in Tesla a little bit. Thank you.

Elon Musk: Yeah, so, I mean I feel confident that we're going to hit the 2000 vehicle a week target by the end of this quarter of which on the order of 40% are X. I'm telling you that's our internal plan and what we expect to meet. There's no question the X is a very difficult car to manufacture. Unquestionably actually the most difficult car in the world to manufacture in the world. Bob Lutz would agree with that, I think he said something to the effect that he thought it wasn't manufacturable or something like that. It's certainly manufacturable it's just a hard thing to [undecipherable].

So, but, we have some internal milestones that we've achieved so far that I'm really excited about. Friday at 3am we achieved our first flawless production of the Model X where it through the whole production process and zero issues. It was a great milestone. If you could have seen us celebrating with the team at 3am it was great. And then now we're starting to get several in a row that are flawless and so it's really gaining momentum very quickly and we feel really good about the trajectory of S and X.

As for convincing all the naysayers, I think that will basically be never. There are always going to be naysayers. And what I find ironic about a lot of the naysayers is that the very same people will transition from saying it was impossible to saying it was obvious. Wait second, was it obvious or was it impossible, it can't be both. Right?

On Model 3 battery capacity

Ben Kallo, Robert Baird: Got it. And the Model 3 battery? We're all analyst here, we stare down a straw dividing by you know 75 kilowatt hours. Is that the right things to do with the Model 3, or should we have a lower number like 40 kilowatt hours or 45? And then you got the guy with the bull making that car saying you know it's going to be ahead of you guys and sell for cheaper than you? So how do I think about GM being able to make a car cheaper than you versus making margin on a Tesla with a lower battery cost. Does that make sense?

Elon Musk: Yeah, I mean, we're not getting into any real specifics on battery pack size but I think it's fair to say like that the average battery pack size for the Model 3 will be less than 75 kilowatt hours. That's..

Ben Kallo, Robert Baird: Sorry, what was that?

Elon Musk: The average energy content of a 3 pack is certainly going to be less than 75. It doesn't really need to be anywhere near 75 to achieve the range of 215 miles. So, we don't want to get into the nitty gritty is probably unwise.

John McNeill: You probably don't have to fixate on the 35 gigawatt hours. We're planning the Gigafactory to meet the production needs of the energy that we know the cars will need. There's not a problem with scaling that as we need to, so obviously internally we know the math and we know what we need to do and we're on track to do it.

On ramping up production

Ben Kallo, Robert Baird: I guess my 18th question is so I'm not a car guy so I have you guys having 40 thousand units of the Model 3 at 2017 and from your commentary and it seems like I need to raise my numbers but how do I think about that ramp up from zero to five hundred thousand over push it from 2018 on. So to go from zero to five hundred thousand over two years or one year or how do we think about that.

Elon Musk: Well I mean obviously if we're saying that we'll have total vehicle production of on the order of five hundred thousand cars in 2018 it can't take more than two years to get there. So, just a sort of another sort of information or indication of production ramp — production ramp is like an S-curve. It's extremely difficult to predict with precision the early part of the S curve because in the early part of the S curve it sort of starts off very slow and then increases exponentially moves to linear and then moves to a logarithmic. So it really is very difficult to predict exactly what the shape of that S curve is and that's when things get tricky because you end up putting quarterly results kind of bracketing somewhere on that S curve. And, depending on where you are on that S curve, it can actually look like a big difference but it could be a shift of a few weeks because of the exponential nature of the beginning of the S curve.

Ben Kallo, Robert Baird: Then my 19th question: Can you make 15% gross margin on it or 20% gross margin or how do you think about margin? People think you can't make it profitably. And thank you, that's it.

Elon Musk: Yeah, we're highly confident that it can be made profitably. And design for manufacturing and economies of scale are the keys to achieving that outcome. Yeah, I think, like GM is not aiming for anything near the volumes that we are. And, so their, despite being a big company, their economies of scale are going to be driven by whatever elements are unique in their EV. And we know for a fact they won't be at the economies of scale that we'll be at for Model 3.

Ben Kallo, Robert Baird: Great, thank you.

On production mix

Dana Hall, Bloomberg News: Yeah, hi, what is the mix in 2018 of the five hundred thousand cars? I mean, it's combined S, X, and 3… Should we think of it as three hundred thousand 3 or what's the mix of those three vehicles.

Elon Musk: You know, I don't think we've got an amazing crystal ball to, you know, figure out exactly what it's going to be. I mean I feel confident about the top line number but the mix internally is difficult to figure out. I mean, Maybe it's something like a 100 to a 150,000 S and X and three to four hundred thousand of 3, but this is… I don't know, it's really hard to say.

Dana Hall, Bloomberg News: And then as you try to attract top manufacturing talent as you begun to ramp have you given any thought to trying to hire a COO? I mean, I'm just thinking about your personal life between Tesla and SpaceX and sleeping in a sleeping bag and working 90 hours a week between two companies. SpaceX has a great COO and has had one for years, but Tesla never has.

Elon Musk: The sort of scope of Tesla's business activity is bolder than SpaceX. SpaceX is sort of a pure technology company and does not have the sort of sales, service, and fleet management and customer financing, all that sort of stuff, that Tesla has. Obviously, John McNeill has that role at Tesla, and then my focus is primarily on technology, design, and manufacturing. We certainly can expect that there will be announcements in the very near future about some great executives joining the ranks.

Dana Hall, Bloomberg News: Okay, great, thank you.

On a million in 2020

Phil LeBeau, CNBC: Hi, Elon, I have a question. It was about 10 minutes ago you made a reference to 1 million vehicles in 2020. Is that a production target, a production goal, or a hypothetical? Just looking for some clarification there.

Elon Musk: That's my best guess. If we're at a half million in 2018 and then sort of roughly 50%-ish growth from there then it's probably around a million in 2020.

Phil LeBeau, CNBC: And do you have an estimate as to how many production plants you will need to make that happen?

Elon Musk: I think it actually feasible, maybe not advisable, but feasible, to do it with just Fremont and the Gigafactory. We actually believe that Fremont and the Gigafactory could scale to a million vehicles. Whether that's actually wise is a separate question. As I said earlier, it's going to make sense to do localized production, at least on a continent basis, otherwise your logistics cost end up being quiet extreme. Your logistics cost start becoming a bigger and bigger percentage of total vehicle costs. That's really why manufacturers build their cars for their local market; they build cars for that market in that market. Logistics cost associated with shipping a one-and-a-half to two-ton vehicle are massively greater than, say, shipping a little consumer electronics device.

Phil LeBeau, CNBC: Great, thank you.

On recruiting manufacturing talent

Alex Sage, Reuters: Elon, you say that you're calling out to the best minds of manufacturing to join Tesla. But at the same time, Google and Apple are giving out the same call. I guess I would wonder what you would say to these people to have them join Tesla over these other companies. Second question is whether you had any takeaway in terms of your suppliers, in terms of this curvature experience, and how you can hold these suppliers' feet to the fire in some of these more complicated tasks that they're asked to fulfill.

Elon Musk: Yeah, in response to your first question: Apple and Google do not manufacture things themselves.

Alex Sage, Reuters: Right, but they are hiring manufacturing people.

Elon Musk: Uh, do to what?

Alex Sage, Reuters: That's a good question, but they are hiring manufacturing people.

Elon Musk: Okay, well, Tesla believes strongly in making things. They do not, that's fine, it's a philosophical difference. We believe that manufacturing technology is itself subject to a tremendous amount of innovation. In fact, we believe that there is more potential in manufacturing than there is the design of the car — by a long shot. Now, this is just a philosophical difference, perhaps we are wrong, but we believe in manufacturing. And we believe that a company that values manufacturing as highly as we do is going to attract the best minds in manufacturing.

Alex Sage, Reuters: But there's a supplier presence.

Elon Musk: I don't know who you're referring too.

Alex Sage, Reuters: Herbiger, the uh…

Elon Musk: Herbiger? I'm not familiar with that name.

Alex Sage, Reuters: It's the original—

Elon Musk: We certainly going to try to draw attention to that we're high standards of manufacturing and suppliers on the Model 3 program, those that didn't perform very well on prior programs are unlikely to be selected for the Model 3 program.