2017 Investor Roundtable:General Discussion

[mmd]

Elon Musk seems to have said on this week's earnings call that mmd is absolutely right. Do you believe that the discussion below can reasonably be interpreted as referring to nothing more than year-end run rates?

Tyler Charles Frank - Robert W. Baird & Co., Inc. (Private Wealth Management)

Okay. So, are you still on track, do you believe, for 1 million vehicles in 2020 and 500,000 in 2018? And then, as a quick follow-up to Model 3, you had previously talked about 20% gross margin on that. When do you think that margin target will be able to be achieved?

Elon Reeve Musk - Tesla, Inc.

Yes. I currently think that we should build to 500,000 vehicles next year and 1 million vehicles by 2020. That's 500,000 vehicles in total, Model S, Model 3, and Model X combined next year should – as far as the information I have at my disposal right now, I believe that is the most likely outcome. And then, with a couple more years, getting to 1 million units. That seems also the most likely outcome. Yes.

I re-read this; this time reading between the lines. My conclusion is, Elon meant "cumulative" production of 500k by next year and 1M by 2020. That's how his option tranches are granted, and that's how he probably thinks. So, he just double spoke a bit to make people think something that he can deny easily he actually said. Come 2020, he will say "Uh? We did get to 1M units in total in 2020. If you got too optimistic, it is your problem. " See the emboldened parts.
BTW, "build to 500,000" seems like a grammatical error.

That's the only way to reconcile the 10-K statement and Elon's statements in the CC, without one of those being an error.

 

[Yggdrasill]

It's not cumulative. Look at the Model 3 plans - 5k/month in 2017 and 10k/month in 2018. Sometime in 2018 they'll be at 480k Model 3 per year production rate (with one week shutdown per quarter), and that's just the Model 3. Add in 100k Model S/X and they'll be at 580k vehicle production rate.

(The actual production should be close to 500k as well in 2018. An average of 7500 Model 3 per week times 48 weeks, plus 100k Model S/X results in 460k vehicles produced in 2018. The may exceed 500k vehicles produced in 2018, but this depends on when in the year they achieve 10k Model 3 per week.)

 

[Yggdrasill]

BTW, "build to 500,000" seems like a grammatical error.

Probably meant "build production capacity to 500,000 vehicles next year and 1 million vehicles by 2020".

 

[PaulRocket]

So I was interested in how much revenue Tesla Energy or more specific the Powerwall might contribute to Q1. Unfortunately it doesn't seem like much. For Q2 however thinks look much rosier. I asked several installers from Australia, the US and Germany about when first installations of the Tesla Powerwall 2 will start.

Green Mountain Power (US):
"We have not received any details from Tesla that would enable us to design the product offering therefore do not have a timeline. Also, the technical specs have not been provided."

Energy Matters (Australia):
"As it stands, the Tesla Powerwall 2's are not in the country yet so in terms of the first installations it is hard to say. We are looking at potential install dates of around April-May time this year."

OffGrid Energy (Australia):
"The latest update from Tesla is that the shipment is expected in late April for the DC Powerwall, and late May for the AC Powerwall. This means that actual installations would start being scheduled for May and June respectively. These estimates are for our customers who already have orders in place for the Powerwall. New orders may have different timeframes depending on when the next shipments from Tesla will be arriving."

Solahart (Australia):
"We are still waiting for the Tesla Powerwall to be released and tested by Solahart Head Office."

The german installers and Tesla Store Frankfurt haven't answered yet, but I will update once I have an answer. From what I'm hearing they also expect installations to start in April.
Elon once said that he expects Powerwall unit sales to exceed vehicle sales. If we add a grain of salt, 50,000 Powerwall sales in 2017 would equal about $300M in revenue. Not bad!

Update:
PVT Hawaii (US):
"We will start installations when we are able to get our first shipment which continues to be delayed."

 

[schonelucht]

Well, calculating gross margin based on revenue listed in 10-K does not tell us much, as we do not know what percentage of the installed BES were paid outright vs. being part of PPA.

I have seen zero indication that a substantial amount of TE storage products was sold as a PPA. There is no discussion about increasing deferred revenue from energy storage systems either in the 10-K (there is plenty discussion about increases from solar energy systems, cars, autopilot, etc..) I suspect the reason is that most TE products are sold through a third party. Any PPA deals going on would be with that 3rd party which pays Tesla in turn in full. Although even for those projects I haven't seen any news coverage mentioning PPAs.

 

[techmaven]

I have seen zero indication that a substantial amount of TE storage products was sold as a PPA. There is no discussion about increasing deferred revenue from energy storage systems either in the 10-K (there is plenty discussion about increases from solar energy systems, cars, autopilot, etc..) I suspect the reason is that most TE products are sold through a third party. Any PPA deals going on would be with that 3rd party which pays Tesla in turn in full. Although even for those projects I haven't seen any news coverage mentioning PPAs.

The Kauai project is a Solarcity PPA, and so now it is a Tesla PPA with 52 MWh:
SolarCity's solar + storage play on Kauai

Since I believe installation completed and turn on happened in Q4, I am unclear what that means for financials.

Here is an interesting article on the CA Aliso Canyon energy utility scale storage situation:
Tesla, Greensmith, AES Deploy Aliso Canyon Battery Storage in Record Time

At the utility level, chemical battery energy storage in large quantities is just starting. It is becoming an accepted solution for a staunchly conservative industry. Looking back at 2014/2015 is instructive, but to a very limited extent. The project sales cycles here stretch to years previously. Now, more and more pilots are being done for thing like mitigating the lack of transmission capacity and avoiding or delaying the creation of additional peak capacity. It takes a while for utilities to move on this, and now we are seeing that movement. 2017 probably marks the turning point, but the 2017 results themselves is unclear. That's why it is interesting that Tesla was so busy making PowerPacks in Q1.

 

[jhm]

I re-read this; this time reading between the lines. My conclusion is, Elon meant "cumulative" production of 500k by next year and 1M by 2020. That's how his option tranches are granted, and that's how he probably thinks. So, he just double spoke a bit to make people think something that he can deny easily he actually said. Come 2020, he will say "Uh? We did get to 1M units in total in 2020. If you got too optimistic, it is your problem. " See the emboldened parts.
BTW, "build to 500,000" seems like a grammatical error.

That's the only way to reconcile the 10-K statement and Elon's statements in the CC, without one of those being an error.

Maybe, but it still sounds like annual production ("build") to me. "That's 500,000 vehicles in total, Model S, Model 3, and Model X combined next year" What is accumulated is models not years. So something like 375k Model 3 and 125k Model S/X would fit the bill. What should be clear to all is that Musk is not saying 500k Model 3 in 2018. That would not reconcile with starting the year at 5k/week for Model 3.

 

[schonelucht]

The Kauai project is a Solarcity PPA, and so now it is a Tesla PPA with 52 MWh:
SolarCity's solar + storage play on Kauai

Since I believe installation completed and turn on happened in Q4, I am unclear what that means for financials.

It's a SolarCity project. SolarCity contracted a PPA and then sourced the batteries from a third party (which happened to be Tesla). Therefore, if Tesla delivered the powerpacks before November 21st, it would be booked full value for TSLA. On the other hand, there are also media reports, that the project is still under construction AES’ New Kauai Solar-Storage ‘Peaker’ Shows How Fast Battery Costs Are Falling If that is true, it shouldn't appear on the books at all (or certainly not at full value). But I must concede that there is more chance that Tesla has PPA on its books for Energy Storage then I originally assumed. Still, I would have expected this to have been mentioned in the 10-K because then it should show up in a line item the report didn't have before. And it isn't as far as I can see.

 

[vgrinshpun]

It's a SolarCity project. SolarCity contracted a PPA and then sourced the batteries from a third party (which happened to be Tesla). Therefore, if Tesla delivered the powerpacks before November 21st, it would be booked full value for TSLA. .

Is this your conjecture, or you have source for this?

On the other hand, there are also media reports, that the project is still under construction AES’ New Kauai Solar-Storage ‘Peaker’ Shows How Fast Battery Costs Are Falling If that is true, it shouldn't appear on the books at all (or certainly not at full value)..

The uncertainty about revenue is another reason one can't use revenue from the 10-K the way you did to calculate what you claim is gross margin. The Mira Loma substation project was completed in late December of 2016, and while all of the PowerPacks installed there were definitely included in the COGS, it is not clear whether SCE completed all the payments and Tesla booked all the revenue associated with this project.

But I must concede that there is more chance that Tesla has PPA on its books for Energy Storage then I originally assumed. Still, I would have expected this to have been mentioned in the 10-K because then it should show up in a line item the report didn't have before. And it isn't as far as I can see

Do not forget that information about what Tesla charges for TE products is highly sensitive and they have every incentive to provide as little information as they can get away with. I think it will be awhile before you can see details about the TE revenue which you expected to see in this 10-K. BTW, I think that was the real reason behind JB's remark that it will be years before TE will appear as a separate line on company's statements...

 

[vgrinshpun]

I have seen zero indication that a substantial amount of TE storage products was sold as a PPA.

I have seen zero indication that Tesla booked full value for the PowerPacks supplied for the Kauai project, other than the claim you made.

 

[vgrinshpun]

Further regarding the uncertainty of the TE revenue included in 10-K, regardless of whether Tesla is part of PPA for the Kauai solar generation project, or supposed to be paid in full upon completion of the project, it is was not scheduled to be completed until early this year, the Powerpacks were delivered back in November of 2016. This means that these PowerPacks are included in COGS, but Tesla either did not or booked very little revenue associated with it.

 

[schonelucht]

Is this your conjecture, or you have source for this?

It is without any doubt a SolarCity project that sourced batteries after a review of third party suppliers of nearly 6 months. See http://kiuc.coopwebbuilder2.com/sites/kiuc/files/PDF/pr/pr2016-0216-solar.pdf As long as Tesla and SolarCity were two independent companies one would have invoiced the other for products delivered.

The uncertainty about revenue is another reason one can't use revenue from the 10-K the way you did to calculate what you claim is gross margin. The Mira Loma substation project was completed in late December of 2016, and while all of the PowerPacks installed there were definitely included in the COGS, it is not clear whether SCE completed all the payments and Tesla booked all the revenue associated with this project.

I guess we will find out next quarter. If margins suddenly turn hugely positive then this may be plausible.

Do not forget that information about what Tesla charges for TE products is highly sensitive and they have every incentive to provide as little information as they can get away with.

Not really. The fact that we can't exactly pinpoint the flow of money means that is small to begin with. As soon as it ramps as projected with revenues/cost of goods hitting tens of millions per month they simply can not hide that kind of movement anymore. Tesla is a listed company.

Anyway, opaque bookkeeping is rightfully seen as a net negative when pricing a stock. There is a point where investing becomes gambling when business practices are hidden too much away in difficult constructs.

 

[AlMc]

Further regarding the uncertainty of the TE revenue included in 10-K, regardless of whether Tesla is part of PPA for the Kauai solar generation project, or supposed to be paid in full upon completion of the project, it is was not scheduled to be completed until early this year, the Powerpacks were delivered back in November of 2016. This means that these PowerPacks are included in COGS, but Tesla either did not or booked very little revenue associated with it.

View attachment 217148

The question regarding revenue recognition sounds like one that Evanston should be able to answer

 

[schonelucht]

This means that these PowerPacks are included in COGS

They could also be in accounts payable.

 

[vgrinshpun]

They could also be in accounts payable.

I hope you are not serious. The Powerpacks were on Site in November. You'll need to subtract shipping time from California to Hawaii, manufacturing time, and shipping time from Japan to California...

 

[vgrinshpun]

It is without any doubt a SolarCity project that sourced batteries after a review of third party suppliers of nearly 6 months. See http://kiuc.coopwebbuilder2.com/sites/kiuc/files/PDF/pr/pr2016-0216-solar.pdf As long as Tesla and SolarCity were two independent companies one would have invoiced the other for products delivered.

The link you provided documents agreement between the Cooperative and Solar City. It has no information on the agreement between the Solar City and Tesla. The claim that Tesla supposed to be paid in full by Solar City is your conjecture.

 

[vgrinshpun]

I guess we will find out next quarter. If margins suddenly turn hugely positive then this may be plausible.

There is no need to wait for the next quarter. Between the two largest Tesla projects, Kauai Island and Mira Loma Substation, at least some revenue was not fully booked, while all cost were incurred during 2016. Not only it is plausible, it is almost certain. You just do not want to acknowledge it because it does not fit your bearish view on TE. We will have to register our disagreement here and move on.

 

[Vitold]

-

Maybe, but it still sounds like annual production ("build") to me. "That's 500,000 vehicles in total, Model S, Model 3, and Model X combined next year" What is accumulated is models not years. So something like 375k Model 3 and 125k Model S/X would fit the bill. What should be clear to all is that Musk is not saying 500k Model 3 in 2018. That would not reconcile with starting the year at 5k/week for Model 3.

That's how I understood it myself, however, cumulative could also mean the rate of production of S,3,X combined.

 

[FredLambert]

Kauai is going online next week. Most of the PW2 were delivered in Jan-Feb.

 

[schonelucht]

If Kauai is PW2, then batteries were manufactured in GF... meaning no COGS for batteries in 2016 (and no revenue either) making @vgrinshpun and my back-and-forth without subject. Negative gross margin for Tesla Energy Storage in 2016Q4 confirmed.

AES was able to outcompete Tesla for the follow-up project on the island bidding 11 cents instead of 13,9 cents per kWh for solar+storage. Kauai Shows Solar + Storage is Starting to Become Cost Competitive With Fossil Fuels, Nuclear It does strengthen the case of AES that being able to freely pick best in class suppliers instead of being bound to in-sourced components increases it competitiveness.