Tesla, TSLA & the Investment World: the Perpetual Investors' Roundtable

[neroden]

I had to replace a tire on my S. While I was in the waiting room, I was talking to another customer who drove a 911 twin turbo. He had driven a 3 performance and was thinking that might be his next car. The girl behind the desk piped in that "Yeh, but they catch on fire." Don't tell me we don't have a brand problem. If we do advertise, we should tout the benefits of EVs first and then mention that Tesla is the leader in EVs.

Actually, if we advertise, we should simply advertise all the gasoline car fires and explosions, and then print the stats for gasoline car fires versus the tiny numbers of EV fires, followed by the "safest car on the road" stats. Maybe how people burning horribly to death in a gasoline car, and then show the guy who calmly got his briefcase out after his Tesla was on fire.
If the fire nonsense is the FUD which stuck, take it on directly.

 

[neroden]

Tesla didn't deliver enough SR+ in Q1 to have a big impact on gross margin.

Q1 GAAP Cost of Automotive Sales was 2856m. This was after a 409m adjustment for sales booked in prior periods that have Resale Value Guarantees. COGS for the cars actually sold in Q1 was thus 3265m. Tesla sold 50,928 Model 3s and an estimated 11,072 S/X. They also leased an estimated 1019 S/X. Note that the lease/sale split could be off by ~250 in either direction. Rearrange the COGS equation to solve for Model 3:

11,072 * S/X COGS + 50,928 * Mod3 COGS = 3265m
Mod3 COGS = [3265m - (11,072 * S/X COGS)] / 50,928

S/X Model 3
75k 47.8k
80k 46.7k
85k 45.6k
90k 44.5k

S/X COGS have historically been around 75k. Poor fixed cost absorption probably drove them to ~80k in Q1. I could be talked into 85k. I can't see how they could possibly be 90k, though. I figure SR+ COGS are roughly 5k less than LR-AWD. I base this on a $3k difference in battery cost (22 kWh * 125/kWh cells + ~250 pack cost savings) and $2k for the front drive unit and power electronics. Battery savings could be a bit higher, but extra AWD cost is probably lower.

Q1 deliveries were 50% US (mix of all variants), 40% Europe (all LR-AWD/P), and 10% China (all LR-RWD and above). You can play around with the numbers all you want, but I always come out with 41-43k SR+ COGS and 46-48k LR-AWD COGS.

Given the teardown numbers, I think there's going to turn out to be a lot of low-hanging fruit, depreciation from scrapped equipment, fixed-cost depreciation, etc. in that number. Historically, cost reduction is something Musk and Tesla have actually been *very good* at. I'm sure they'll get it down.

They got the COGS down for the Roadster as time went on; they did for the S; they did for the X. It may take time, but I'm not concerned about that.

 

[anon]

At the end of November, an e-mail from Musk to employees said that the cost of a standard range Model 3 at that time would be $38k. It's unclear if that was for the basic SR that was never made or for the SR+. Hopefully the cost has come down since then, especially with the ramp to 7k/week this quarter, but perhaps not if that was the SR price when they are now selling SR+ instead.
Tesla achieves Model 3 production of 1,000/day, pushes to maintain it and reduce costs - Electrek

 

[anon]

I base this on a $3k difference in battery cost (22 kWh * 125/kWh cells + ~250 pack cost savings) and $2k for the front drive unit and power electronics. Battery savings could be a bit higher, but extra AWD cost is probably lower.

I expect considerably higher battery pack cost savings - in late 2018 (probably Q3 earnings call, can't remember for sure) Musk talked about how they had designed a different battery pack for the SR to give much larger cost savings than were possible for the MR. This suggests a lot more than $250 on top of the cell cost to me. The savings are not only in the components of the pack, but also in a cheaper production process, presumably due to greater automation with new equipment installed in February.

 

[Zaxxon]

Interesting video from down under: the Victoria Police now have a Model X patrol car, and they mention integrated MCU controls for their lights/siren/body cams/etc. Also, that Tesla engineers are assisting with the integration and will be checking back to see what works well and what could use improvement. Maybe the start of Tesla having a formal law enforcement mod?

Victoria Police on Twitter

 

[Hock1]

I disagree in the following sense:

Rivian has a real announced pickup, not a concept vehicle, shown both at LA and NY auto shows, and it will be manufactured later this year. I have seen it up close (they had one at an outdoor expo in Arizona a few weeks ago) and talked to some of the team. Vehicle and specs and timeline are very credible, including a factory in Illinois that they got at rock bottom prices.

Tesla does not have an announced pickup vehicle. I hope they do soon.

Yes. But, they do have 8 years of mass manufacturing experience. That's not nothing.

 

[SPadival]

Rumor in The Netherlands is that Elon will be here on Monday, Tuesday and Wednesday to attend The Global Entrepreneurship Summit in The Hague and the YES!Delft incubator at Delft Technical University. I can’t imagine him staying that long though.

(A heads up, so we don’t have to start speculating about the European Gigafactory when his jet crosses the Atlantic)

Here comes Euro Gigafactory

 

[gene]

If the renter is willing to pay for it with a licenced electrician the landlord almost always says yes.

In California it is the law that the renter can install charging outlet.

Electric Vehicle Charging Stations for California Landlords - Kimball Tirey & St. John LLP

 

[AcesDealt]

Tesla didn't deliver enough SR+ in Q1 to have a big impact on gross margin.

Q1 GAAP Cost of Automotive Sales was 2856m. This was after a 409m adjustment for sales booked in prior periods that have Resale Value Guarantees. COGS for the cars actually sold in Q1 was thus 3265m. Tesla sold 50,928 Model 3s and an estimated 11,072 S/X. They also leased an estimated 1019 S/X. Note that the lease/sale split could be off by ~250 in either direction. Rearrange the COGS equation to solve for Model 3:

11,072 * S/X COGS + 50,928 * Mod3 COGS = 3265m
Mod3 COGS = [3265m - (11,072 * S/X COGS)] / 50,928

S/X Model 3
75k 47.8k
80k 46.7k
85k 45.6k
90k 44.5k

S/X COGS have historically been around 75k. Poor fixed cost absorption probably drove them to ~80k in Q1. I could be talked into 85k. I can't see how they could possibly be 90k, though. I figure SR+ COGS are roughly 5k less than LR-AWD. I base this on a $3k difference in battery cost (22 kWh * 125/kWh cells + ~250 pack cost savings) and $2k for the front drive unit and power electronics. Battery savings could be a bit higher, but extra AWD cost is probably lower.

Q1 deliveries were 50% US (mix of all variants), 40% Europe (all LR-AWD/P), and 10% China (all LR-RWD and above). You can play around with the numbers all you want, but I always come out with 41-43k SR+ COGS and 46-48k LR-AWD COGS.

Yes. I took s/x up close to 85k because it was the only way to get the actual gross margin on model 3 high enough to hit 15% which is explicitly claimed. So by looking at revenue you further close in on the answer.

I've actually been favoring Tesla by giving them a 50% implied incremental margin on upgrade delta between sr+ and higher trims, but the margin is likely higher which is actually worse for the cost picture.

People shouldn't allow others to derail them from thinking about this. It is the most important fact right now, and the most logical reason the stock has been imperiled. If costs remain this high, Tesla can only crawl out of the profit hole by selling a highly unlikely number of 45k$+ configurations and the unit sales count for SR+ ends up being fairly inconsequential.

 

[RFernatt]

Actually, if we advertise, we should simply advertise all the gasoline car fires and explosions, and then print the stats for gasoline car fires versus the tiny numbers of EV fires, followed by the "safest car on the road" stats. Maybe how people burning horribly to death in a gasoline car, and then show the guy who calmly got his briefcase out after his Tesla was on fire.
If the fire nonsense is the FUD which stuck, take it on directly.

Create an ad for a fictitious insurance product to cover your family in the event of a gasoline car fire since X thousand occur each year resulting in X deaths. Fine print points you to EV stats.

Can't take credit for the concept. Just read an article about LA doing this same thing around firearm violence.

 

[RFernatt]

That is ludicrous!
You can get about an 8kW PV system installed for that price.

But, then again, the Powerwall as it is right now is not something to buy for economic reasons.

Agreed. It would never pencil out in raw numbers due to our net metering policy, relatively low electric cost, and no ToU pricing. You just have to assign a large personal value to it to keep the solar system in operation during outages and for backup power. After being without power for 2 weeks following a derecho event and living with unreliable power for years, your assigned value increases, but not double what I expected to pay!

 

[gene]

I finally got a quote on a single PW in a simple installation for $19,150. To be located in an attached garage with concrete floor, 2 feet away from the main breaker box and solar inverter with exposed conduit. We're not in an incorporated city limits, so no real building codes, just the loose county/state stuff. I would like to have one, but that's just unreasonable. In WV, there is only one certified installer and I can't seem to get nearby out of state installers to be interested. I can be in 3 other states in 20 minutes and DC in 90 minutes. I will bide my time.

I am in California where we have the SGIP incentive. I have two Powerwalls arriving in July.

BatteryTesla Powerwall 2 AC
x2 with instal, permitting etc.
Cash purchase
Total amount

$23,094.50

Federal tax credit**

$6,928.35

SGIP incentives & Rebates**

$8,120.00

Net cost

$8,046.15

 

[EVNow]

I figure SR+ COGS are roughly 5k less than LR-AWD. I base this on a $3k difference in battery cost (22 kWh * 125/kWh cells + ~250 pack cost savings) and $2k for the front drive unit and power electronics. Battery savings could be a bit higher, but extra AWD cost is probably lower.

I think that difference is too optimistic. Then there are some other things that are different too, apart from extra motor & battery.

Yes, Q1 had too little SR+ to actually figure out the COGS. A bit of change in ASP of LR/AWD and a few options on SR+ can easily change a lot of things.

If we look forward and assume 50/50 mix of SR+ and LR/AWD/P and see what Tesla needs for 20% margin, looks to me either they need some 10% margin on SR+ or almost 40% margin on others.

ps :

Q1 deliveries were 50% US (mix of all variants), 40% Europe (all LR-AWD/P), and 10% China (all LR-RWD and above). You can play around with the numbers all you want, but I always come out with 41-43k SR+ COGS and 46-48k LR-AWD COGS.

​

You can come up with any COGS for SR+ with low % of SR+ sales in Q1 by making small changes in LR COGS.

 

[Artful Dodger]

Won't happen. The Roadster battery replacement program worked out to be expensive and frustrated owners (go over to that section of this forum to learn); Tesla's learned their lesson and is not going to offer deep retrofits any more.

With the Model S body life expectancy of 25 years, and the bty pack half of that, owners of Model S cars will demand battery replacements. As this bty swap video shows, this isn't a "deep retrofit"; its a planned feature. It just a feature the will just be used at end-of-bty-life rather than an alternative to supercharging:

Roadster bty replacement has exactly ZERO in common with the technology. 'Roadster owner frustration' over a feature never intended for Roadster 1 is understandable, but irrelevant to the Model S feature we are discussing.

The original Roadster also doesn't support supercharging. That would be a deep retrofit. Just like adding AP2 to AP1 Model S cars, which is also something Tesla will not do because it is cheaper to buy a new car.

Tesla has learned that its customers demand the best tech, and they have shown a willingness to provide it as a retrofit when it extends the life cycle of the vehicle. There's 2,000 roadsters and there will be 100s of thousands of Model S cars. So you bet, it'll happen en mass, just not yet.

Cheers!

 

[⚡️ELECTROMAN⚡️]

If the brand "is just fine", why is the share price in the crapper? I ask as a MS owner and long share holder.

Apparently the health of a company and it’s stock price aren’t necessarily related to each other.

 

[EVNow]

It is the most important fact right now, and the most logical reason the stock has been imperiled.

There is no logic to the market. All the shorts (and analysts) are saying there is no demand. Not sure why you would invent a "logical reason" that nobody is talking about.

 

[⚡️ELECTROMAN⚡️]

Rivian gets decent press, but let's be real guys, they are not competition that Tesla should be worried about any time soon.

I actually don’t see any competition that Tesla should be worried about any time soon. They let Tesla get too far ahead.

 

[Artful Dodger]

Another interesting tidbit.

The SpaceX Performance Package for the Roadster 2 provides 3 g’s of thrust. Meaning it can accelerate at 2 g’s straight up, or provide roughly 15,000 lbs of thrust (assuming weight of 5k lbs and I’m not getting confused)

Ergo, the quickest car 0-60 in the world even going straight up! This seems so astounding as to be unbelievable!

Ever stood on deck of a ferry, just as the air horn goes off? Yeah, its gonna be that loud. Got a private race track? That's were it'll be legal to use.

The 200-odd pounds of air in the COPV will last about 1/75 th of a second at that flow rate. Think millisecond bursts intended to 'save your ass' instead. But if you're driving that hard, you shouldn't be in a street legal car at all.

Cheers!

 

[⚡️ELECTROMAN⚡️]

I disagree in the following sense:

Rivian has a real announced pickup, not a concept vehicle, shown both at LA and NY auto shows, and it will be manufactured later this year. I have seen it up close (they had one at an outdoor expo in Arizona a few weeks ago) and talked to some of the team. Vehicle and specs and timeline are very credible, including a factory in Illinois that they got at rock bottom prices.

Tesla does not have an announced pickup vehicle. I hope they do soon.

Bring it on!

 

[bhtooefr]

This is strange. They removed P +/- differences to streamline production and now those differences are back?

To be fair, the difference between Dual Motor and Performance Minus, when they were simultaneously produced, was one bit in the software - see all the reports of Dual Motor and Performance Minus getting mixed up, and Tesla unlocking Dual Motors to Performance Minus when someone bought a Performance Minus and got a Dual Motor delivered to them.

And, IIRC, Dual Motor, Performance Minus, and Performance Plus already get the same rear drive unit part number as a service part anyway, so what that tells me is that the motor yields are to the point that Tesla decided to flip the switch for all RHD cars to Performance Minus default instead of Dual Motor default.

I expect CCS to catch up to and possibly surpass CHAdeMO in terms of useful locations (but perhaps not - as most will likely be dual CHAdeMO / CCS installations, they may end up as a "tie" for locations). Once something close to parity happens, CCS will be the clear preference for Tesla owners as they are more likely to get a higher charge rate from CCS than CHAdeMO.

Dual CHAdeMO/CCS installs often have a single token CHAdeMO for LEAF owners, and then multiple CCS stalls, so in terms of plugs, CCS will win, too.

When driving between destinations without customers, an empty FSD car can be driving super conservatively: yield, keep to the speed limit, factor in weather conditions, etc. etc. Since the violence of traffic accidents scales with energy, and energy scales with the square of speed, just dropping the speed to the speed limit has a significant effect on insurance costs and liabilities. The FSD car will also not be annoying customers by driving in a timid, easily bullied fashion.

Unless it arrives late to the pickup...

It also means that somehow, 44% of Mod3 owners find a way to charge without wiring their garage.

Not necessarily - that 44% includes rented homes, and it includes owners that rent garage spaces and got those wired. And, the 56% that are homeowners may not have wired their garages, too...

Won't happen. The Roadster battery replacement program worked out to be expensive and frustrated owners (go over to that section of this forum to learn); Tesla's learned their lesson and is not going to offer deep retrofits any more.

It wasn't the swappability which was the issue. Go into the details. The new battery didn't perform as well as people were hoping, and it led to disappointment. It also involved a lot of service labor -- which is in short supply -- and design work with fairly low profit margins. Even with the simpler swap on Model S... still labor time and time in the service centers.

There is a reason that Tesla basically does not offer hardware battery pack upgrades on Model S -- though if your pack fails (so they already have to do a replacement) and you offer to pay them at that time, you can get them to give you a different size pack. They don't think the hassle is worth it. They aren't going to offer a limited edition upgrade for a small number of cars. Their service is stretched too thin for it.

That's the thing, the new battery for the Roadster 1.x/2.x was a custom design for that car, with cells chosen for that car. The performance is downright dreadful, and Tesla really does need to do something for those owners, but it's not indicative of what would happen on a later model.

A 2020 Model S battery is presumably a unit designed to go into the same unibody mounting points as a 2012 Model S battery, with similar wiring and cooling connections. So, a battery upgrade program for the Model S is orders of magnitude easier than the one for the Roadster (it's retrofitting an existing pack meant to go into that unibody, instead of designing and validating (or not validating, apparently) a new pack inside of the old case), it's just that for reasons like what you mentioned, they don't want to actually do it.