Long-Term Fundamentals of Tesla Motors (TSLA)

[erha]

Pretty sure this will still be a niche compared to the grid that controls home electricity usage and commercial entities on land, with the home usage being the largest consumer, this is just a hunch though, I am sure the numbers are relatively easy to find.

Agree, I see the word huge came twice there unintentionally. There are a lot more households than ships

 

[ckessel]

Interesting quote from CS report:

"Electric vehicles are inherently better than Internal Combustion (ICE) vehicles," said Gaives. "If Tesla can get to cost-parity with ICE and still offer $1,400-$2,500 per year fuel savings to the consumer, it won't be a fair fight."

Read more: http://www.benzinga.com/analyst-rat...overage-on-tesla-says-its-not-a#ixzz3AN0g8cOl

I think the first part of that statement is the part that's really not understood by the public. Electric engines are simply superior (that's why they're used for almost everything instead of ICE). I certainly had no idea before I looked at Tesla. Or, more accurately, it wasn't that I didn't know in theory based on facts rattling about in my head, it's that I never gave it a thought at all.

Most EVs just aren't good enough to get the general population to leave the devil they know (ICE) and go look at them. That's not to say something like the LEAF isn't good, it's just not good enough to get people to break away and reach mass appeal against it's ICE competition. Once EVs break down the cost, range, and recharge time barriers (much like the S has done relative to it's competition), it's going to be a quick tipping point.

One of the early Model S reviewers (Dan Snyder maybe?) said, sort of prophetically, that he "couldn't shake the feeling of a corner turned". Once the Model 3 comes out, it's going to become startling clear that the entire market is ready to make that turn.

 

[Perfectlogic]

I disagree with your line of thinking:

The battery storage market is a lot better business, because it doesn't require the huge infrastructure that building and servicing vehicles requires.

And yes, TSLA will be able to charge 50%+ gross margins since they will be by far the lowest cost battery storage producer in the world. If they can make batteries at $100/kWh and the rest of the world is making them at say $200/kWh, then TSLA can Charge $250/kWh and make a 60% gross margin, while the rest of the world is selling at the same price for a 20% gross margin.

Those margins will come down over time, but still there are a lot of benefits to simple selling batteries over cars. The only reason Tesla is selling cars is because there is not enough demand for battery storage today at current prices, so TSLA is doing things based on market conditions.

Selling cars might make the top line look better, but selling batteries is significantly better for the bottom line. And the bottom line is all that matters, since that is what goes to the shareholders.

So you are saying that the more simple the industry the higher the margin opportunity? I am confused, it is generally accepted that a commodity will be sold at a lower margin compared to products with more qualitative differentiation.

I agree that Tesla has an edge in the battery business, but I think producing at half the cost of competitors is being overly optimistic, LG just announced plans to produce a battery giving a car a 200 mile range intended to be sold around $35k. A part of Teslas battery edge lies in the quality of the battery as I have understood it, and that part doesn't really have much value in storage as you won't need to pull out large amounts of electricity at a time.

 

[erha]

So you are saying that the more simple the industry the higher the margin opportunity? I am confused, it is generally accepted that a commodity will be sold at a lower margin compared to products with more qualitative differentiation.

I agree that Tesla has an edge in the battery business, but I think producing at half the cost of competitors is being overly optimistic, LG just announced plans to produce a battery giving a car a 200 mile range intended to be sold around $35k. A part of Teslas battery edge lies in the quality of the battery as I have understood it, and that part doesn't really have much value in storage as you won't need to pull out large amounts of electricity at a time.

That's what I was wondering about as well, but this situation is a little bit special because there are no true competitors in the battery storage industry so far. In the car industry Tesla is competing with ICE's, not the Leaf and the i3. Tesla's battery management systems are superior everything else making it possible to use cheap batteries. If it would stay like this forever then Tesla should try to use all it's batteryproduction for storage units, but sooner or later there is going to be competition bringing the margins down to a more sensible commodity level. After all, the chemistry in Tesla's batteries is not Tesla's edge.

 

[eepic]

Pretty much every household has a car, if every household would have a storage unit too in a bullish scenario (wouldn't it be more efficient to have grid storage more centralized?) it still wouldn't come close to the auto market as I see it, as the battery is only 20-25% of the cars cost and the battery for the household would probably be significantly smaller than the car battery. I'm also pretty sure you won't see even close to 50% margins on a commodity like batteries in the future.

Just doing a bit of back of napkin math.

Let's say over the next few years/decades renewables+batteries because economically attractive, and we transition over 20 years to a system that supports a stationary storage buffer of about 5% of daily energy usage.

Using 143,851 TWh/year from World energy consumption - Wikipedia, the free encyclopedia

143,851 TWh/year = 394,000,000,000 kWh/year * 5% = 19,700,000,000 kWh stationary storage

Let's say that storage is introduced at a steady pace over twenty years, then each year there's a demand for 985,000,000 kWh stationary storage. Let's say they've streamlined the supply chain with multiple gigafactories, and have a $50/kWh cost vs $125/kWh pricing.

$125 kWh x 985,000,000 = $123 billion per year revenue
$75 x 985,000,000 = $73.9 billion per year in gross margin

Of course, it's going to be production constrianed to the ability of Tesla to scale up to this. But like Elon alluded to before, if stationary storage becomes economically attractive the demand will be quasi-infinite for decades.

 

[sleepyhead]

So you are saying that the more simple the industry the higher the margin opportunity? I am confused, it is generally accepted that a commodity will be sold at a lower margin compared to products with more qualitative differentiation.

I agree that Tesla has an edge in the battery business, but I think producing at half the cost of competitors is being overly optimistic, LG just announced plans to produce a battery giving a car a 200 mile range intended to be sold around $35k. A part of Teslas battery edge lies in the quality of the battery as I have understood it, and that part doesn't really have much value in storage as you won't need to pull out large amounts of electricity at a time.

You are also underestimating the fact that TSLA is specifically building a $5b, 35GWh gigafactory to minimize costs as much as possible. Nobody else in the world is doing that, so competitor's costs will not be anywhere near as close to Tesla's. But even if they were, there will still be a huge battery shortage at $100/kWh, which will naturally lead to higher margins for the first years, if not decades.

Also, we know that Tesla had a meeting with Apple. Everyone was speculating takeover bid. But I think it had to do with creating solar cells etched into glass (or sapphire to be specific). Apple is working with GTAT on a sapphire project, but I have a feeling that it will be a lot more. Apple also is working on etching solar cells directly into the sapphire of the phone. When the iphone6 comes out in 4 weeks, I am certain that you will see a sapphire screen. Then in a couple of years, you will see sapphire front and back. This will allow them to etch solar cells directly into sapphire and make them invisible. Your phone will then charge by itself any time you lay it down on the table; or at a minimum keep the battery from draining in stand-bye mode.

GTAT has this one of a kind Hyperion technology that can revolutionize the world. They can use it to creat ultra-thin 20 micron silicon carbide (SiC) substrates; that can be used to grow graphene electrodes that would make batteries 10x better. This could be a potential gamechanger; and I fully expect Elon Musk to be working with GTAT in using their technology; especially when it comes to SCTY but also for Tesla. Also, I would not be surprised to see TSLA incorporate solar cells into the glass of the car, and we will have a self-charging car in a few short years (maybe decade).

The future is bright, and a lot more "science fiction-ee" than most people can imagine.

 

[eepic]

So you are saying that the more simple the industry the higher the margin opportunity? I am confused, it is generally accepted that a commodity will be sold at a lower margin compared to products with more qualitative differentiation.

I agree that Tesla has an edge in the battery business, but I think producing at half the cost of competitors is being overly optimistic, LG just announced plans to produce a battery giving a car a 200 mile range intended to be sold around $35k. A part of Teslas battery edge lies in the quality of the battery as I have understood it, and that part doesn't really have much value in storage as you won't need to pull out large amounts of electricity at a time.

The battery edge Tesla has is that they are able to use one of the best "quality" chemistries (particularly superior in degradation/cycle life) and innovate in the pack in such a way to produce a safe and efficient pack that's still multiples cheaper than competitors. Of course, that's not to say others can't catch up.

 

[sleepyhead]

Just doing a bit of back of napkin math.

Let's say over the next few years/decades renewables+batteries because economically attractive, and we transition over 20 years to a system that supports a stationary storage buffer of about 5% of daily energy usage.

Using 143,851 TWh/year from World energy consumption - Wikipedia, the free encyclopedia

143,851 TWh/year = 394,000,000,000 kWh/year * 5% = 19,700,000,000 kWh stationary storage

Let's say that storage is introduced at a steady pace over twenty years, then each year there's a demand for 985,000,000 kWh stationary storage. Let's say they've streamlined the supply chain with multiple gigafactories, and have a $50/kWh cost vs $125/kWh pricing.

$125 kWh x 985,000,000 = $123 billion per year revenue
$75 x 985,000,000 = $73.9 billion per year in gross margin

Of course, it's going to be production constrianed to the ability of Tesla to scale up to this. But like Elon alluded to before, if stationary storage becomes economically attractive the demand will be quasi-infinite for decades.

Thanks for running the numbers, so now we know what I am alluding to here.

I would just point out that $50/kWh production cost is very ambitious since Elon has said that a battery is made up of $60-$80/kWh of raw materials. I would shoot for $80-$100/kWh cost and $150-$200/kWh installation cost in the long run. Initial margins will be a lot higher and then slowly taper down over many years, but will be more than offset by higher volume.

Also, I think you meant $73.9b in gross profit and not gross margin. Sorry, but the accounting nerd in me... :wink:

 

[SteveG3]

I disagree with your line of thinking:

The battery storage market is a lot better business, because it doesn't require the huge infrastructure that building and servicing vehicles requires.

And yes, TSLA will be able to charge 50%+ gross margins since they will be by far the lowest cost battery storage producer in the world. If they can make batteries at $100/kWh and the rest of the world is making them at say $200/kWh, then TSLA can Charge $250/kWh and make a 60% gross margin, while the rest of the world is selling at the same price for a 20% gross margin.

Those margins will come down over time, but still there are a lot of benefits to simple selling batteries over cars. The only reason Tesla is selling cars is because there is not enough demand for battery storage today at current prices, so TSLA is doing things based on market conditions.

Selling cars might make the top line look better, but selling batteries is significantly better for the bottom line. And the bottom line is all that matters, since that is what goes to the shareholders.

Until someone else starts producing cells on a Gigafactory scale Tesla will indeed have a large cost advantage. I 'm not however convinced we should expect a business with sustainable 60% margins. Yes, Tesla's cost per kWh for packs blows away other automakers but that's because they're all going for larger format cells... none of them have gone for the small cells Tesla has made. I've seen a variety of thoughts on why this is so and what I've seen centers around the other automakers having perceived challenges with using small cells for a pack in an automotive application. No such issue with grid storage. Thus I only see supply constraints limiting the other battery producers from jumping into storage if they see Tesla & Panasonic making such fat margins. Again, Tesla would continue to do quite well here, likely for many years to come, but it's I just not clear to me 60% margins would have a long shelf life. I see it depending on the size of demand. If demand for grid storage is massive at $200-300 kWh as Tesla gets down to $100-150, your 60% margins may have legs... it would simply take demand large enough that even with Tesla/Panasonic and all the competition going after the demand at full tilt the grid storage market remained supply constrained for decades. I think this is possible,

 

[erha]

You are also underestimating the fact that TSLA is specifically building a $5b, 35GWh gigafactory to minimize costs as much as possible. Nobody else in the world is doing that, so competitor's costs will not be anywhere near as close to Tesla's. But even if they were, there will still be a huge battery shortage at $100/kWh, which will naturally lead to higher margins for the first years, if not decades.

Also, we know that Tesla had a meeting with Apple. Everyone was speculating takeover bid. But I think it had to do with creating solar cells etched into glass (or sapphire to be specific). Apple is working with GTAT on a sapphire project, but I have a feeling that it will be a lot more. Apple also is working on etching solar cells directly into the sapphire of the phone. When the iphone6 comes out in 4 weeks, I am certain that you will see a sapphire screen. Then in a couple of years, you will see sapphire front and back. This will allow them to etch solar cells directly into sapphire and make them invisible. Your phone will then charge by itself any time you lay it down on the table; or at a minimum keep the battery from draining in stand-bye mode.

GTAT has this one of a kind Hyperion technology that can revolutionize the world. They can use it to creat ultra-thin 20 micron silicon carbide (SiC) substrates; that can be used to grow graphene electrodes that would make batteries 10x better. This could be a potential gamechanger; and I fully expect Elon Musk to be working with GTAT in using their technology; especially when it comes to SCTY but also for Tesla. Also, I would not be surprised to see TSLA incorporate solar cells into the glass of the car, and we will have a self-charging car in a few short years (maybe decade).

The future is bright, and a lot more "science fiction-ee" than most people can imagine.

Sounds cool, but I think to keep the solar cells at roofs are better. You want them to be positioned perpendicular to the sunlight in order to maximize the flux. You also want them to be positioned at a place where they receive sunlight for the longest time possible each day. Hence, a car parked in a garage is a waste of solar cells or solar cell roads. A car has a far too small area to generate enough energy to make a real difference

 

[Perfectlogic]

$125 kWh x 985,000,000 = $123 billion per year revenue
$75 x 985,000,000 = $73.9 billion per year in gross margin

Again I would argue that 60% gross margins for a commodity is extremely optimistic, you also imply that Tesla will have a market share of 100%, again very optimistic. Your math uses 5% of a years energy usage, or more than 18 days of energy usage as buffer, I don't know if this is necessary or not tbh, but either way overall a very optimistic outlook. Say Tesla had 50% market share which I still would say is optimistic and a gross margin of 30% (also optimistic imo), then revenue would be $61,5B with $18,45B in gross profit. I think Tesla's car sales will crush even this optimistic storage forecast going well into the hundreds of billions in sales with time.

 

[ItsNotAboutTheMoney]

Sounds cool, but I think to keep the solar cells at roofs are better. You want them to be positioned perpendicular to the sunlight in order to maximize the flux. You also want them to be positioned at a place where they receive sunlight for the longest time possible each day. Hence, a car parked in a garage is a waste of solar cells or solar cell roads. A car has a far too small area to generate enough energy to make a real difference

Rear windshield
- good sized area of glass, especially in semi-kammback aerodynamic cars
- less subject to damage that requires replacement, so a higher cost isn't that big a deal
- already has power for heating
- ... so transparency is obviously not such a big deal
- allows a lot of heat into the car; reducing heat helpful in hot locations

PiP uses solar to run a ventilation fan and keep the interior temperature close to ambient
Leaf uses solar to trickle charge the 12V

I think integrating solar into the rear windshield would be a good idea. Then there are panoramic roofs with more area.

 

[eepic]

Again I would argue that 60% gross margins for a commodity is extremely optimistic, you also imply that Tesla will have a market share of 100%, again very optimistic. Your math uses 5% of a years energy usage, or more than 18 days of energy usage as buffer, I don't know if this is necessary or not tbh, but either way overall a very optimistic outlook. Say Tesla had 50% market share which I still would say is optimistic and a gross margin of 30% (also optimistic imo), then revenue would be $61,5B with $18,45B in gross profit. I think Tesla's car sales will crush even this optimistic storage forecast going well into the hundreds of billions in sales with time.

Very rough back of napkin math, with very basic assumptions. But I would argue these are one *plausible* outcome, with my justification below (of course, there are many other ways this can play out).

If others can barely fathom producing storage solutions at $100/kWh cost in 15-20 years (hence the analyst "whoa" on the recent Q2 call), then it has several implications.

Let's say at $125/kWh storage, the solar+storage solution becomes economically advantageous to the traditional grid, with a huge amount of demand there. While others are slowly weening down their cost basis towards a competing product, Tesla would be the *only* supplier able to supply the market at that price, independent of Tesla's cost basis (sub $100/kwh). Of course in the longer term competition will bring the percentage margin much lower, but that could be many years. Assuming equal technology improvements, someone else will have to produce at the same streamlined scale to bring margin pressures.


However, I don't see the relevance of days-per-year metric of storage requirements you mentioned. Renewables like solar only produce power part of the day (and could produce little to no energy for periods of days due to weather), which is why storage is needed in the first place... in a world with purely renewables it would be like 30-50% storage requirements. An ideal long term balance would still have some fossil fuel and nuclear sources as a base load to reduce storage requirements. If renewables become economic without subsidies, 5% storage is not an unreasonable estimate if you look out a quarter century.

I would argue the realistic market share Tesla can capture with autos has a much lower ceiling (Toyota has < 10% market share and has a huge lineup of products), due to styling and personal preferences.

 

[austinEV]

We are getting pretty off topic, but remember even ignoring renewables there is a market for storage. Homeowners and businesses can avoid peak useage fees in the summer and/or power companies can use them to similarly bridge peak times. iirc, a typical fossil fuel grid has power plant capacity that is idle 95% of the time and only needed for that last bit of peak demand. Eliminating the need for that last 5% is disproportionately valuable.

 

[Perfectlogic]

eepic, you were the one using 5% of yearly energy consumption in your calculation, you wrote 5% of daily energy usage, but you are clearly using 5% of yearly

143,851 TWh/year = 394,000,000,000 kWh/year * 5% = 19,700,000,000 kWh stationary storage

Tesla might be able to sell batteries for storage with a high margin for 5 years or whatever till competition arrives (which obviously will happen relatively quickly if they sell commodities with a margin of 50%) but the profit opportunity of that relatively short period with Tesla low production earmarked storage is low looking at the big picture so it is not something I emphasize in my evaluation of Tesla.

in a world with purely renewables it would be like 30-50% storage requirements

30-50% of yearly energy consumption stored as a buffer? That seems quite high to me, I think there will be other solutions for when solar/wind doesn't produce enough like nuclear as you say, or fossils as a backup in the rare cases where shortages happen to keep the storaged buffer at 5% of yearly usage max.

I would argue the realistic market share Tesla can capture with autos has a much lower ceiling (Toyota has < 10% market share and has a huge lineup of products), due to styling and personal preferences.

I guess I agree, but the thing is that the auto industry is enourmous, it's $2 trillion which is much larger than even your highly optimistic storage estimates, and the auto market is expected to see a significant growth in the coming years. Looking at how innovative Tesla is in its business and their quality of execution compared to Toyota with their horribly bet on fool cells I do think Tesla can achieve a higher than 10% market share in time, and with higher margins than what Toyota is doing given their vertical integrated business and head start in the new technology.

 

[ItsNotAboutTheMoney]

Let's say at $125/kWh storage, the solar+storage solution becomes economically advantageous to the traditional grid, with a huge amount of demand there. While others are slowly weening down their cost basis towards a competing product, Tesla would be the *only* supplier able to supply the market at that price, independent of Tesla's cost basis (sub $100/kwh). Of course in the longer term competition will bring the percentage margin much lower, but that could be many years. Assuming equal technology improvements, someone else will have to produce at the same streamlined scale to bring margin pressures.

Well, you can run theoretical numbers and there's a reason why they'd do bulk or sell through solar installers.

Let's consider a 5kW/10kWh system. (Tesla specs C/2 for their static storage).
Capacity: 10kWh * $125/kWh = $1,250
Power: 5kW * $0.10/W = $500 (JB Straubel, Keynote Q&A)
Total: $1,750
Cost: $0.35/W
Life: 10 years
25 year cost: $0.875/W
Even at that price it's not that super cheap, and that price excludes 2.5 x delivery, installation, margin; and then also consider % usable, degredation, charge/discharge loss and utilization. I see most play in areas where per Watt installation costs would be lower, and that's in larger storage for business and utilities.

Under the current retail pricing of electricity, there may be more room for home storage, but I'm sure utilities would act to shift the pricing for people who install solar+battery. But, the good thing is that any pricing shift would favor PEVs and help Tesla. There might be a large battle overcoming the lobbying power of monopoly utilities, although I think that globally there would be some keen markets, enough to build volume, and get other power-hungry businesses onside.

I would argue the realistic market share Tesla can capture with autos has a much lower ceiling (Toyota has < 10% market share and has a huge lineup of products), due to styling and personal preferences.

While Tesla wouldn't necessarily capture a large percentage of auto sales, they can be a drivetrain supplier. If you look at car components instead of cars, there are indeed suppliers with high market share.

 

[SteveG3]

Tesla might be able to sell batteries for storage with a high margin for 5 years or whatever till competition arrives (which obviously will happen relatively quickly if they sell commodities with a margin of 50%) but the profit opportunity of that relatively short period with Tesla low production earmarked storage is low looking at the big picture so it is not something I emphasize in my evaluation of Tesla.

Even if it is no mystery how to produce the batteries, if it is economically impossible to raise the capital (i.e. hundreds of billions of dollars, or even a trillion dollars or more, (see below)) to build up supply to meet demand these batteries will not be a commodity in practice... that is Tesla will continue to command very high margins. of course, if it takes far more modest capital to build supply to meet demand, yes the batteries will drop toward commodity pricing.

For perspective, it's easy to do some back of the napkin that at 100 million vehicles sold per year, and $5 billion to build infrastructure to make batteries for 500K vehicles, it's going to take a trillion in investment for supply to meet demand (once nearly everyone is convinced EVs are superior to ICE). I used the car example because they are round simple numbers and I'm familiar with them, but if demand is large enough for grid storage the same principle will apply. That is, if it takes something like a trillion dollars in capex to build enough infrastructure to build battery supply that meets demand, Tesla will not have its margins squeezed. I think the key question is figuring out the arc of demand for the storage application over the coming decades at the kind of pricing that will be attractive to Tesla and the buyer.

 

[eepic]

eepic, you were the one using 5% of yearly energy consumption in your calculation, you wrote 5% of daily energy usage, but you are clearly using 5% of yearly

Tesla might be able to sell batteries for storage with a high margin for 5 years or whatever till competition arrives (which obviously will happen relatively quickly if they sell commodities with a margin of 50%) but the profit opportunity of that relatively short period with Tesla low production earmarked storage is low looking at the big picture so it is not something I emphasize in my evaluation of Tesla.

30-50% of yearly energy consumption stored as a buffer? That seems quite high to me, I think there will be other solutions for when solar/wind doesn't produce enough like nuclear as you say, or fossils as a backup in the rare cases where shortages happen to keep the storaged buffer at 5% of yearly usage max.

Apologies, typo on my part! The 394,000,000,000 kWh is per day (143,851 TWh/year divided by 365 = 394 TWh/day), no change to the rest of the numbers. But if there is enough demand that production can't even be met at $125/kWh, even if competition can scale up to producing tens to hundreds of GWh a year, they would just sell at the same price rather than undercut. (Again, this is a hypothetical example price for the sake of discussion)

If we take the limit of 100% of world energy production being solar/wind, then I believe we would need quite a bit of storage as a percentage of daily usage (30-50% of daily usage on average). In climates where solar can be less reliable (i.e. not California), there'll be a greater need for storage. But as I alluded to earlier the ideal long term mix would still have some traditional "always-on" grid sources that provide a steady production for base power load 24/7, but arguably the renewables can be the majority.

@ItsNotAboutTheMoney - completely agree that at today's prices the economics for solar+storage does not make sense in an unsubsidized market. My (admittedly optimistic) argument is based on the premise that unsubsidized solar may become more cost efficient than traditional sources at some point, although this is straight from the horse's mouth (below).

"Goal is for unsubsidized solar power to cost less than grid electricity from coal or fracked gas"
Twitter / elonmusk: Goal is for unsubsidized solar ...

 

[Perfectlogic]

Steve, with your math you are implying that the storage industry will have a size of $1T a year, that is insanity. The capex requirements will definately be high in order to meet demand in batteries and will probably push margins up temporarily due to constrained supply, but if this happens then big investments will start to flow in from several companies trying to get a piece of the pie, thinking margins on batteries will be extremely high for decades to come due to supply constraint is unlikely. Just look at how fast the supply caught up in solar despite the industry growing rapidly actually creating an oversupply resulting in paper thin margins and bankrupt companies (I don't actually think this will happen but it's just an example to show how quickly supply can catch up).

- - - Updated - - -

Ah, I see, makes sense now. Well that was positive as I am a happy long as we probably all are here, then I guess you are right in that energy storage could get to $123B or even $200B (consumption will probably rise) in 10-15 years when solar/wind really catches on, which I do believe it will looking at how fast the solar prices are dropping. I still think margins will be around 25-30% at best but it does create a significant opportunity for Tesla, that is very far from priced in at this point.

 

[SteveG3]

Steve, with your math you are implying that the storage industry will have a size of $1T a year, that is insanity.

I think you now see that I was not saying that... I was saying the cost to build the infrastructure (i.e. Gigafactories) to produce enough battery packs to meet upcoming battery storage demand was the key question. if that cost, was in hundreds of billions of dollars, or even a trillion dollars (which for perspective is a back of the envelope estimate of building up enough GFs to convert all vehicle production to long range EVs), we wont see commodity prices for perhaps decades... if the that cost is more modest, i.e. tens of billions of dollars, competition would likely start squeezing margins on a Tesla energy storage business before it ever grew to the size of Tesla's auto business (not that the margins would necessarily be squeezed to the point that Tesla wouldn't want to keep growing the business).

this is why I think the key question is, what do we think the upcoming battery storage demand will be.