Buckminster
Well-Known Member
Do we know how many GWh Lathrop and Nevada is doing?
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Value Of The Megapack Business
Hi, all --I'm trying to eyeball the value of the Megapack business. I've looked at variety of forecast for BESS over the next decade or so, all of which look something like this recent McKinsey piece:
Enabling renewable energy with battery energy storage systems
The market for battery energy storage systems is growing rapidly. Here are the key questions for those who want to lead the way.
Here's the money graph:
Roughly, volume quintuples while unit price halves.
What follows is *not* a forecast, it's a math exercise with very round #s.
The midpoint of 2030 market size is $135B. Say Tesla gets 10% of that with a 10% *net* margin. That's $1.35B net. Divide by 3.5B shares = $0.38 EPS. Slap a 30x multiple (2030 rev. growth is 12%; 30x might be too high or too low) on that and you've $11.4/share in value in 2030. Cut that in half to discount back to today, and the megapack business worth about $5.70/share today, or $40B market cap in 2030.
Again, this math exercise, not a forecast. Anyone want to argue for a much higher/lower market share or margin? Could anyone share research supporting a much higher/lower market size? NB, I have not read the entire thread, just the last 5 pages or so.
Yours,
RP
Last edited:
SageBrush
REJECT Fascism
Funny, I was just looking at the Tesla Q3/2023 financial 'stack' this morning. Tesla reported ~ 33% gross profit on storage (1.5E9 revenue, 1E9 cost)
Hi, SageBrush --
I get: (1,559 - 1,178) / 1,559 = 24.4%. I'm interested in net margins to enable an EPS calculation. If you want to assume GM stays stable at, say, 25%, then, using Tesla's Q2 & Q3 as a guide (in both periods GM was 25%), you'd wind with ~14.5% net. There's no reason to assume Megapacks would have the same economics as Tesla as a whole, it's just a convenient example.
Yours,
RP
I get: (1,559 - 1,178) / 1,559 = 24.4%. I'm interested in net margins to enable an EPS calculation. If you want to assume GM stays stable at, say, 25%, then, using Tesla's Q2 & Q3 as a guide (in both periods GM was 25%), you'd wind with ~14.5% net. There's no reason to assume Megapacks would have the same economics as Tesla as a whole, it's just a convenient example.
Yours,
RP
SageBrush
REJECT Fascism
I understand. I'm more inclined to ask how much Tesla battery production for storage might be available by 2030, and then assume a 20% margin at half the current price. This is similar to your exercise, but it puts more emphasis on the idea that Tesla can sell however much it can scale (the 4680, dry electrode process.)
During battery day Tesla talked about reaching 3 TWh of battery production and 20 million annual car production by 2030. If a car is a 75 kWh pack then 1.5 TWh goes to cars and 1.5 TWh goes to storage. If we presume Tesla sells storage at a future cost of $200 per kWh and gross profit is 20% then Tesla profit is $40 per kWh ($40 per 1E3 Wh) = $0.04 per Wh. That works out to $0.06E12, or $6E10 = $60B in 2030 dollars.
That is the Elon scenario, so take it as best case, probable but not in the timeframe mentioned
Any errors in my arithmetic ?
Hi, SageBrush --
> Any errors in my arithmetic ?
No, but,
> During battery day Tesla talked about reaching 3 TWh of battery production and 20 million annual car production by 2030. If a car is a 75 kWh pack then 1.5 TWh goes to cars and 1.5 TWh goes to storage.
is what I'd call a "much larger market" scenario, where Tesla has 100% of a market that is 2x-3x bigger than McKinsey's forecast of total 2030 volume. I don't think Tesla would have any problems *making* 1.5TWh if they could get the batteries, capital costs here seem to be low; *selling* 1.5TWh on the other hand ...
I'm not wedded to McKinsey's #s. I'd be interested to see independent research that's supportive of a larger market than they're calling for. But, they're ballparky around what I've seen from the others, and given the lags involved in these projects, should have a decent handle on at least the next couple of years. I'm also not wedded to the #s in my arithmetic exercise either; if you double all of them (market size, market share, margins, PE), you'd get $80/share. But I'm having a hard time getting to, "Megapacks worth as much as the auto business," unless the auto business is worth a *lot* less than the last quote.
Minor quibble with your arithmetic: 20% gross margins should turn into something like 12% net.
Yours,
RP
> Any errors in my arithmetic ?
No, but,
> During battery day Tesla talked about reaching 3 TWh of battery production and 20 million annual car production by 2030. If a car is a 75 kWh pack then 1.5 TWh goes to cars and 1.5 TWh goes to storage.
is what I'd call a "much larger market" scenario, where Tesla has 100% of a market that is 2x-3x bigger than McKinsey's forecast of total 2030 volume. I don't think Tesla would have any problems *making* 1.5TWh if they could get the batteries, capital costs here seem to be low; *selling* 1.5TWh on the other hand ...
I'm not wedded to McKinsey's #s. I'd be interested to see independent research that's supportive of a larger market than they're calling for. But, they're ballparky around what I've seen from the others, and given the lags involved in these projects, should have a decent handle on at least the next couple of years. I'm also not wedded to the #s in my arithmetic exercise either; if you double all of them (market size, market share, margins, PE), you'd get $80/share. But I'm having a hard time getting to, "Megapacks worth as much as the auto business," unless the auto business is worth a *lot* less than the last quote.
Minor quibble with your arithmetic: 20% gross margins should turn into something like 12% net.
Yours,
RP
SageBrush
REJECT Fascism
I estimate that gross profit on a Tesla car is somewhere in the range of $7,500 - $10,000. Tesla 'invested' ~ 75 kWh, which works out to 10¢ per Wh on the low end. This is 2.5x the gross profit margin I estimated for a large scale battery storage business. This reinforces the idea that Tesla will first saturate the EV market, and then divert the unused battery fraction to storage. Both businesses are huge profit machines in their own right, but cars are the crown jewel
My honest opinion is that the storage business' unsung value is in letting Tesla scale battery production as fast as possible, unencumbered by the need to hedge car macroeconomic trends and cycles. Tesla profits will rise and fall, but I don't expect to see unsold batteries filling up warehouses.
Megapacks currently use LFP, likely in a different form factor than vehicles, from 3rd party suppliers. That's outside the cell production of Tesla itself, though they could start in-sourcing LFP production.
Powerwall 3 is also LFP.
Hi, all --
I'm hoping to stimulate further conversation on this. The value of the Megapack business (or anything, really!) is:
(At terminal date, market size * market share * margin * multiple) * discount factor. Note that "multiple" is intended to capture forward expectations at the terminal date. The numbers I threw out for the arithmetic exercise were:
$135B Market size--mid-point of McKinsey's estimates.
10% Market share -- Round #, Tesla is currently somewhere around there.
10% *Net* Margin -- Round #, Tesla is currently somewhere around there.
30x Terminal Multiple -- Round #, a reasonable growth but not hyper-growth multiple.
50% Discount -- Round #, roughly a 10% annualized return.
, which after some tedious multiplication gets you around $6/share today. If you double everything except for discount factor, you get $96 (I was rounding down to $5 earlier when I said $80).
Let me add some commentary and personal opinion to each of those #s.
1: Market size. I have nothing intelligent to say about this. I'm nowhere close to being able to come up with an independent estimate, I'm completely reliant on industry forecasters. I do not have access to proprietary research. I do an occasional Google binge to see if there's anything new in the public domain. If anyone has anything to share, I'd be grateful.
My gut feeling is that the market is more likely to be 2x McKinsey than 0.5x. Note, though, that a major source of demand is intermittent power, ie, solar and wind, and last time I looked S&W stocks were in the toilet.
2: Market share. Getting market share data is also really difficult. My current sense is that Tesla is a major player, with say 10%-15% share. My guess is that they're more likely to lose than gain share. This is because:
A: I expect the market to be fiercely competitive. Capital costs are low and manufacturing complexity minimal. To get a sense of capital costs, check out this Google map of the Lathrop plant:
www.google.com
Just tool around, zoom in and out. This modest sized building in an undistinguished industrial area has 40GWh capacity, or about 1/3 of McKinsey's 2023 global estimates.
Manufacturing complexity is: You're shoving boxes into a box and attaching some electronics. Note the implications for capital costs.
There's a very important software component. I just don't think it's all that hard. (As a side note, my cousin used to work for Hewlett Packard on the software for the printer head of their ink-jet printers (remember those?) He was not a genius.) This stuff is tricky and specialized, but well within "team of competent and experienced engineers" territory.
Finally, I don't see why battery manufacturers don't become major competitors. BYD, CATL and LG already are. I can see why a battery manufacturer wouldn't want to directly compete with customers in automobiles (it's rumored that BYD is looking at spinning its battery division off for just this reason), but don't get the same vibe from BESS. It's possible that Tesla becomes a major battery manufacturer, but that's a separate conversation.
Geopolitics is a major wild-card. I could see Tesla winding up with a much larger share of a smaller market.
3: Margins. I think 5% more likely than 20% -- remember, these are net, not gross -- due to the market share discussion above. Honestly, I think 20% net pretty much impossible. Like, no f*cking way possible. Note the geopolitics caveat.
4: Multiple. No sense spending much time discussing this, it's a guess as to what people 7 years from now will think the next 7 years looks like. FWIW, I find 15x more likely than 60x.
5: Discount Factor. I feel no desire to fiddle with this.
To sum up, relative to *very round numbers* I threw out for the arithmetic exercise, I get:
Market Size: +
Market Share: = (Note I think this implies some share loss, I think Tesla is > 10% now.)
Margins: - (Again, I think this implies a decline from current levels.)
Multiple: = , but who knows.
Discount Factor: =
, with geopolitics being a huge swing factor that I'm not even going to try to handicap.
Yours,
RP
I'm hoping to stimulate further conversation on this. The value of the Megapack business (or anything, really!) is:
(At terminal date, market size * market share * margin * multiple) * discount factor. Note that "multiple" is intended to capture forward expectations at the terminal date. The numbers I threw out for the arithmetic exercise were:
$135B Market size--mid-point of McKinsey's estimates.
10% Market share -- Round #, Tesla is currently somewhere around there.
10% *Net* Margin -- Round #, Tesla is currently somewhere around there.
30x Terminal Multiple -- Round #, a reasonable growth but not hyper-growth multiple.
50% Discount -- Round #, roughly a 10% annualized return.
, which after some tedious multiplication gets you around $6/share today. If you double everything except for discount factor, you get $96 (I was rounding down to $5 earlier when I said $80).
Let me add some commentary and personal opinion to each of those #s.
1: Market size. I have nothing intelligent to say about this. I'm nowhere close to being able to come up with an independent estimate, I'm completely reliant on industry forecasters. I do not have access to proprietary research. I do an occasional Google binge to see if there's anything new in the public domain. If anyone has anything to share, I'd be grateful.
My gut feeling is that the market is more likely to be 2x McKinsey than 0.5x. Note, though, that a major source of demand is intermittent power, ie, solar and wind, and last time I looked S&W stocks were in the toilet.
2: Market share. Getting market share data is also really difficult. My current sense is that Tesla is a major player, with say 10%-15% share. My guess is that they're more likely to lose than gain share. This is because:
A: I expect the market to be fiercely competitive. Capital costs are low and manufacturing complexity minimal. To get a sense of capital costs, check out this Google map of the Lathrop plant:
Google Maps
Find local businesses, view maps and get driving directions in Google Maps.
www.google.com
Just tool around, zoom in and out. This modest sized building in an undistinguished industrial area has 40GWh capacity, or about 1/3 of McKinsey's 2023 global estimates.
Manufacturing complexity is: You're shoving boxes into a box and attaching some electronics. Note the implications for capital costs.
There's a very important software component. I just don't think it's all that hard. (As a side note, my cousin used to work for Hewlett Packard on the software for the printer head of their ink-jet printers (remember those?) He was not a genius.) This stuff is tricky and specialized, but well within "team of competent and experienced engineers" territory.
Finally, I don't see why battery manufacturers don't become major competitors. BYD, CATL and LG already are. I can see why a battery manufacturer wouldn't want to directly compete with customers in automobiles (it's rumored that BYD is looking at spinning its battery division off for just this reason), but don't get the same vibe from BESS. It's possible that Tesla becomes a major battery manufacturer, but that's a separate conversation.
Geopolitics is a major wild-card. I could see Tesla winding up with a much larger share of a smaller market.
3: Margins. I think 5% more likely than 20% -- remember, these are net, not gross -- due to the market share discussion above. Honestly, I think 20% net pretty much impossible. Like, no f*cking way possible. Note the geopolitics caveat.
4: Multiple. No sense spending much time discussing this, it's a guess as to what people 7 years from now will think the next 7 years looks like. FWIW, I find 15x more likely than 60x.
5: Discount Factor. I feel no desire to fiddle with this.
To sum up, relative to *very round numbers* I threw out for the arithmetic exercise, I get:
Market Size: +
Market Share: = (Note I think this implies some share loss, I think Tesla is > 10% now.)
Margins: - (Again, I think this implies a decline from current levels.)
Multiple: = , but who knows.
Discount Factor: =
, with geopolitics being a huge swing factor that I'm not even going to try to handicap.
Yours,
RP
Hi, all --
FWIW, McKinsey's volume forecasts are on the high end; I went with them because they have $ market size estimate. Here are some others that center around 100GW (I suspect GW is the right metric here, not gWH)/400 GWh:
www.rystadenergy.com
"According to the research by Bloomberg, the global installed energy storage capacity additions are expected to hit a record in 2023, with 42GW/99GWh. And is expected to grow at a CAGR of 27% through 2030, with annual additions of 110GW/372GWh in 2030, which is 2.6 times the expected figure for 2023."
www.linkedin.com
Hmm, I thought I had more of these, but maybe not.
On the other hand, BESS manufacturing (and thus barriers to entry) might be more complicated than I thought, based on this piece about what can go wrong:
www.cea3.com
Yours,
RP
FWIW, McKinsey's volume forecasts are on the high end; I went with them because they have $ market size estimate. Here are some others that center around 100GW (I suspect GW is the right metric here, not gWH)/400 GWh:
New battery storage capacity to surpass 400 GWh per year by 2030
The era of battery energy storage applications may just be beginning, but annual capacity additions will snowball in the coming years as storage becomes crucial to the world’s energy landscape.
"According to the research by Bloomberg, the global installed energy storage capacity additions are expected to hit a record in 2023, with 42GW/99GWh. And is expected to grow at a CAGR of 27% through 2030, with annual additions of 110GW/372GWh in 2030, which is 2.6 times the expected figure for 2023."
Market Insight: Global Energy Storage Market Outlook to 2030
According to the research by Bloomberg, the global installed energy storage capacity additions are expected to hit a record in 2023, with 42GW/99GWh. And is expected to grow at a CAGR of 27% through 2030, with annual additions of 110GW/372GWh in 2030, which is 2.
www.linkedin.com
Hmm, I thought I had more of these, but maybe not.
On the other hand, BESS manufacturing (and thus barriers to entry) might be more complicated than I thought, based on this piece about what can go wrong:
Maximizing BESS Performance: Insights and Quality Assurance Strategies from CEA's Experience — Clean Energy Associates
Explore the evolution and challenges in battery energy storage systems (BESS) with Chi Zhang and George Touloupas of Clean Energy Associates. Learn about common manufacturing defects, the shift in battery chemistries, and the importance of rigorous quality assurance in ensuring safe, efficient, and
www.cea3.com
Yours,
RP
Buckminster
Well-Known Member
damn, I got rid of my Smart things and got Google Nest instead
jhm
Well-Known Member
On market size: Conservatively, getting to net zero emissions will require 4 hours of battery storage. Many people will argue the need is much greater from 8 hours to 90 day! But generally those who are most opposed to renewable energy want people to believe the battery requirement is super high. So let's go with 4 hour as a pretty confident lower bound.
Global electricity generation in 2022 was 3.55TW. So without factoring in growth in power demand, the grids of the world could easily use some 14 TWh of batteries.
Allowing demand to grow about 3.5%/y to 2030 brings avg generation to 4.7TW and 4 hours of batteries to 18.7 TWh.
Basically, if Tesla and other batterymakers can produce 1TWh in 2030, this comes nowhere close to saturating the market.
Global electricity generation in 2022 was 3.55TW. So without factoring in growth in power demand, the grids of the world could easily use some 14 TWh of batteries.
Allowing demand to grow about 3.5%/y to 2030 brings avg generation to 4.7TW and 4 hours of batteries to 18.7 TWh.
Basically, if Tesla and other batterymakers can produce 1TWh in 2030, this comes nowhere close to saturating the market.
Last edited:
UkNorthampton
TSLA - 12+ startups in 1
jhm
Well-Known Member
UkNorthampton
TSLA - 12+ startups in 1
jhm
Well-Known Member
Total genaration in 2022 was 28,661TWh. 28,661TWh/365/24h = 3.27 TW.
Electricity Mix
Explore data on where our electricity comes from, and how this is changing.
ourworldindata.org
Somehow, I had misremembered the 3.55TW figure.
SageBrush
REJECT Fascism
'Net' zero is too fuzzy a notion to be useful to me.
I'm much more comfortable with either total annual emissions (what actually matters), or reduction from some arbitrary baseline.
My WAG is that for non-extreme climates, where local complementary clean resources have been well developed, 4 hour grid storage covers 95% - 99% of annual usage with clean energy. That is nothing to sneeze at, since it can get done in the next 10 years.
I used to think that HVDC long transmission was the answer for the remainder -- and I'm still pretty sure that it has a role to play. But lately I'm enamored with the potential of of EVs providing both a sink and source for clean energy production and storage. For residential that is on the order of 100 - 150 kWh per household, so about 125E3 * 150E6 Wh = 20 TWh. An almost universal EV transition pretty much replaces the utility 4 hour battery storage.
Last edited:
SageBrush
REJECT Fascism
Pardon me for quoting myself -- I want to expand on Residential EV storage
My home has PV, and we have two EVs with a combined 140 kWh of batteries. We bought the Tesla to have a long drive option for which the Chevy Bolt is not a great option. But for 95+ percent of our days, we drive about 60 miles a day combined. Optimal use of battery could easily cover that with 20 kWh, implying that 85% of our household battery is unused for 310 days of the year. It is that vastly under-utilized resource that can serve grid clean energy storage.
Frankly, I find it hard to imagine a reasonable scenario where EV storage is not a linchpin of the clean energy economy.
UkNorthampton
TSLA - 12+ startups in 1
Time of Use pricing is key. Automated charging/other use based on pricing better still.
UK: price guarantee recently risen to 28 pence per kWh (most people without an EV pay this). Overnight pricing, maybe 5p/kWh, heavy EV users might have.
We are light EV users at present as work from home etc.
My pricing is look ahead whole day according to a published formula. So I look to see what today's and tomorrow's prices are.
If cheap (down to 10p over Xmas), EVs charged, washing/drying plus dehumidifier all on. Goes above 20p and avoid charging and clothes washing.
Expanding number if people doing this, using solar, batteries at every level, infra red/storage heat, heatpumps in addition to gas boilers (use cheapest energy) would dent the peak aspect, especially if commercial usage adjusts.
Switching from gas to electric heating when electricity prices are low would also reduce gas spot price, gas-generated electricity prices would lower, more electricity use rather than gas, gas peaker plants less viable - a virtuous circle towards electrification.
Office buildings in UK are awful. If working weekends with heating off, huge difference due to low thermal mass and bad insulation.
Last edited:
Similar threads
- Article
- Article
