Near-future quarterly financial projections

[EVNow]

1.8m makes very little sense to me. They just did 439k in 4Q22...which is an annualized rate of 1.756m. Either they're sandbagging 2023 badly or they're implicitly forecasting Shanghai and/or Fremont will slow down from their 4Q22 rate, because presumably Austin & Berlin will continue ramping this year.

I think it has two sides
- Shanghai / Fremont will cut back expensive production i.e. where earlier they may have streatched at the cost of higher COGS to produce more, they will cut back to make production much more efficient
- Some sandbagging after record deliveries in Q4 was branded a big miss

 

[bkp_duke]

Help me out with back of the napkin math here.

Tesla management is guiding for 1.8m deliveries in 2023.
Tesla made about $9k in net profit per vehicle sold in 4Q22 (also close in 3Q22), depending on how much net income you attribute to Energy & Service.
They just cut prices by $6k per vehicle. Let's say they recognize a cost savings of $1.5k/vehicle in 2023. That's now $4.5k/vehicle in net profit.

Well, 1.8m x $4.5k would equate to $8.1bn in net profit, without any price changes to the lineup. Add in $900m of net profit from energy and that's $9bn.

That's ~$2.60 EPS for FY23 given the guidance from management and the above assumptions of the price cuts.

You completely forget the ~$4000/vehicle sold in the USA that has battery incentives from the IRA, that goes directly to Tesla (not the consumer).

You also assume that Tesla was selling in Q4 cars at list - that is wrong. On the call yesterday they stated that many cars were still being fulfilled at much lower, older prices (2021). So the cut per vehicle is not $6k as you assume, but lower.

Additionally, your assumption for cost savings for vehicle is pretty low. Berlin and Austin are no longer going to be huge drains on capex now that their RAMPS are above 3k/wk.

Finally, input costs (Al, Steel, etc.) and shipping costs are WAY WAY down compared to 2022. I've posted about this before, but commodity prices on everything by Lithium and Nickel are actually below what they were in 2021. These will continue to work their way through the supply chain.

 

[transpondster]

You completely forget the ~$4000/vehicle sold in the USA that has battery incentives from the IRA, that goes directly to Tesla (not the consumer).

You also assume that Tesla was selling in Q4 cars at list - that is wrong. On the call yesterday they stated that many cars were still being fulfilled at much lower, older prices (2021). So the cut per vehicle is not $6k as you assume, but lower.

Additionally, your assumption for cost savings for vehicle is pretty low. Berlin and Austin are no longer going to be huge drains on capex now that their RAMPS are above 3k/wk.

Finally, input costs (Al, Steel, etc.) and shipping costs are WAY WAY down compared to 2022. I've posted about this before, but commodity prices on everything by Lithium and Nickel are actually below what they were in 2021. These will continue to work their way through the supply chain.

We know that Panasonic and Tesla will split subsidies, but don't know ratio... Models S/X only will get pack subsidies and LFP packs 0.
Lithium rpcie is way higher than 2021 and Musk(?) said they expect on avarage 2023 will have higher than 2022, most likely because spot prices were priced in new longterm contracts

 

[bkp_duke]

We know that Panasonic and Tesla will split subsidies, but don't know ratio... Models S/X only will get pack subsidies and LFP packs 0.
Lithium rpcie is way higher than 2021 and Musk(?) said they expect on avarage 2023 will have higher than 2022, most likely because spot prices were priced in new longterm contracts

There is only about 5-10kg of Li in a tesla battery pack. Unlike Nickle, it's not the major constituent (the name Li-Ion comes from the fact the the Lithium is the mobile ion in the cell, not the major constituent).

So Li pricing, while important, is not actually the main driver of pack price. Elon has said that is Nickle, and was one of the main factors for many companies to research using LFP in automotive

 

[Doggydogworld]

There is only about 5-10kg of Li in a tesla battery pack.

I usually see 10 kg quoted. Spot lithium carbonate was $84/kg for a while last year, up from 12-15. LiCO3 is 2/7th lithium, so $84/kg of lithium carbonate is ~300/kg of lithium. That's 3k per pack. I doubt Tesla ever paid 84/kg spot, nor are this year's contracts struck at that price, but there's a reason Musk complained about lithium multiple times and called lithium refineries a license to print money. It's not just a rounding error.

 

[bkp_duke]

I usually see 10 kg quoted. Spot lithium carbonate was $84/kg for a while last year, up from 12-15. LiCO3 is 2/7th lithium, so $84/kg of lithium carbonate is ~300/kg of lithium. That's 3k per pack. I doubt Tesla ever paid 84/kg spot, nor are this year's contracts struck at that price, but there's a reason Musk complained about lithium multiple times and called lithium refineries a license to print money. It's not just a rounding error.

Oh, I agree. It's not inconsequential. But it's still a fraction of the cost of the Nickel.

Obviously we want them all to come down in price.

 

[transpondster]

Oh, I agree. It's not inconsequential. But it's still a fraction of the cost of the Nickel.

Obviously we want them all to come down in price.

except at current spot prices Lithium cost more than Nickel, it doesnt't matter that there is 9 times more Nickel than Lithium in battery if it costs 12 times more.

 

[EVNow]

I usually see 10 kg quoted. Spot lithium carbonate was $84/kg for a while last year, up from 12-15. LiCO3 is 2/7th lithium, so $84/kg of lithium carbonate is ~300/kg of lithium. That's 3k per pack. I doubt Tesla ever paid 84/kg spot, nor are this year's contracts struck at that price, but there's a reason Musk complained about lithium multiple times and called lithium refineries a license to print money. It's not just a rounding error.

If we assume $100/kWh cost, 100 kWh would cost $10k. Assuming 10 kg of Li per pack,

@ $10 -> $350 of Li -> 4% of 10k battery
@ $40 -> $1400 of Li -> 13% of 11k battery
@ $84 -> $2940 of Li -> 23% of 12.5k battery

Because of the long lead time to start new factories to mine & refine Li, we'll have to expect high variability in Li prices, even going forward.

 

[bkp_duke]

what will be the year that tesla sold their bitcoin?
what do you think?

Bulk of the BTC has already been sold. What's left is not really of much consequence. It could go to 0 and the impact on earnings would be pretty small.

 

[petit_bateau]

I figure this is the thread where folk may be more interested in this type of analysis. I'm continuing to look at this and have now reworked the annual exercise using the quarterly data.

I guess this is the sort of thing one ought to be able to get out of a Bloomberg terminal if one had access and some training.

A lot of people who are active in the market operate just on the basis of PE x EPS. Most retail probably fall into that category, quite a few pros as well. Enough that it likely has a significant influence. Perhaps not the dominant influence, but non-trivial.

The red line in the yellow-shaded forecast area is simply driven by a 40x of the current quarter EPS (annualised, i.e. 4 * EPS * 40). A $170 TSLA shareprice corresponds to 40x PE ratio which is why I've used that PE number.

A PE of 40 is not particularly high for a growth share like TSLA.

In the highish case (i.e. the 2.2m in 2023, not the 1.8m guidance) the EPS will roughly treble over the next two years.

If share price follows EPS in that 40x fashion then TSLA is setting up for another significant disconnect between the EPS-driven share price and the NPV-driven valuation. That in turn would suck in all the momentum players. Which might in turn resolve through a subsequent correction.

The key of course is to have a good enough model to generate the EPS series. Hopefully this motivates some of you to persevere with that effort for a while longer, as it could get quite interesting. The next peak into nosebleed teritory might be at $550-$600 in 24-30 months.

(I'm away to think through lower volumes, revenues, and margins next.)

 

[heltok]

What are the current estimates for 2023 deliveries? It seems that Tesla intends to adapt prices to make sure demand matches their production and now we hear that Shanghai is increasing output. So where are we at?

A napkin estimate is to start with Q4*4= 439000*4=1756000. Then add the ramp from Austin and Berlin, add a few semi and cybertrucks, some gains from Shanghai and Freemont. So where does that land? Is there any way to get below 2M?

I know Tesla said 1.8M and 2M during the talk, but that was assuming unlikely events such as a large recession or unexpected disruption.

 

[petit_bateau]

What are the current estimates for 2023 deliveries? It seems that Tesla intends to adapt prices to make sure demand matches their production and now we hear that Shanghai is increasing output. So where are we at?

A napkin estimate is to start with Q4*4= 439000*4=1756000. Then add the ramp from Austin and Berlin, add a few semi and cybertrucks, some gains from Shanghai and Freemont. So where does that land? Is there any way to get below 2M?

I know Tesla said 1.8M and 2M during the talk, but that was assuming unlikely events such as a large recession or unexpected disruption.

I am normally on the optimistic side. I have 2.2m. I am struggling to square that with Tesla's 1.8m guidance and their statement that Shanghai is at capacity for the year.

Best I can think of is that a capacity reduction at Fremont for a substantial site redevelopment might be balanced by the increases at Berlin and Austin during the year.

Also no-one in the know is saying what is going on at Shanghai re expansion plans. Previous indications were that the nearby land that had all the earth piled on it (for dewatering / ground pre-loading) should now be ready for the next phase of expansion construction. But that has not happened, and no regulatory filings have been observed. Which in my mind leads to all sorts of speculation. My best guess is that the additional LFP capacity coming onstream that had been intended to feed that is instead feeding Berlin. And the 4680 machinery that had been intended to go to Berlin has instead been sent to Austin. Which is to an extent driven by slower than hoped 4680 ramp, and to an extent driven by US-IRA. After all, it really is about the cell supply.

These are just my speculations, with very little clear evidence, so lots of missing pieces in the puzzle.

As a result I am struggling to bridge the gap between 1.8m and 2.2m.

 

[Doggydogworld]

I am struggling to square that with Tesla's 1.8m guidance and their statement that Shanghai is at capacity for the year.

When did they say that?

Best I can think of is that a capacity reduction at Fremont for a substantial site redevelopment might be balanced by the increases at Berlin and Austin during the year.

Troy forecasts falling Y output at Fremont, more than offset by rising Austin output. I don't know how much more money they'll throw at Fremont, it's a high cost site and Elon wants to stick it to the "unelected fascists".

Also no-one in the know is saying what is going on at Shanghai re expansion plans. Previous indications were that the nearby land that had all the earth piled on it (for dewatering / ground pre-loading) should now be ready for the next phase of expansion construction. But that has not happened, and no regulatory filings have been observed. Which in my mind leads to all sorts of speculation. My best guess is that the additional LFP capacity coming onstream that had been intended to feed that is instead feeding Berlin. And the 4680 machinery that had been intended to go to Berlin has instead been sent to Austin. Which is to an extent driven by slower than hoped 4680 ramp, and to an extent driven by US-IRA. After all, it really is about the cell supply.

In the call they said they weren't cell-limited. I see no reason to expand Shanghai 3/Y output and they won't make PointyTruck there, so it all comes down to Model 2.

 

[petit_bateau]

I am struggling to square that [my 2.2m forecast] with Tesla's 1.8m guidance and their statement that Shanghai is at capacity for the year.

When did they say that?

25-Jan-2023

"Since our Shanghai factory has been successfully running near full capacity for several months, we do not expect meaningful sequential volume increases in the near term. "

Tesla Investor Relations

Tesla's mission is to accelerate the world's transition to sustainable energy. Today, Tesla builds not only all-electric vehicles but also infinitely scalable clean energy generation and storage products.

ir.tesla.com

I genuinely am struggling to find plausible reasons to close the gaps between 1.8m TSLA guidance and the 2.2/2.3m that I think ought to be achievable and necessary this year.

 

[Doggydogworld]

"Since our Shanghai factory has been successfully running near full capacity for several months, we do not expect meaningful sequential volume increases in the near term. "

Thanks. Shanghai produced 233k cars in Q4, so that implies a bit over 900k this year. 500k for Fremont and 200k each for Berlin and Austin gives 1.8m.

 

[heltok]

500k for freemont seems like a very low number. In q3 they were at 12k/week aiming for 14k/week with a stated capacity for 650k/year after having sent the shanghai team there to optimize things. S/X is still not available in many countries so doubt they will need to slow down those. Y maybe getting saturated at some point, not for a while with these prices decreases.

 

[bkp_duke]

500k for freemont seems like a very low number. In q3 they were at 12k/week aiming for 14k/week with a stated capacity for 650k/year after having sent the shanghai team there to optimize things. S/X is still not available in many countries so doubt they will need to slow down those. Y maybe getting saturated at some point, not for a while with these prices decreases.

Rumor is one of the 3 lines is down for Highland upgrades. This may be a consistent theme in 2023.

 

[DarkandStormy]

500k for freemont seems like a very low number. In q3 they were at 12k/week aiming for 14k/week with a stated capacity for 650k/year after having sent the shanghai team there to optimize things. S/X is still not available in many countries so doubt they will need to slow down those. Y maybe getting saturated at some point, not for a while with these prices decreases.

Stated capacity has never equaled actual production.

Troy is at 518k for Fremont for FY23, fwiw.

 

[petit_bateau]

Stated capacity has never equaled actual production.

Troy is at 518k for Fremont for FY23, fwiw.

I think Fremont + Sparks did 548k in 2022, so Troy's 518k would represent a reduction.

Methinks there is likely some fire amidst the smoky rumours of a rebuild at Fremont.

 

[DarkandStormy]

https://twitter.com/x/status/1604579924903223296

This tweet about hitting 3k/week run rate...coupled with the below January numbers...

doesn't make a ton of sense to me. Berlin didn't pause until January 30th for upgrades, but doing only 9k deliveries in Europe? Berlin entered 2023 at a 3k/week run rate (or greater) plus I was told there were a record number of ships (for being so early in the quarter) from Shanghai arriving. Berlin alone should have been able to provide 12k vehicles for the month, no?