Near-future quarterly financial projections

[JohnnyEnglish]

So inventory in Europe is gonna go up a lot in Q1:

https://twitter.com/x/status/1632036662380249090

It is always interesting to see Troy's estimates but they are only one data point to be taken into account when gazing into our crystal balls. It is clear that his estimates are at odds with the strategy stated by Tesla on several occasions and reiterated by @Dreadnought above: Tesla will ramp as quickly as they can (whilst not paying over the odds to solve logistics problems) and adjust pricing to stimulate demand when necessary. So far it seems to me that the hard data that we are seeing in terms of price adjustments and production figures seem to indicate that Tesla is indeed walking the talk.

I am looking forward to seeing the CPCA February production figures which hopefully will be released next week (Rob Maurer generally comments on these in his podcast). If they are close to the 74k retail figures that will be another indication that Shanghai is being operated close to production capacity (allowing for known model 3 line downtime), Fremont is at record production (Investor day) and Berlin is continuing to ramp. I have not seen any indication that Austin is not ramping as planned.

Unfortunately hard data is infrequent and between times it is not possible to refute hypotheses based on softer indicators implying that Tesla is somehow not operating in line with their stated strategy.

 

[EVNow]

Tesla will ramp as quickly as they can (whilst not paying over the odds to solve logistics problems) and adjust pricing to stimulate demand when necessary. So far it seems to me that the hard data that we are seeing in terms of price adjustments and production figures seem to indicate that Tesla is indeed walking the talk.

Zach has gone on record with his over 20% GM comment. Thats the constraint they have to work with. We saw last quarter they had left over inventory and only in Jan they reduced prices. They might do it again this time in EU ... making sure GM is above guidance, even if inventory builds up a bit (which can be blamed on unwinding).

BTW, according to what Rob said - Tesla is down 19% from last quarter looking at insurance data for Jan & Feb. BYD is down more - and overall China auto market is down 20%. Looks like China had big Covid wave, reopening & recession - all in quick time

 

[petit_bateau]

I was looking for those takt times and then @Singuy gives them

- but Fremont missing
- 3 missing
- is it per line, or per factory

 

[Doggydogworld]

BTW, according to what Rob said - Tesla is down 19% from last quarter looking at insurance data for Jan & Feb. BYD is down more - and overall China auto market is down 20%. Looks like China had big Covid wave, reopening & recession - all in quick time

Tesla's insurance regs grew strongly again last week, to 13,266. That's more than double a couple weeks ago. I can't explain the wild swings. The uptick pushed Feb retail sales to 33,993 (or maybe 33,923, still checking) vs. my 32k estimate from last week.

Tesla China Jan+Feb retail sales were 60.8k. That's down 24% from Oct+Nov, but up 43% y/y.

I use rolling 3 month numbers for Tesla, to eliminate shipping cadence weirdness. Rolling 3 month retail sales are 102.8k, down 10% Y/Y. December's higher prices seem to be the main culprit, so I think Q1 will end up showing a little y/y growth. The picture brightened considerably the past two weeks.

Rolling 3 month exports were 93.6k, up 26% y/y. They added some smaller markets like Aus/NZ, but it still seems the vast majority went to Europe. That plus a ramping Berlin is a lot to absorb. Maybe the UK will come to the rescue, March is a huge month there due to a bizarre registration scheme.

 

[Thekiwi]

Tesla's insurance regs grew strongly again last week, to 13,266. That's more than double a couple weeks ago. I can't explain the wild swings. The uptick pushed Feb retail sales to 33,993 (or maybe 33,923, still checking) vs. my 32k estimate from last week.

Tesla China Jan+Feb retail sales were 60.8k. That's down 24% from Oct+Nov, but up 43% y/y.

I use rolling 3 month numbers for Tesla, to eliminate shipping cadence weirdness. Rolling 3 month retail sales are 102.8k, down 10% Y/Y. December's higher prices seem to be the main culprit, so I think Q1 will end up showing a little y/y growth. The picture brightened considerably the past two weeks.

Rolling 3 month exports were 93.6k, up 26% y/y. They added some smaller markets like Aus/NZ, but it still seems the vast majority went to Europe. That plus a ramping Berlin is a lot to absorb. Maybe the UK will come to the rescue, March is a huge month there due to a bizarre registration scheme.

Zach has gone on record with his over 20% GM comment. Thats the constraint they have to work with. We saw last quarter they had left over inventory and only in Jan they reduced prices. They might do it again this time in EU ... making sure GM is above guidance, even if inventory builds up a bit (which can be blamed on unwinding).

BTW, according to what Rob said - Tesla is down 19% from last quarter looking at insurance data for Jan & Feb. BYD is down more - and overall China auto market is down 20%. Looks like China had big Covid wave, reopening & recession - all in quick time

As Rob pointed out:

Last quarter in China Tesla had factory shutdown in December, whereas this quarter the shutdowns were in the 2 months already passed. So of course production will be down vs the first 2 months of this quarter, but the 3rd month will be significantly higher production than happened in 3rd month of last quarter.

 

[JohnnyEnglish]

<...>
I am looking forward to seeing the CPCA February production figures which hopefully will be released next week (Rob Maurer generally comments on these in his podcast). If they are close to the 74k retail figures that will be another indication that Shanghai is being operated close to production capacity (allowing for known model 3 line downtime), Fremont is at record production (Investor day) and Berlin is continuing to ramp. I have not seen any indication that Austin is not ramping as planned.

Unfortunately hard data is infrequent and between times it is not possible to refute hypotheses based on softer indicators implying that Tesla is somehow not operating in line with their stated strategy.

Rob Maurer gave the February MIC production as 72,332 with 51,839 Model Y and 20,493 Model 3. This gives weekly production rates of 12,960 for the Y and 6750 for the 3 given the week of downtime for line upgrades. This confirms these lines are effectively running at capacity.

Zach has gone on record with his over 20% GM comment. Thats the constraint they have to work with. We saw last quarter they had left over inventory and only in Jan they reduced prices. They might do it again this time in EU ... making sure GM is above guidance, even if inventory builds up a bit (which can be blamed on unwinding).

BTW, according to what Rob said - Tesla is down 19% from last quarter looking at insurance data for Jan & Feb. BYD is down more - and overall China auto market is down 20%. Looks like China had big Covid wave, reopening & recession - all in quick time

As stated in my previous post it is not possible to refute/prove these hypotheses until we have some further hard data, in this case the Q1 P&D figures. Only 3 weeks to wait.

 

[EVNow]

As stated in my previous post it is not possible to refute/prove these hypotheses until we have some further hard data, in this case the Q1 P&D figures. Only 3 weeks to wait.

I’ve no interest in refuting or proving hypothesis - only in figuring out the possibilities and risks. Not recognizing the risks is the primary reason for losing a lot of money for many people last year.

 

[petit_bateau]

FWIW, courtesy @Buckminster and @Troy1466 (?)

or from

https://twitter.com/x/status/1635688449188540416

 

[Doggydogworld]

FWIW, courtesy @Buckminster and @Troy1466 (?)

A month ago Troy estimated 430k for Q1. A couple weeks ago he dropped China from 129k to 110k and dropped the overall number to 412k. Since then he's kept the total in the 410k area but bumped China back up to the near-130k range and dropped USA (mostly Austin) to compensate.

Tesla China's insurance registrations have been all over the map this quarter. Near-zero before the price cut, then a jump to 13k/week after the cut. Then a gradual fade to a horrible sub-6k by mid-February. Then, just as Q1 was looking pretty grim, deliveries took off, notching a terrific 17k last week. If they maintain 17k/week they'll be around 135k for Q1. If they keep ramping into EOQ they'll do 140k+. That's way above Troy's estimate.

I don't understand the wild swings the past six weeks. They had inventory. Wait times never really moved the way they used to during the big "wave" quarters. Other OEMs (e.g. BYD) have been pretty steady week-to-week, though Nio took a hit lately due to holding the line on prices while everyone else cut. Some companies are doing backdoor promotions -- e.g. Li Auto now offers 90 day Price Protection (if they cut within 90 days of your purchase they'll refund you the difference). Maybe Tesla has something going on at the store level that's motivating people to pull the trigger?

 

[Singuy]

I was looking for those takt times and then @Singuy gives them

- but Fremont missing
- 3 missing
- is it per line, or per factory

View attachment 915445

It's total Y production per factory.

 

[heltok]

https://twitter.com/x/status/1635927349337235457

Again Troy stands firm in his belief that Austin>Berlin. Even though Berlin has had a confirmed higher weekly production rate at the start and at the end, but one week downtime inbetween. Imo the downtime should be ~10% less cars, but their weekly rate has been more than that higher. This is not deliveries this is production. So clearly he believes that Berlin has had lots of downtime due to no demand. Let's see how it goes!

 

[petit_bateau]

I too don't understand how Troy thinks Austin > Berlin. It is possible that is correct, but it doesn't align well with the various pieces of info that surface (e.g. reaching the '000 unit milestones from each factory). It is quite plausible that in due course Austin will pass Berlin if cell supply limitations affect Berlin as clearly Austin is getting the 4680 love. To what extent do we think that Troy has better information sources than are available to each/any of us that allow Troy to stand up his views ?

Regarding the China numbers for the other OEMs is it possible that the wild insurance swings are evidence of channel stuffing behaviour. The rumour is that BYD has been selling to dealers to artificially inflate numbers, but I presume dealers would not insure. So as dealers then shift inventory onwards this might cause fluctuations. I appreciate that would not (or should not) affect Tesla's numbers, so that would only be a partial hypothesis.

If I leave my Q1 2023 model on autopilot I have the following table :

Comments I have are :
- My Austin does not yet reflect the shutdown & supposed factory mods they had, so Troy's number looks more plausible than mine, and I will likely adjust down substantially.
- My Shanghai, ditto inc CNY effects.
- My Berlin : it is plausible that some Q1 decelleration of ramp, but the observations in the wild don't seem to justify the dialdown that Troy has.
- My Austin : here I am low vs Troy. I get the feeling that he has private sources ....
- Overall 4 x 485k = 1.94m which a) presupposes no further change quarterly but b) is fairly close to Musk's blurt target of 2m. My full year is at 2,070k so if I do dial down my Q1 a tadge then that pulls back to Musk's blurt.
- If one were to add in a certain amount of wave unwinding then sales could come down a smidge further. I've held constant at 13d inventory so that means that inventory climbs from 70k to 84k, i.e. 14k extra flows into stock on autopilot. Any further unwinding would naturally increase that.
- For solar & storage I live in hope !

We are really lacking good factory and line and model level info sources to keep track of this. It would be nice to see more transparency in this respect from Tesla.

Add some dialdowns and wave unwinds and one could see a pretty significant decelleration in EPS growth in Q1. Here is my autopilot table for 9% growth to $1.27 (GAAP) but I am ordinarily 10% high

And that all comes a fully rational end Q1 shareprice of $182 from an NPV model. But of course none of this is remotely rational so the current shareprice being $183 is purely co-incidence.

 

[Troy]

Giga Berlin normally operates 5 days a week but rarely, they have half-day Saturday shifts. Naturally, they break the weekly production record during those long weeks and that's what Tesla tweets. What they don't mention is that they go back to 5 workdays afterward.

 

[petit_bateau]

Giga Berlin normally operates 5 days a week but rarely, they have half-day Saturday shifts. Naturally, they break the weekly production record during those long weeks and that's what Tesla tweets. What they don't mention is that they go back to 5 workdays afterward.

Thanks Troy, that makes sense.

 

[Singuy]

Just for context, a flat QOQ or slightly down is a NORMAL Q1.

From the last 6 years, 3/6 Q1 was up by less than 3% compared to Q4, 2/6 Q1 was down substantially, and only 1/6 when Tesla delivered like 20k cars from 2016->2017 Tesla beat by over 10%.

So the expectation should be a flat QOQ or slightly down. Even according to Tesla's guide, you shouldn't expect Q1 to be 450k or whatever. Conservatively we should expect something like 410k+440+480+500 to hit 1.8M.

 

[JohnnyEnglish]

Giga Berlin normally operates 5 days a week but rarely, they have half-day Saturday shifts. Naturally, they break the weekly production record during those long weeks and that's what Tesla tweets. What they don't mention is that they go back to 5 workdays afterward.

Just wanting to check I have understood your info:

Normal operation is 5 days at 3 shifts a day = 15 shifts
Long week is 5 days at 3 shifts a day plus half shift on Saturday = 15.5 shifts

Have I got this right?

 

[Dikkie Dik]

Just wanting to check I have understood your info:

Normal operation is 5 days at 3 shifts a day = 15 shifts
Long week is 5 days at 3 shifts a day plus half shift on Saturday = 15.5 shifts

Have I got this right?

I think half a day of shifts is 1.5 shifts.

So 16.5 shifts for a long week.

 

[ByeByeJohnny]

I think half a day of shifts is 1.5 shifts.

So 16.5 shifts for a long week.

I don't think that is certain.

Maybe Troy knows, maybe he's guessing on the hours, but I doubt they would get more than one shift to come in on a Saturday. Seems very unlikely that would be a 12-hour shift so more likely 8 or maybe 10 hours.

More importantly though. If they are now at 15 shifts a week, which would then have started in the last couple of months, would they have already also needed an extra Saturday shift at all in the time since? I doubt that very much. If they have, that would mean the demand is incredibly strong.

My estimate is that they have never done more than 15 shifts. So there are no 16 shift weeks. In which case 'regular' workweeks would be equal to best production weeks.

If they are still at 10 shifts + occasional 1 Saturday shift. That extra shift would only add around 10% more than a regular five day week. As soon as the third shift actually starts (again, if it hasn't already) that record production would almost immediately be beaten anyway.

So any Saturday bump included in some 'record week number' would only be valid very short term.

 

[Doggydogworld]

Tesla ramped China sales again last week, to 18.7k based on insurance registrations. Holding that rate the final 12 days of Q1 would put them at 137k, well above Q4's 121.6k. Further acceleration into EOQ could push them to 140k. That's still less than 30% y/y growth, but the sequential growth shows the price cuts are definitely working.

The price cuts are also pressuring BYD, who finally reacted (a little) in late February and is now curtailing production. Nio and XPeng sales are down significantly, though Li Auto is doing well.

Troy has been bringing his China numbers up. I think he might still be a bit low in Europe, too. A lot depends on the UK, which has a ridiculous sales surge every March due to a registration fee quirk.

 

[mongo]

Tesla ramped China sales again last week, to 18.7k based on insurance registrations. Holding that rate the final 12 days of Q1 would put them at 137k, well above Q4's 121.6k. Further acceleration into EOQ could push them to 140k. That's still less than 30% y/y growth, but the sequential growth shows the price cuts are definitely working.

The price cuts are also pressuring BYD, who finally reacted (a little) in late February and is now curtailing production. Nio and XPeng sales are down significantly, though Li Auto is doing well.

Troy has been bringing his China numbers up. I think he might still be a bit low in Europe, too. A lot depends on the UK, which has a ridiculous sales surge every March due to a registration fee quirk.

Sure, that would be <30%YoY, but what is the significance of that number? At some point, Shanghai production will max out at the top of the S curve.