Near-future quarterly financial projections

[EVNow]

Aluminum - Price - Chart - Historical Data - News

Aluminum increased 72 USD/Tonne or 3.02% since the beginning of 2024, according to trading on a contract for difference (CFD) that tracks the benchmark market for this commodity. Aluminum - values, historical data, forecasts and news - updated on April of 2024.

tradingeconomics.com

Cobalt - Price - Chart - Historical Data - News

Cobalt decreased 585 USD/T or 2.01% since the beginning of 2024, according to trading on a contract for difference (CFD) that tracks the benchmark market for this commodity. Cobalt - values, historical data, forecasts and news - updated on April of 2024.

tradingeconomics.com

Nickel - Price - Chart - Historical Data - News

Nickel increased 2,894.50 USD/MT or 17.68% since the beginning of 2024, according to trading on a contract for difference (CFD) that tracks the benchmark market for this commodity. Nickel - values, historical data, forecasts and news - updated on April of 2024.

tradingeconomics.com

There are also some pretty good forecasts available.

The issue is - we don't know when Tesla had to renegotiate contracts ....

 

[petit_bateau]

credit @Chunky Jr.

 

[zach_]

We know that Giga Texas produced 10k Model Ys between Sept 17 and Oct 30 for an average rate over that period of 1600/week. If we approximate and say that means on Oct 1 the rate was 1000/week, and Dec 15 was 3000/week. I realize the ramp is not linear, but if we just estimate say for 11 weeks the average was 2000, and then 3000 per week for the 12th and 13th weeks of the quarter, that's 28,000 from Texas.

With Giga Berlin, the estimated production rate based on VINs to norway has been > 3000/week for quite some time. Obviously this is not perfect, but I feel estimating an average of 2750/week for the quarter is very safe. That's 33,000 from Berlin.

We can just say a slight bump in Fremont to 145,000 (mostly S/X).

Lastly we already know Oct/Nov for Giga Shanghai was 87,000 and 88,000 roughly. Given the repeated denials of the shutdown, if we assumed another 87,000 from Shanghai that's a total of 262,000.

All added up? that's 468,000.

I don't think the Berlin/Texas estimates are too aggressive (I could see Texas being a slight overestimation, but Berlin a slight underestimation). So if the Shanghai production holds I just simply don't buy that deliveries will be ~430,000 like Troy is estimating. Another 40,000 seems like an unreasonably large amount of inventory to add. So I either think production is much lower or deliveries are much higher, but we have more insight into production so I lean more towards an upside surprise in deliveries. Thoughts?

 

[petit_bateau]

We know that Giga Texas produced 10k Model Ys between Sept 17 and Oct 30 for an average rate over that period of 1600/week. If we approximate and say that means on Oct 1 the rate was 1000/week, and Dec 15 was 3000/week. I realize the ramp is not linear, but if we just estimate say for 11 weeks the average was 2000, and then 3000 per week for the 12th and 13th weeks of the quarter, that's 28,000 from Texas.

With Giga Berlin, the estimated production rate based on VINs to norway has been > 3000/week for quite some time. Obviously this is not perfect, but I feel estimating an average of 2750/week for the quarter is very safe. That's 33,000 from Berlin.

We can just say a slight bump in Fremont to 145,000 (mostly S/X).

Lastly we already know Oct/Nov for Giga Shanghai was 87,000 and 88,000 roughly. Given the repeated denials of the shutdown, if we assumed another 87,000 from Shanghai that's a total of 262,000.

All added up? that's 468,000.

I don't think the Berlin/Texas estimates are too aggressive (I could see Texas being a slight overestimation, but Berlin a slight underestimation). So if the Shanghai production holds I just simply don't buy that deliveries will be ~430,000 like Troy is estimating. Another 40,000 seems like an unreasonably large amount of inventory to add. So I either think production is much lower or deliveries are much higher, but we have more insight into production so I lean more towards an upside surprise in deliveries. Thoughts?

Thoughts,

1. PRODUCTION

Your estimates look good to me. Personally I hold 462,711 in my 4Q production estimate. All that means is I am as likely as you to be wrong

2. DELIVERIES

If they are serious about unwinding the EOQ wave, then inventory ought to grow more, say by another 20k vehicles.

My sneaking suspicion based upon the operational info we have re shipping and sales patterns, is they will once again try to game the wave yet again so as to (ideally in their view) reduce the inventory that built last quarter. So the reverse of unwinding the wave ! Again, this means I have no clue.

 

[adiggs]

We know that Giga Texas produced 10k Model Ys between Sept 17 and Oct 30 for an average rate over that period of 1600/week. If we approximate and say that means on Oct 1 the rate was 1000/week, and Dec 15 was 3000/week. I realize the ramp is not linear, but if we just estimate say for 11 weeks the average was 2000, and then 3000 per week for the 12th and 13th weeks of the quarter, that's 28,000 from Texas.

With Giga Berlin, the estimated production rate based on VINs to norway has been > 3000/week for quite some time. Obviously this is not perfect, but I feel estimating an average of 2750/week for the quarter is very safe. That's 33,000 from Berlin.

We can just say a slight bump in Fremont to 145,000 (mostly S/X).

Lastly we already know Oct/Nov for Giga Shanghai was 87,000 and 88,000 roughly. Given the repeated denials of the shutdown, if we assumed another 87,000 from Shanghai that's a total of 262,000.

All added up? that's 468,000.

I don't think the Berlin/Texas estimates are too aggressive (I could see Texas being a slight overestimation, but Berlin a slight underestimation). So if the Shanghai production holds I just simply don't buy that deliveries will be ~430,000 like Troy is estimating. Another 40,000 seems like an unreasonably large amount of inventory to add. So I either think production is much lower or deliveries are much higher, but we have more insight into production so I lean more towards an upside surprise in deliveries. Thoughts?

FWIW my working assumption for undelivered cars / inventory at end of quarter is 10% of quarterly production. This is my attempt at estimating the size of finished inventory in the brave new world of the unwound wave. I realize that might not yet be true, or even that the wave can be wound back up in pursuit of a one-time juice to quarterly deliveries.

I believe there were 20k undelivered vehicles from Q3 - that would get me to 468k production - 47k undelivered this quarter + 20k undelivered from last quarter = 441k Q4 deliveries. With 47k undelivered vehicles going into next quarter, an increase of 27k over last quarter.


This is just my thumb to the wind with an explanation of how I get there. I figure something like this is the relationship we should see until Tesla starts holding inventory at sales locations to serve the "need it now" car market.

If this is fairly accurate then the undelivered units will grow quarter over quarter right along with production. And the quarterly FUD articles are practically already written "Tesla inventory balloons as unsold cars doubles quarter over quarter".

 

[EVNow]

If they are serious about unwinding the EOQ wave, then inventory ought to grow more, say by another 20k vehicles.

I think they will "unwind" the wave when Berlin is ramped up.

Then, we should expect about 2 weeks of production to be in the inventory (or better yet, take last week of quarter off for maintenance/upgrades).

 

[Dreadnought]

In a perfect wave-unwound state, the number of cars in transit is determined by daily production multiplied by shipping time.

Right now, Shanghai and Fremont dominate and 10% (or 9 days) seems like a good approximation. As Berlin and Austin ramp up, their influence on the average transit times will become more significant. More factories means shorter distances and less overseas transport. Over time, the average number of days in transit will be reduced as long as every car on the way is already sold and gets delivered immediately.

 

[wipster]

It's a good point you bring up. I don't do enough of a deep dive in this area. I have a 1% inflation factor Q4 vs Q3 for COGS material components.
The Producer Price Index (PPI) has increased about 0.8% in the last 3 months but I know that the overall PPI is not the basket of goods that Tesla is purchasing. But I also don't have cost improvements such as removing ultrasonic sensors. So hopefully some missing cost savings offset any missing cost increases. Some areas are more art than science in my model.

Are you going to post another comparison spreadsheet with the best guesses for the 4th qtr? Sure enjoyed the last couple, even if most folks overestimated by a substantial amount.

Thanks again!

 

[Doggydogworld]

I believe there were 20k undelivered vehicles from Q3

I show 42.7k undelivered at end of Q3. Tesla reported 8 days supply. Based on their 75 day math that implies 34.4k-39k.

Does everyone here disagree with Troy's delivery forecast? Other than maybe being 5-10k low in China I don't see much to argue with.

 

[wipster]

Thoughts,

1. PRODUCTION

Your estimates look good to me. Personally I hold 462,711 in my 4Q production estimate. All that means is I am as likely as you to be wrong

2. DELIVERIES

If they are serious about unwinding the EOQ wave, then inventory ought to grow more, say by another 20k vehicles.

My sneaking suspicion based upon the operational info we have re shipping and sales patterns, is they will once again try to game the wave yet again so as to (ideally in their view) reduce the inventory that built last quarter. So the reverse of unwinding the wave ! Again, this means I have no clue.

I would bet that deliveries are pretty tight this quarter and we know how important they are, especially at the end of this quarter in particular. As such, Tesla is making every effort to get inventory vehicles into the hands of new owners in order to get a 50% increase for the year on the objective counts. And I think that's great. Tesla can and should work on unwinding the wave in 2023 when they're making even more. I personally think it's amazing that they do what they do especially with the logistics involved... trains, ships, trucks, at the factory, delivered to your home, pick up at the service center, etc.

Unwinding the wave should hopefully make some blood pressures go down next December!

 

[petit_bateau]

I show 42.7k undelivered at end of Q3. Tesla reported 8 days supply. Based on their 75 day math that implies 34.4k-39k.

Does everyone here disagree with Troy's delivery forecast? Other than maybe being 5-10k low in China I don't see much to argue with.

My maths was 30k inventory at end of Q3. Is there an error in my maths, or my understanding ?

Anyway if they are serious about unwinding the wave that might grow further to 50k. My suspicion is they are not serious, and the urge will take them.

(To be honest I've not paid a great deal of attention to anyone else's numbers. I have a lot on my plate here that has nothing to do with Tesla.)

 

[Doggydogworld]

My maths was 30k inventory at end of Q3. Is there an error in my maths, or my understanding ?

View attachment 885876

They base it on 75 selling days. Q3 deliveries were 343,830, divided by 75 that's 4584 per day. Times 8 days is 36,675.

I get 42.7k by tracking production vs. deliveries over time. My number is a bit high since some cars never get delivered (e.g. wrecked in transit or while serving as store demos, maybe some used for engineering then scrapped). Tesla's "days of supply" number is usually a day or so less than my number would indicate, though once in a while it swings the other direction.

 

[elasalle]

I show 42.7k undelivered at end of Q3. Tesla reported 8 days supply. Based on their 75 day math that implies 34.4k-39k.

Does everyone here disagree with Troy's delivery forecast? Other than maybe being 5-10k low in China I don't see much to argue with.

Troys numbers haven’t factored in the latest discounts yet
They have deal for Shanghai only for like 3.5K … so worldwide they could maybe move another 15-20K of product with the discounts.. cheers!!

 

[The Accountant]

Are you going to post another comparison spreadsheet with the best guesses for the 4th qtr? Sure enjoyed the last couple, even if most folks overestimated by a substantial amount.

Thanks again!

Yes - I will put that together again this quarter.

 

[Chunky Jr.]

I wonder what this quarter's one time unexpected negative drag on earnings will be? Seems like there is one every quarter.

 

[The Accountant]

I wonder what this quarter's one time unexpected negative drag on earnings will be? Seems like there is one every quarter.

So far I see a small bitcoin impairment charge.
I still have to check my calculation again but a month ago I calculated approx a $40m charge impacting EPS by $0.01.
The recent price cuts and incentives may hit margins more than expected if we haven't estimated them properly.
That's all that I see at the moment.

 

[bkp_duke]

So far I see a small bitcoin impairment charge.
I still have to check my calculation again but a month ago I calculated approx a $40m charge impacting EPS by $0.01.
The recent price cuts and incentives may hit margins more than expected if we haven't estimated them properly.
That's all that I see at the moment.

Any chance for the deferred tax stuff to be claimed? Doesn't some of it "expire" if not used by end of year?

 

[zach_]

So far I see a small bitcoin impairment charge.
I still have to check my calculation again but a month ago I calculated approx a $40m charge impacting EPS by $0.01.
The recent price cuts and incentives may hit margins more than expected if we haven't estimated them properly.
That's all that I see at the moment.

In the US ASPs could come in a bit lower if FSD take rate goes down. I know it's somewhat counter to what we'd expect with a wide release now, but with car loan rates going way up and people just tightening the belt a bit I could see a non-trivial step down in the amount of people who splurge on a $15k option.

 

[petit_bateau]

I appreciate this is the near-future thread, but I cannot recall where the long-term thread is. In any case I suspect the more interested readership is here.

I put this article (see link below) aside a few days ago to read more thoroughly and have now done so. It sets out the hypothesis that most so-called exponential growth companies are at best quadratic in nature or more likely - though it does not say this - linear. I make the point re linear as when I read the graphs many of them look pretty linear to me, though the article (oddly) call those quadratic, perhaps out of politeness. It also makes points regarding genuine quadratic growth and multiple sequential product releases; releases into successive geographies; and logistics curve (S-curve) growth with good examples. And regarding unit sales (or users) vs revenues, and in changing market sizes. I'm not sure it says anything particularly new, but it does pul it all together in a coherent and useful manner. Therefore it may interest others here.

https://longform.asmartbear.com/exponential-growth/

Personally, in my longer term modelling I do adopt an S-curve (logistics) approach that takes Tesla at their word re both vehicles and energy. Since I also break out my own view of the likely Tesla-product line up (Tesla's push) and what I know about whole-world market vehicle requirements (the world's pull) it is interesting to compare the result to the stacked quadratic the article considers. They look to my eyes to be pretty similar.

As a final comment one could view Tesla's current strategy as being a layered exploitation of three successive meta-product areas:
- vehicles;
- energy;
- autonomy;

I've not yet thought that there is enough evidence to specifically model a revenue line to autonomy (FSD, Optimus, Dojo-as-service, Robotaxi, etc) but I have included a revenue line for energy product sales (but not specifically any energy operations or service revenue, though there is a little bit embedded in the grey S&S as this includes the charger network) so here is that graph for info.

 

[The Accountant]

Any chance for the deferred tax stuff to be claimed? Doesn't some of it "expire" if not used by end of year?

I have an opinion on this which I have hesitated to communicate because it could be wrong.
Not only is the deferred tax topic complex but the Inflation Reduction Act (IRA) has made it more complex.
But let me share my thinking with the caveat it could be very wrong.

I think Tesla will not recognize any of the Deferred Tax Asset due to the IRA Minimum Corporate Tax of 15%.
Only financial reporting net operating losses can be applied to the 15% (known as "Book" NOLs) which Tesla has none. Tesla has "Tax" net operating losses (which generates the deferred tax asset) and that cannot be used to lower the 15%. This is my understanding.

The good news (again, from my understanding) is that the Mfg tax credit in the IRA legislation along with some other tax credits can reduce the 15%.