Near-future quarterly financial projections

[CorneliusXX]

Thanks!


So, theoretically, even if Tesla went to 0% automotive gross margin, they would still be making money ($620 million a quarter at that level) because depreciation is an accounting thing?

Exactly. but making money in this scenario is just the per unit direct revenue received above the direct cost of the labour/materials. Admin/R&D/overheads, etc could lead to a net negative cash position at the company level even if per vehicle cash margins are 0% or better. I haven't reconciled that the $620m is all dropping through COGS but most of it should be.

That's also why many valuations are done on EBITDA multiples - which is (arguably) a better guestimate of ongoing product margins.

 

[damonbrodie]

Anybody have any Q4 EPS thoughts?
What about Q1?

 

[The Accountant]

Nice.

If I was setting out that table in that way I'd include a GM% line as well as a GM line.

Your volume assumption is 113%, i.e. more than doubling. I expect you are doing that calc for USA. However two things:

1) A lot of sales are non-USA and prices have reduced all over the world; but IRA is US-specific. Maybe two customer-view tables are required.
2) The 2023 volume is likely to be of the order of 2.2m vs the 2022 volume of 1.3m, i.e. 72% volume growth, a long way below the 113% you are selecting.

Clearly this is pricing to run Tesla at max capacity during what we expect to be a high-volume increase 2023 year of ?? 72%. The 2022 was a low point at only (!) 40%. The amazing thing is that 2024 is also likely to be high and 81% is my model for that, so 4.1m.

I wonder which will be the first legacy manufacturer to go under.

I ran a scenario on only the Model Y LR in the US because this model/region got the steepest cut at 20%.
I do think the Model Y LR will double in 2023 in the US with Austin ramping. Of course this means other models/trims elsewhere will grow below 50% . . model S, X in the US and model 3 in China, for example, will see much lower growth but their cuts were much lower than 20%.

 

[The Accountant]

How did you derive the COGS number? I've gotten approximately similar guesses for my bull case but I'm wondering what your math looks like. Especially, is that including the $45/kWh battery credit or is that the unsubsidized COGS?

In future revisions do you intend to estimate any increases in Options per Car? This is almost certain to be nonzero but the question is how much. Especially with software features not disqualifying buyers from the IRA $7.5k credit. Normally when prices for goods drop this much, consumers will exploit some of the increased buying power to upgrade instead of just buying the exact same version at a lower price.

Edit: Just saw your footnotes. So, to highlight for others, the $21.3k margin per car estimate excludes software revenue, which will likely increase above 2022 levels.

I think you are likely right; with such a large price drop and rebate available, customers who would not normally spring for Autopilot may now purchase it.
And I was not aware until today that software price does not get included in the cap limit. This was spelled out by Tesla here:

 

[Gigapress]

I think you are likely right; with such a large price drop and rebate available, customers who would not normally spring for Autopilot may now purchase it.
And I was not aware until today that software price does not get included in the cap limit. This was spelled out by Tesla here:
View attachment 895582

Take rates for all of the upsells will likely increase. In general when a product gets cheaper for the base version then some customers will use the savings to get upgrades. For those American customers who qualify for the $7500 credit the hardware upgrades are now unattractive for the Y LR because that’ll push it over the $55k limit, but for those whose income makes them ineligible anyway, I bet many will decide to select more red paint, tow hitches, etc.

 

[petit_bateau]

I think I've previously posted my Q4-2022 forecast but if not here it is again.

Revenue = $25.837 bn
Auto GM% = 28%
GAAP EPS = $1.19
Non-GAAP EPS = $1.29 approx (I'm just adding 10c back in, nothing too sophisticated)

My calculations don't aim for anything near @The Accountant 's level of detail or accuracy so I'm happy to keep on posting mine up for the time being. I only aim for sufficient as to spot and understand trend changes before being smacked in the face by them.

 

[Chunky Jr.]

Lots of estimates around $1.15-$1.20, but since there always seems to be an unexpected one time charge, I'd guess closer to $1.05

 

[navguy12]

I think I've previously posted my Q4-2022 forecast but if not here it is again.

Revenue = $25.837 bn
Auto GM% = 28%
GAAP EPS = $1.19
Non-GAAP EPS = $1.29 approx (I'm just adding 10c back in, nothing too sophisticated)

My calculations don't aim for anything near @The Accountant 's level of detail or accuracy so I'm happy to keep on posting mine up for the time being. I only aim for sufficient as to spot and understand trend changes before being smacked in the face by them.

View attachment 898851

View attachment 898852

Silly question: where are carbon credits listed?

 

[Mengy]

Lots of estimates around $1.15-$1.20, but since there always seems to be an unexpected one time charge, I'd guess closer to $1.05

This does so often seem to be the case. Maybe for Q4 the incentives and discounts for the last few weeks of December will hit margins more than expected? Maybe we'll be pleasantly surprised to the upside for an unusual change?

We'll know in two days!

 

[petit_bateau]

Silly question: where are carbon credits listed?

No idea !

I'd have to trawl through the Tesla quarterly disclosures to figure that out. And that is not how my model operates. It is just a GM% and ARPV style model. That is why it is simple enough for me to maintain without too much effort.

(I'm sure @The Accountant can say where to look in the full Tesla quarterlies)

 

[Doggydogworld]

Silly question: where are carbon credits listed?

Tesla includes regulatory credits in Automotive Sales and Leasing Revenue. They disclose the amount in the "Financial Summary" page of their earnings reports. It's on page 4 of the Q3 report, and shows reg credits were 1.5% of Auto Revenue (286m of 18,692m).

 

[petit_bateau]

Tesla includes regulatory credits in Automotive Sales and Leasing Revenue. They disclose the amount in the "Financial Summary" page of their earnings reports. It's on page 4 of the Q3 report, and shows reg credits were 1.5% of Auto Revenue (286m of 18,692m).

Thanks.

In that case it is a part of my assumed 28% automotive GM%.

(Overall I've held the 28% from Q3 as I have no firm evidence base to vary it in any direction.)

 

[petit_bateau]

What I would really like to see is everyone else's - including Wall Street analysts' - projections for FY 2023 before they magically **puff** and go.

Mine are :
- 2.263m vehicles sold
- 27.9% automotive GM%
- total revenues $145.384bn
- GAAP EPS $7.37

Anyone fancy collating those and reminding us in a year ?

 

[damonbrodie]

What I would really like to see is everyone else's - including Wall Street analysts' - projections for FY 2023 before they magically **puff** and go.

Mine are :
- 2.263m vehicles sold
- 27.9% automotive GM%
- total revenues $145.384bn
- GAAP EPS $7.37

Anyone fancy collating those and reminding us in a year ?

So 70ish percent vehicle growth? That pretty aggressive! Let’s see how that jives with their guidance.

 

[The Accountant]

What I would really like to see is everyone else's - including Wall Street analysts' - projections for FY 2023 before they magically **puff** and go.

Mine are :
- 2.263m vehicles sold
- 27.9% automotive GM%
- total revenues $145.384bn
- GAAP EPS $7.37

Anyone fancy collating those and reminding us in a year ?

Where you have GAAP EPS for 2023 at $7.37, UBS just published their number at $3.38 (and $5.52 for 2024).
I think Wall Street easily lowers revenue for price cuts but has a hard time lower cost of goods sold because they don't usually forecast things until they see it (poor forecasting). COGS should come down due to Berlin/Austin ramping, improved purchasing power for materials at Fremont and Shanghai as volumes grow, lower cost of critical minerals, fixed costs spread over more units produced, manufacturing process improvements, part redesign improvement, etc.
Tesla does not stand still.

 

[The Accountant]

So 70ish percent vehicle growth? That pretty aggressive! Let’s see how that jives with their guidance.

Guidance will likely be 50% but 2.2m+ deliveries are very doable for 2023. I've done the math.
If you just take Shanghai and Fremont exit rate (using November) and assume Berlin and Austin get to 220k each, you are already at 2,070k. That's a 57% increase to 2022. Remember that 2022 is low due to Shanghai COVID slowdown. Also, I have no Cybertruck or Semi in my numbers.
Shanghai and Fremont will not stand still and will deliver above their 2022 exit rate by installing new capacity or through mfg innovations and that would take us to 2.2m+ deliveries for 2023. Of course this assumes no supplier interruptions, no covid issues, no black swan events. Demand will be there.
There are always hiccups along the way so I am modeling 2.1m for 2023 at the moment.

 

[petit_bateau]

Guidance will likely be 50% but 2.2m+ deliveries are very doable for 2023. I've done the math.
If you just take Shanghai and Fremont exit rate (using November) and assume Berlin and Austin get to 220k each, you are already at 2,070k. That's a 57% increase to 2022. Remember that 2022 is low due to Shanghai COVID slowdown. Also, I have no Cybertruck or Semi in my numbers.
Shanghai and Fremont will not stand still and will deliver above their 2022 exit rate by installing new capacity or through mfg innovations and that would take us to 2.2m+ deliveries for 2023. Of course this assumes no supplier interruptions, no covid issues, no black swan events. Demand will be there.
There are always hiccups along the way so I am modeling 2.1m for 2023 at the moment.

I've not accounted for any change in stocks. So my 2.263m is the same for both produced and delivered.

End Q4-22 inventory at 13-days sales (using the 75 days/qtr) was 71,098 vehicles. Between Q4-22 and Q4-23 the prod rate will grow by approx 50%. So if they stay at 13d inventory - which would be impressive - then end Q4-23 inventory would be 35.5k higher, i.e. 106,647 vehicles.

So if I were to allow for this then my produced 2.263m corresponds to delivered 2.228k vehicles for 2023. I'm not sure that trying to build in the additional complexity in my model is sensible given the other bigger factors flying around with bigger error bars on them.

---

I think that Wall St analyst projections are becoming increasingly tight by the end of each quarter, at least that is my impression. They've certainly been closer than many of us here on TMC of late. I do however have a sneaking suspicion that their early-in-the-quarter projections are less good. And correspondingly their early-in-the-year projections are also not so good. Whether my suspicions are justified, and if so what might the reasons for that, I'll leave for those of you who study Wall St behaviour more than I do.

 

[JohnnyEnglish]

Hopefully the announcement today regarding a new factory in Nevada for Semi production (see Teslarati article) will provide some indications of planned capacity and ramp timelines so that we can incorporate them into our models. Presumably it will also prompt some mention in the Q4 ER.

In addition to the stuff covered by the SAY questions I am also hoping for a progress update re the first Dojo Exapod which should be closing on completion. Perhaps that will be deferred to the March 1st shareholder update. It would be interesting to know how many Exapod months of compute are the equivalent of the $10B of Azure credits Microsoft just granted to Open AI

 

[DarkandStormy]

What I would really like to see is everyone else's - including Wall Street analysts' - projections for FY 2023 before they magically **puff** and go.

Mine are :
- 2.263m vehicles sold
- 27.9% automotive GM%
- total revenues $145.384bn
- GAAP EPS $7.37

Anyone fancy collating those and reminding us in a year ?

Are you asking for individual analyst forecasts or just a consensus?

This is per TSLA IR - you can see the 2023 columns on the right. Not sure if you prefer average or median, but both are available for comparison. These might be lower than other compiled forecasts as I think most internally compiled estimates come in a touch low so the company can "beat" forecasts, but who knows.

Per Yahoo Finance at the moment for 2023 forecasts...
-$4.23 avg EPS (non-GAAP, fully diluted I believe) from 31 analysts
-$105.13bn avg revenue from 36 analysts
Fwiw, the highest estimate is $6.19 EPS and $137.35bn in revenue

Marketwatch is showing $4.71 EPS average.

https://twitter.com/x/status/1617633000174518272

I believe Gary is using Bloomberg consensus to represent "Wall Street" here. You can obviously read his forecast vs WS consensus.

 

[petit_bateau]

Are you asking for individual analyst forecasts or just a consensus?

View attachment 899207

This is per TSLA IR - you can see the 2023 columns on the right. Not sure if you prefer average or median, but both are available for comparison. These might be lower than other compiled forecasts as I think most internally compiled estimates come in a touch low so the company can "beat" forecasts, but who knows.

Per Yahoo Finance at the moment for 2023 forecasts...
-$4.23 avg EPS (non-GAAP, fully diluted I believe) from 31 analysts
-$105.13bn avg revenue from 36 analysts
Fwiw, the highest estimate is $6.19 EPS and $137.35bn in revenue

Marketwatch is showing $4.71 EPS average.

https://twitter.com/x/status/1617633000174518272

I believe Gary is using Bloomberg consensus to represent "Wall Street" here. You can obviously read his forecast vs WS consensus.

Thanks. I am not too fussed either way. Really it is just to have a proper comparison in 12-months time of how they have (or have not) walked their estimates up (or down) in the course of the year.

Looking at that Wall St average is only dialling in a rerun of Q4-2022 four times over. Why do they bother doing this ? They obviously know how to do it better. Is this laziness or conspiracy ?

(Thanks again)