Near-future quarterly financial projections

[dc_h]

For Q2, assuming May 11th Fremont opening, I think TE will pickup due to storage, but lag still due to solar. Fremont with 6 weeks of deliverable cars and 7 weeks of build-able cars, should be able to deliver about 50,000 cars in the USA and Canada. Maybe they can ship one ro-ro of cars to Europe and S. Korea, but still limited to 50,000 total deliverable. Shanghai could be over 3000 per week, but 40,000 seems like a reasonable number for MIC Model 3 delivered. That would be more deliveries than Q2 2019, but probably a bit less in revenue, with a lower SX mix (14517 SX & 72531 M3 in Q2/2019). Margins could still be good, with a higher mix is MIC and some expected process improvements in Fremont. I don't want to be looking through rose colored glasses, but it seems like Q2 could be revenue neutral. Every quarter they can stay revenue neutral through this epidemic is a victory and if they can show a profit, it really positions them to exit the crisis with incredible momentum.
Tesla also predicted MIC Model 3 production would be 4000 per week by mid 2020. That's a loose way of saying June, so they should be over 3000 per week for most of this quarter.

 

[Doggydogworld]

They deployed a record 530MWh energy storage - but Energy revenue as a whole was down 33% over Q4.

They deployed 260 MWh in Q1, down >50% Q/Q.

They disclosed new info on energy sales and leasing on page 10 of the 10-Q:

1Q20 1Q19
173 212 Energy generation and storage sales
120 112 Energy generation and storage leasing

Sales were down 39m Y/Y. 260 vs. 229 MWh of storage at $500/kWh is +$16.5m. 35 vs. 47 MW of solar would be -$24m assuming 70% cash/loan sales and only token amounts of solar roof. Cash/loan percentage must have fallen off a cliff, which makes some sense with their new $50/month marketing push and the fact that they suddenly stopped touting cash/loan percentage last summer.

I always estimated solar leasing at 70m in Q1 and Q4, 100m in Q2/Q3. That's based on old SCTY SEC filings and Powerpoint slides that disclosed typical kWh production and per kWh pricing. It's also consistent with ~200m annual depreciation and the 32% margins they used to disclose. 120m doesn't really fit with any of those previously disclosed numbers. Do they lease lots of Powerwalls and Powerpacks? The Australia deal was a sale and the others I've heard about (e.g. Nantucket, Moss Landing) all seem to be sales. And there's no billion dollar storage lease portfolio on the balance sheet.

I really don't get this.

 

[EVNow]

They deployed 260 MWh in Q1, down >50% Q/Q.

Oh yes - I misapplied the news report that was posted.

​

They disclosed new info on energy sales and leasing on page 10 of the 10-Q:

1Q20 1Q19
173 212 Energy generation and storage sales
120 112 Energy generation and storage leasing

Sales were down 39m Y/Y. 260 vs. 229 MWh of storage at $500/kWh is +$16.5m. 35 vs. 47 MW of solar would be -$24m assuming 70% cash/loan sales and only token amounts of solar roof. Cash/loan percentage must have fallen off a cliff, which makes some sense with their new $50/month marketing push and the fact that they suddenly stopped touting cash/loan percentage last summer.

I always estimated solar leasing at 70m in Q1 and Q4, 100m in Q2/Q3. That's based on old SCTY SEC filings and Powerpoint slides that disclosed typical kWh production and per kWh pricing. It's also consistent with ~200m annual depreciation and the 32% margins they used to disclose. 120m doesn't really fit with any of those previously disclosed numbers. Do they lease lots of Powerwalls and Powerpacks? The Australia deal was a sale and the others I've heard about (e.g. Nantucket, Moss Landing) all seem to be sales. And there's no billion dollar storage lease portfolio on the balance sheet.

I really don't get this.

Here is an interesting lease deal for Power Wall - though I think it must have been paid by the utility ?

2019 Tesla Powerwall Installation Case Study | EnergySage

It would make sense for PowerPacks to be leased by commercial customers who don't have much cash (or may be even otherwise).

 

[kengchang]

Leasing Powerwall for residential customer is not available even for the solar lease customer. Not sure about the commercial side though.

 

[hobbes]

Found a twitter thread by carsonight from 3 days ago, supporting the idea that storage might "save" Q2 results-
posted this on the main thread before but I think it is relevant here, too:
Carsonight on Twitter

Before someone asks, carsonight is a well-known credible source with good connections to GF1.

 

[hobbes]

Order page for all European countries (o.k., I checked about 10) and Australia says August for deliveries. That likely means they will produce those cars in July, so Q2 is North America only. Assuming that demand drop due to Corona is less than production drop, they can probably deliver all cars made in Q2, no guesswork with ship timing and no switching between North America and European Models.

 

[Doggydogworld]

Order page for all European countries (o.k., I checked about 10) and Australia says August for deliveries. That likely means they will produce those cars in July, so Q2 is North America only. Assuming that demand drop due to Corona is less than production drop, they can probably deliver all cars made in Q2, no guesswork with ship timing and no switching between North America and European Models.

Tesla is steering European customers to shop inventory instead of ordering a custom configuration which they might not be able to deliver this quarter. If they had restarted production today they might have changed the European ordering pages. Alas, the fascists of Alameda county are once again stopping Elon from saving the planet.

Assuming they are allowed to restart later this month, this would be an ideal quarter to unwind the wave.

 

[EVNow]

Order page for all European countries (o.k., I checked about 10) and Australia says August for deliveries. That likely means they will produce those cars in July, so Q2 is North America only. Assuming that demand drop due to Corona is less than production drop, they can probably deliver all cars made in Q2, no guesswork with ship timing and no switching between North America and European Models.

That's unfortunate. No EU deliveries means no FCA credits.

I'm trying to guess optimistic Q2 production / deliveries to model Q2 + deliveries from inventory of 11k+20k+200 S&X, 3 and Y.
GF3 : 40k / 35k + 5k
M3 in Fremont : 30k / 25k + 10k - 5k/wk for 6 weeks
Y in Fremont : 10k / 8k + 200
S&X : 8k/6k + 3k

Total production : 88k (vs 101k in Q1)
Total deliveries : 92.2k - 75k 3, 8.2 Y, 9k S&X (vs 88.4k in Q1) <-- this is only 5k less than my pre-covid estimate.

This gives me a GAAP loss of $214M and non-gaap profit of $85M. Similarly to what I forecast for Q1. The difference is $200M of regulatory credits compared to Q1 actuals.

 

[mrdoubleb]

That's unfortunate. No EU deliveries means no FCA credits.

I'm trying to guess optimistic Q2 production / deliveries to model Q2 + deliveries from inventory of 11k+20k+200 S&X, 3 and Y.
GF3 : 40k / 35k + 5k
M3 in Fremont : 30k / 25k + 10k - 5k/wk for 6 weeks
Y in Fremont : 10k / 8k + 200
S&X : 8k/6k + 3k

Total production : 88k (vs 101k in Q1)
Total deliveries : 92.2k - 75k 3, 8.2 Y, 9k S&X (vs 88.4k in Q1) <-- this is only 5k less than my pre-covid estimate.

This gives me a GAAP loss of $214M and non-gaap profit of $85M. Similarly to what I forecast for Q1. The difference is $200M of regulatory credits compared to Q1 actuals.

@Troy estimates ~7500 inventory in Europe. In April they sold ~2900. Don't expect this to go to zero, but still a couple thousnds to sell in Q2.

 

[vikings123]

That's unfortunate. No EU deliveries means no FCA credits.

I'm trying to guess optimistic Q2 production / deliveries to model Q2 + deliveries from inventory of 11k+20k+200 S&X, 3 and Y.
GF3 : 40k / 35k + 5k
M3 in Fremont : 30k / 25k + 10k - 5k/wk for 6 weeks
Y in Fremont : 10k / 8k + 200
S&X : 8k/6k + 3k

Total production : 88k (vs 101k in Q1)
Total deliveries : 92.2k - 75k 3, 8.2 Y, 9k S&X (vs 88.4k in Q1) <-- this is only 5k less than my pre-covid estimate.

This gives me a GAAP loss of $214M and non-gaap profit of $85M. Similarly to what I forecast for Q1. The difference is $200M of regulatory credits compared to Q1 actuals.

The absence of regulatory credits gets offset by the recognition of deferred FSD revenue, No? I think Q2 GAAP profit is not completely off the table if we start production next week.

 

[StarFoxisDown!]

The absence of regulatory credits gets offset by the recognition of deferred FSD revenue, No? I think Q2 GAAP profit is not completely off the table if we start production next week.

Also, Tesla furloughed a lot of it's workforce for a month and a half. Would that not reduce costs somewhat?

 

[EVNow]

The absence of regulatory credits gets offset by the recognition of deferred FSD revenue, No? I think Q2 GAAP profit is not completely off the table if we start production next week.

It’s going to be a small FSD revenue recognition. They delivered a small piece of city NOA.

They have a total of $600M deferred revenue - may be they can recognize $60M, esp because the feature is US only.

More importantly what about the 3k deliveries in China in April. @FrankSG what do you make of this ? I don’t see how we get to 30k or 40k deliveries in China from here. I guess there was no lack of inventory. We also saw Musk reduce price in China during ER. I think we have to rethink deliveries.

 

[dc_h]

It’s going to be a small FSD revenue recognition. They delivered a small piece of city NOA.

They have a total of $600M deferred revenue - may be they can recognize $60M, esp because the feature is US only.

More importantly what about the 3k deliveries in China in April. @FrankSG what do you make of this ? I don’t see how we get to 30k or 40k deliveries in China from here. I guess there was no lack of inventory. We also saw Musk reduce price in China during ER. I think we have to rethink deliveries.

Troy commented in twitter that he thought the mix of SR was too high and people wanted the LR. With the price reduction for the SR and qualification for subsides SR sales will take off in May and June. I would also guess there may be some mini wave as they restock show rooms and ship product farther away from Shanghai. Elon must be tough on the finance people, maximize cash flow every quarter! Also wonder if any MIC could be exported in region? China to Australia & S. Korea would be a lot faster and cheaper.
With production starting this week, it appears they will get 1-2 ships to Europe this quarter. The first truck off the lot appeared to show Euro style plates, so some of the load appears europe bound.
Production from Fremont this week not likely over 3000. Next week higher production, but what percent of capacity will they be able to run? I'd guess they will run over 70% of capacity and expect they will be looking at updating assembly process and component design where distancing and manual assembly is still required. More model Y wiring and components in the 3. Will be interesting how much process was updated during plant closing.

 

[damocles11]

That's unfortunate. No EU deliveries means no FCA credits.

I'm trying to guess optimistic Q2 production / deliveries to model Q2 + deliveries from inventory of 11k+20k+200 S&X, 3 and Y.
GF3 : 40k / 35k + 5k
M3 in Fremont : 30k / 25k + 10k - 5k/wk for 6 weeks
Y in Fremont : 10k / 8k + 200
S&X : 8k/6k + 3k

Total production : 88k (vs 101k in Q1)
Total deliveries : 92.2k - 75k 3, 8.2 Y, 9k S&X (vs 88.4k in Q1) <-- this is only 5k less than my pre-covid estimate.

This gives me a GAAP loss of $214M and non-gaap profit of $85M. Similarly to what I forecast for Q1. The difference is $200M of regulatory credits compared to Q1 actuals.

In Q1, there was 14K more of cars produced than sold, which makes sense as finished goods inventory increased from $1.3B to $1.9B, and this $600M increase averages to around $40k cost per car which also makes sense (as $50k per ASP makes 20% margin).

However, there is still $1.3B of finished goods inventory on hand at the end of Q1, excluding the 14k cars. I do not know the breakout between finished goods energy products and finished goods cars as they are lumped together. However, if you make a guess that half of the $1.3B is cars, there could be potentially another ~16K of cars on hand, if you do the math of $650M divided by $40k avg cost of car.

TLDR, I think it's possible that there may have been as many as 30K cars available for sale at the end of Q1. Therefore I am anticipating that deliveries may be significantly greater than production in Q2, assuming there is no demand problem.​

 

[The Accountant]

In Q1, there was 14K more of cars produced than sold, which makes sense as finished goods inventory increased from $1.3B to $1.9B, and this $600M increase averages to around $40k cost per car which also makes sense (as $50k per ASP makes 20% margin).

However, there is still $1.3B of finished goods inventory on hand at the end of Q1, excluding the 14k cars. I do not know the breakout between finished goods energy products and finished goods cars as they are lumped together. However, if you make a guess that half of the $1.3B is cars, there could be potentially another ~16K of cars on hand, if you do the math of $650M divided by $40k avg cost of car.

TLDR, I think it's possible that there may have been as many as 30K cars available for sale at the end of Q1. Therefore I am anticipating that deliveries may be significantly greater than production in Q2, assuming there is no demand problem.​

30k autos in inventory is a good estimate. I believe Troy has this estimate as well.

Q2 Shipments may evolve as follows:
- 30k autos in inventory at end of Q1 - these get delivered
- China delivers locally produced vehicles of 30k.
- That is a total of 60k
- Fremont produces 8k in May and 30k in June (Total 38k)
- That is a total of 98k units available for delivery
- back out 8k undelivered and left in inventory = 90k delivered in Q2
- This assumes no parts shortages.

How likely is 90k deliveries? Not sure...but it is possible. This is all back of the envelope.
I will be watching the Fremont production ramp-up closely to develop a better estimate by quater end.

 

[EVNow]

TLDR, I think it's possible that there may have been as many as 30K cars available for sale at the end of Q1. Therefore I am anticipating that deliveries may be significantly greater than production in Q2, assuming there is no demand problem.

I track inventory and I have given my estimates in my post. Yes about 31k.

 

[Doggydogworld]

30k autos in inventory is a good estimate. I believe Troy has this estimate as well.

Q2 Shipments may evolve as follows:
- 30k autos in inventory at end of Q1 - these get delivered
- China delivers locally produced vehicles of 30k.
- That is a total of 60k
- Fremont produces 8k in May and 30k in June (Total 38k)
- That is a total of 98k units available for delivery
- back out 8k undelivered and left in inventory = 90k delivered in Q2
- This assumes no parts shortages.

How likely is 90k deliveries? Not sure...but it is possible. This is all back of the envelope.
I will be watching the Fremont production ramp-up closely to develop a better estimate by quater end.

8k inventory is too low. They had 8k in Model 3 inventory alone at the end of 2018, with zero international cars and a full court press to deliver every single US car possible before the clock struck midnight and the incentive got cut 50%. 8k is also pretty close to the minimum S+X inventory. Thousands of those care are loaners and demos, technically for sale but with very low turnover. The rest are cars that get stuck in transit, refused at delivery or simply can't be matched to a buyer for that exact configuration in that exact location as the quarter closes.

15k at EOQ would be a very good result, 12k would be outstanding. On the flip side, they could in theory produce more than 30+38k this quarter. No way to know, unless you have factory sources you'd like to share with us

I don't know what to make of China's 3k April deliveries. Data coming out of China is always a little suspect, anyway.

 

[The Accountant]

8k inventory is too low. They had 8k in Model 3 inventory alone at the end of 2018, with zero international cars and a full court press to deliver every single US car possible before the clock struck midnight and the incentive got cut 50%. 8k is also pretty close to the minimum S+X inventory. Thousands of those care are loaners and demos, technically for sale but with very low turnover. The rest are cars that get stuck in transit, refused at delivery or simply can't be matched to a buyer for that exact configuration in that exact location as the quarter closes.

15k at EOQ would be a very good result, 12k would be outstanding. On the flip side, they could in theory produce more than 30+38k this quarter. No way to know, unless you have factory sources you'd like to share with us

I don't know what to make of China's 3k April deliveries. Data coming out of China is always a little suspect, anyway.

Thanks for the feedback on the 15k EOQ number. I'm sure your estimate is better than mine.
Hopefully we'll come across some clues alerting us when Fremont is at full production again

 

[kengchang]

For people who are forecasting Q2 result, is the salary cut and no equity grant in Q2 are being factored in?

 

[EVNow]

For people who are forecasting Q2 result, is the salary cut and no equity grant in Q2 are being factored in?

Not yet in my model - but really first we need a good delivery estimate.

BTW, we have to see if they reserve something for CEO compensation. That can really hurt gaap bottomline.

8k inventory is too low.

Yes, that’s too low. My assumption was about 10k less than in Q1 end. BTW, Q1 end inventory was quite low in US because last week production stopped. Of course in EU they couldn’t deliver towards the end.

We have to wonder how much of old inventory they can liquidate because of production disruptions.