-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
I may be wrong, but I predict a small profit.
Rationale: Start with Q3 reported profits. Subtract an amount for fewer cars delivered. Add Q3 reported SCTY profits (SCTY accounting is wacky but they can arrange to report profits by opening a smaller dollar value of new PPAs/loans than the dollar value of those they monetize). Subtract something for integration costs. Add an estimate for battery sales profits...
Profits from battery sales are uncertain due to uncertain volume of sales and uncertain gross margins. I suspect it will balance out somewhere between half and all of the loss from selling fewer cars. The SCTY numbers are so driven by financial engineering that they're manipulable (unfortunately), but anyway I'm pretty sure they can book enough of a profit off of them to pay for the integration costs and the rest of the car sales miss. One wild card: accounting costs may skyrocket for a little while, but I suspect that will happen in Q1 instead of Q4.
------
Later quarters have interesting accounting changes coming. I am no expert on accounting standards, but this is my layman's interpretation of what seems to be happening:
The revenue recognition standard must be adopted in Q1 2018 and can be adopted as early as Q1 2017 (but not earlier).
This, if I'm correct (and I might not be) causes variable income from PPAs (where SCTY owns the panel and the homeowner contracts to pay for the power) to be recognized upfront upon contract signing and panel installation, rather than over time. (Defaults on payment or bad estimates of production level would lead to write-downs later.) This makes SCTY look *much* more profitable. I don't know how the transition works but there should be an enormous one-time accounting change profit.
It also cleans up the ungodly accounting mess related to the resale value guarantee. I might be wrong, but I believe the full value of the sales of cars with the RVG gets recognized immediately (apart from the part already recognized). The RVG probably gets entered as a liability to compensate. Should be a large accounting-change profit.
Later, the lease accounting changes. This is required to be adopted in 2019, but can be adopted early.
If I'm right, this changes most fixed-payment SCTY leases from "operating" leases to "sales-type leases", causing revenue to be recognized upfront upon sale. (Again, defaults == write-downs.) It probably changes some to "direct financing leases".
These are both going to be nontrivial to adopt. Every one of the hundreds of specialized LLCs used for financing by SCTY has to redo its accounting. In order to present "comparable results", the accounting probably has to be redone going back an entire year. It might be worth seeing whether there are filings relating to these accounting changes at the LLCs as an early hint. Accounting costs may be very large for a couple of quarters. :-(
Anyway, right now the accounting for SCTY is whacked: it's very close to cash accounting, reporting every single sale as a loss now and income later, totally inconsistent with accrual accounting principles. That's desirable for a cash flow statement, but way off for a P&L statement. These changes should fix that.
I don't know when Tesla is going to adopt the lease accounting, though if I were them I'd be doing it as soon as I could. (The revenue accounting they *have* to adopt for Q1.) But it'll cause a large one-time accounting-change-driven improvement in the balance sheet, which will end up being booked as profit (since that's the only way to make the books balance).
The short-sellers will no doubt find ways to say bad things about this, but the fact is the new accounting standards make a lot more sense than the old ones. (It means SolarCity accounting will be pretty transparent. You'll be able to tell whether the monetization transactions were at a profit or a loss, you'll be able to tell what the effective interest rates on the leases were, etc.)
----
So I'm currently anticipating a profitable Q4, and a ridiculously profitable Q1 due primarily to one-time accounting changes. The big question: can they pull off a profitable Q2, or will ramping up for Model 3 put them back into GAAP losses?
Rationale: Start with Q3 reported profits. Subtract an amount for fewer cars delivered. Add Q3 reported SCTY profits (SCTY accounting is wacky but they can arrange to report profits by opening a smaller dollar value of new PPAs/loans than the dollar value of those they monetize). Subtract something for integration costs. Add an estimate for battery sales profits...
Profits from battery sales are uncertain due to uncertain volume of sales and uncertain gross margins. I suspect it will balance out somewhere between half and all of the loss from selling fewer cars. The SCTY numbers are so driven by financial engineering that they're manipulable (unfortunately), but anyway I'm pretty sure they can book enough of a profit off of them to pay for the integration costs and the rest of the car sales miss. One wild card: accounting costs may skyrocket for a little while, but I suspect that will happen in Q1 instead of Q4.
------
Later quarters have interesting accounting changes coming. I am no expert on accounting standards, but this is my layman's interpretation of what seems to be happening:
The revenue recognition standard must be adopted in Q1 2018 and can be adopted as early as Q1 2017 (but not earlier).
This, if I'm correct (and I might not be) causes variable income from PPAs (where SCTY owns the panel and the homeowner contracts to pay for the power) to be recognized upfront upon contract signing and panel installation, rather than over time. (Defaults on payment or bad estimates of production level would lead to write-downs later.) This makes SCTY look *much* more profitable. I don't know how the transition works but there should be an enormous one-time accounting change profit.
It also cleans up the ungodly accounting mess related to the resale value guarantee. I might be wrong, but I believe the full value of the sales of cars with the RVG gets recognized immediately (apart from the part already recognized). The RVG probably gets entered as a liability to compensate. Should be a large accounting-change profit.
Later, the lease accounting changes. This is required to be adopted in 2019, but can be adopted early.
If I'm right, this changes most fixed-payment SCTY leases from "operating" leases to "sales-type leases", causing revenue to be recognized upfront upon sale. (Again, defaults == write-downs.) It probably changes some to "direct financing leases".
These are both going to be nontrivial to adopt. Every one of the hundreds of specialized LLCs used for financing by SCTY has to redo its accounting. In order to present "comparable results", the accounting probably has to be redone going back an entire year. It might be worth seeing whether there are filings relating to these accounting changes at the LLCs as an early hint. Accounting costs may be very large for a couple of quarters. :-(
Anyway, right now the accounting for SCTY is whacked: it's very close to cash accounting, reporting every single sale as a loss now and income later, totally inconsistent with accrual accounting principles. That's desirable for a cash flow statement, but way off for a P&L statement. These changes should fix that.
I don't know when Tesla is going to adopt the lease accounting, though if I were them I'd be doing it as soon as I could. (The revenue accounting they *have* to adopt for Q1.) But it'll cause a large one-time accounting-change-driven improvement in the balance sheet, which will end up being booked as profit (since that's the only way to make the books balance).
The short-sellers will no doubt find ways to say bad things about this, but the fact is the new accounting standards make a lot more sense than the old ones. (It means SolarCity accounting will be pretty transparent. You'll be able to tell whether the monetization transactions were at a profit or a loss, you'll be able to tell what the effective interest rates on the leases were, etc.)
----
So I'm currently anticipating a profitable Q4, and a ridiculously profitable Q1 due primarily to one-time accounting changes. The big question: can they pull off a profitable Q2, or will ramping up for Model 3 put them back into GAAP losses?
Last edited:
cross posting from a post I made several weeks ago
----------------------------------------------------
So I sat down and made some assumptions and did some calculations to try to predict the coming ER. Here are the assumptions and results
1. S and X deliveries as reported yesterday, S 12700 and X 9500
2. ASP for S at $87 k. This is based on in Q3 ER they said ASP for S decreased QoQ by 6.5% due to late introduction of P100D (4.5%) and the discounts (2%). I simply added these back because we had a full Q of P100D and no widespread discount happened. In addition the price hike of AP 2.0 would also contribute here. Note that I am accounting $1.5 k for this because out of the full 8k price of AP 2.0, I am assuming half of it (AP 1.0 function parity not achieved at end of year) got recognized. And not all deliveries in Q4 got AP 2.0 (it came out in Oct after all). The result is $87k.
3. ASP for X at $104 k. Same approach as above, only not including adding 2% back for discounts because there were few in Q3 to begin with.
4. Gross margin of S increased to 31%. Aside of a flat 2% due to no discounts, the high margins of P100D, AP 2.0, and continue improvement of supply chain/manufacture efficiency all contributed for another 2%.
5. Gross margin of X increased to 28%. Mainly due to similar reasons except for the 2% discount but more gains from manufacture efficiency.
6. Regulatory credit contribution to gross margin balanced out with the strong dollar. This is because Tesla have no horded ZEV to sell so the positive impact on gross margin would be far less. No net impact on overall automotive gross margin.
7. A nice increase in the revenue from services and others to $200 m in total mainly due to TE initial ramping up. However, I still consider gross margins on TE in Q4 very poor (around 0) so gross profit for this section is still very low - $10 m.
8. Overall I think expansion of stores and superchargers was on a similar pace with Q3, so OpEx increase the same rate QoQ, $593 m compared to $551 m in Q3.
9. No idea on the interest side but made similar assumptions in OpEx, net loss $62 m compared to $56 m in Q3.
10. Same provision for income taxes of $8 m.
With these assumptions, total revenue would be $2,295 m with $2,095 m coming from automotive. Gross profit $630 m compared to $638 m in Q3. After deducting OpEx, interest expenses, and income taxes, net loss is $33 m. However, this number is very small and could easily swing $100 m to either side. But overall I would expect to see a small loss of EPS for Q4.
As for cash flow, since in my calculation, gross profit is stable thanks to higher gross margins, I continue to assume cash flow provided by operating activities be about the same as in Q3, or $424 m. Deducting $1 b of predicted CapEx results in a negative $576 m. Full AP 2.0 could provide at least $50 m additional cash in Q4. In addition, Tesla said 2750 cars were not delivered but already paid in full. These won't count into income but I think goes in to cash flow. A nice $250+ m. Now we're at about $275 m negative on the free cash flow. However, there's also the increase of $500 m in ABL and accounts payable to soak up cash drain so maintaining a positive free cash flow is too out of this world I think. Even if SolarCity part didn't do great, the small size of this segment of the company won't make things too bad.
TL;DR. I expect close to breaking even on the EPS and slightly negative free cash flow for Q4.
----------------------------------------------------
So I sat down and made some assumptions and did some calculations to try to predict the coming ER. Here are the assumptions and results
1. S and X deliveries as reported yesterday, S 12700 and X 9500
2. ASP for S at $87 k. This is based on in Q3 ER they said ASP for S decreased QoQ by 6.5% due to late introduction of P100D (4.5%) and the discounts (2%). I simply added these back because we had a full Q of P100D and no widespread discount happened. In addition the price hike of AP 2.0 would also contribute here. Note that I am accounting $1.5 k for this because out of the full 8k price of AP 2.0, I am assuming half of it (AP 1.0 function parity not achieved at end of year) got recognized. And not all deliveries in Q4 got AP 2.0 (it came out in Oct after all). The result is $87k.
3. ASP for X at $104 k. Same approach as above, only not including adding 2% back for discounts because there were few in Q3 to begin with.
4. Gross margin of S increased to 31%. Aside of a flat 2% due to no discounts, the high margins of P100D, AP 2.0, and continue improvement of supply chain/manufacture efficiency all contributed for another 2%.
5. Gross margin of X increased to 28%. Mainly due to similar reasons except for the 2% discount but more gains from manufacture efficiency.
6. Regulatory credit contribution to gross margin balanced out with the strong dollar. This is because Tesla have no horded ZEV to sell so the positive impact on gross margin would be far less. No net impact on overall automotive gross margin.
7. A nice increase in the revenue from services and others to $200 m in total mainly due to TE initial ramping up. However, I still consider gross margins on TE in Q4 very poor (around 0) so gross profit for this section is still very low - $10 m.
8. Overall I think expansion of stores and superchargers was on a similar pace with Q3, so OpEx increase the same rate QoQ, $593 m compared to $551 m in Q3.
9. No idea on the interest side but made similar assumptions in OpEx, net loss $62 m compared to $56 m in Q3.
10. Same provision for income taxes of $8 m.
With these assumptions, total revenue would be $2,295 m with $2,095 m coming from automotive. Gross profit $630 m compared to $638 m in Q3. After deducting OpEx, interest expenses, and income taxes, net loss is $33 m. However, this number is very small and could easily swing $100 m to either side. But overall I would expect to see a small loss of EPS for Q4.
As for cash flow, since in my calculation, gross profit is stable thanks to higher gross margins, I continue to assume cash flow provided by operating activities be about the same as in Q3, or $424 m. Deducting $1 b of predicted CapEx results in a negative $576 m. Full AP 2.0 could provide at least $50 m additional cash in Q4. In addition, Tesla said 2750 cars were not delivered but already paid in full. These won't count into income but I think goes in to cash flow. A nice $250+ m. Now we're at about $275 m negative on the free cash flow. However, there's also the increase of $500 m in ABL and accounts payable to soak up cash drain so maintaining a positive free cash flow is too out of this world I think. Even if SolarCity part didn't do great, the small size of this segment of the company won't make things too bad.
TL;DR. I expect close to breaking even on the EPS and slightly negative free cash flow for Q4.
FWIW, I expect massively high margins on battery sales. When you compare the price on the website ($400/kwh+) with Tesla's claims about production cost for battery packs (<$190/kwh, probably as low as $133/kwh), a reasonable estimate for inverter costs ($25/kwh), and something for installation (a plausible but maybe-too-high estimate is $75/kwh), the gross margins just look enormous to me.
Eventually, in order to be cheap enough to undercut the competition from grid fossil fuel plants, Tesla will have to cut the battery prices. So I wouldn't expect these high margins to last for years; they won't. But right now they're selling to the desperate. So they might as well take the high gross margins now, when they need the cash and the "desperate for batteries" market hasn't been sated yet.
Eventually, in order to be cheap enough to undercut the competition from grid fossil fuel plants, Tesla will have to cut the battery prices. So I wouldn't expect these high margins to last for years; they won't. But right now they're selling to the desperate. So they might as well take the high gross margins now, when they need the cash and the "desperate for batteries" market hasn't been sated yet.
I don't know if I pay attention enough or can find good enough data to come up with worthwhile quarterly estimates anymore. My thoughts over the last few quarters are basically like Fallen, on the one hand they definitely have the possibility of having good numbers if things fall the right way, but of course this is a company that isn't really trying much to make short-term profits. Moreover imo the accounting stuff you are talking about could be a very significant thing if they decided it was time.
With the accounting fees I wouldn't expect them to be that important relative to other expenses, but I don't know they have a lot going on so maybe.
With regard to scty q4, I think they've said multiple times that they expected scty to be a cash generator, and possibly profitable since they were planning to switch to completely loan/cash deals rather than loans which makes sense from a gaap perspective and cash-flow.
With regard to the larger accounting changes, one thing that I think is a possibility like you said, is that they will make the switch right now. They might announce it before hand to give analysts a chance to adjust, or they might just go ahead and do it, which I think would be a big surprise, and I don't know of any rule that says they have to announce ahead of time. Also, it's not really just a one time accounting change, it will apply retroactively as well, so the veil that was covering scty sleeper status might come off, and it will make all the lease accounting stuff look better. Just a possibility, maybe they wait till next year, maybe it won't be seen as that big of a deal. But I do think it is basically like an ace up or more up Tesla's sleeve, so it's hard for me to imagine it not being a big deal if they do it. I don't want to stoke false hope but I have to wonder what would happen if a) tsla posts profit in q4 b) it's big enough make 2016 profitable c) they make the accounting switch d) they also announce 2017 guidance which of course will be big e) any number of other catalysts in the pipeline
MitchJi
Trying to learn kindness, patience & forgiveness
One thing that you should be aware of is that the $190 per kWh pack figure is for car packs, which have a greater energy density, hence a lower cost per kWh .
Some people disputed that fact after I stated it a few days ago. In this short talk by JB he says the same thing at about 18 minutes. At the beginning of the talk he explains the reasons for founding Tesla, (for those who were aurguing about that):
Ummmm.... this doesn't make sense to me. Do you mean that the car CELLS have a greater energy density than the stationary CELLS? Because if the car packs are simply packing the cells closer together, that doesn't really change the cost per kwh at all.
Well, assuming a fixed markup from cost of production, a lower specific energy cell would cost more per kWh. And NMC cells do cost more than NCA ones... maybe it's the much higher cobalt content. In any case, I believe the stationary storage NMC cells to cost substantially more than the automotive NCA cells. Further, we do not know the costs for the production of the stationary storage products beyond the cells. The only metrics we have is stationary storage dropped into a category with many other things and an overall poor margin. Certainly, at high volume, we expect solid margins. We just don't know where they are in the volume ramp and therefore the cost curve.
MitchJi
Trying to learn kindness, patience & forgiveness
Yes. They use cells that are optimized more for high cycle life in the TE products and are more optimized for energy density.
Remember when they introduced the V1 TE products between the two there was a substantial difference in the per kWh price? The increased cost between the two was more than the difference between the two types so there were other factors as well. @techmaven did a good job taking about some of the other factors.
The reason I pointed out the price difference between two types of TE cells is that demonstrates that it makes sense for car packs to cost less per kWh than TE packs if there is a substantial difference between two types if TE packs.
My wag is that the car packs cost about 10-25% less than the TE packs. The main reason I decided to post this now is because I forgot to include that in my earlier $80-95 pack price estimate which is for cars.
Last edited:
MitchJi
Trying to learn kindness, patience & forgiveness
Edit Addition: After I wrote this post I noticed that you said small loss. Maybe the small profit I remember was after Q4? In any case I think the remaining part of my post is accurate.
That was when they were producing packs by hand, which is part of scale, which I agree does matter a lot. Those margins were also using-18650's produced in Japan. They stated it as a positive. Starting production and having low volume production they still made a small profit.
In any case the current production is so different that I don't believe that we know enough to to use that to draw any meaningful conclusions.
Last edited:
They reported a small gross profit for Q1 2016 and said that in the letter. No information for Q2 2016. For Q3 2016 they didn't say anything in the letter, but provided some numbers in the 10-Q report about the increased revenue and cost for TE and the result was a negative gross margin.
There are more cost to this. The entire GF they have built, no matter if only 10% or 90% of the space is utilized, is a depreciation cost on the activities happening there. In Q3 they already have the robots assembling the pack so it shouldn't be a big factor to the then high cost. But the mostly empty space of the huge factory shared by so little output is the biggest cost item.
Just to illustrate my point of the depreciation cost of the GF. I think up to now the total accumulative capex on the GF is something on the order of $800M. Tesla depreciates buildings over a 30 years timeframe or 120 quarters and much shorter (3-12 years) for machinary and equipment. Even if we take a flat 120 quarters, that's $6.7M per quarter. Seems small right? But the problem is the output of TE is not big either. Compare to Q3 2015, when TE was really not even happening (reveal event was in May 2015), Q3 2016 TE revenue increased $22M in revenue, or roughly 55MWh with an approximate $400/kWh price to customer.. So that's $128/kWh of depreciation cost of the GF alone.
The good thing is, this cost should be negligible once TE scales up, which looks more and more promising. Even if we take JB's word (not material to break out as a stand alone in several years) in the recent article at face value, and use 10% as the break out threshold, that could mean revenue level of $200-300M per quarter even without considering the revenue of Model 3, or ~%750M when using a 500k sales of Model 3 per year. At that level, the depreciation cost of GF is just a few bucks per kWh.
The good thing is, this cost should be negligible once TE scales up, which looks more and more promising. Even if we take JB's word (not material to break out as a stand alone in several years) in the recent article at face value, and use 10% as the break out threshold, that could mean revenue level of $200-300M per quarter even without considering the revenue of Model 3, or ~%750M when using a 500k sales of Model 3 per year. At that level, the depreciation cost of GF is just a few bucks per kWh.
One thing I was wondering about - Tesla said in late 2016 that they didn´t need to raise money though they might still do it if a chance comes up IIRC. Wouldn´t now after the run up be a good time to do that? Then, doesn´t the fact that they haven´t done it yet point to a likely good Q4 report, which they expect to positively affect the share price before raising money?
A bit OT, but just to follow-up quickly on the post above, the cash generation (cash flow) from sales to leasing partners over the past five quarters has been $930M:
Q3 2016 $173M
Q2 2016 $143M
Q1 2016 $242M
Q4 2015 $209M
Q3 2015 $163M
(Numbers are from quarterly SH letters or earnings reports).
But due to lease accounting, only a small fraction of this cash generation is recognized and contributes to the bottom line in the quarter when the vehicle is leased. In fact, this "cash generation" is not even formally recognized as "cash flow," so it seems to get discounted by the market even though it is actual dollars in the door. And the same is true for top line revenue and profits/earnings.
If I am understanding @neroden's post correctly, the accounting change described could have a huge impact on 2017 financials and also significantly increase reported cash flow. If they are able to do this for 2017, I can't imagine why they wouldn't do it starting in Q1. Would be good to pin this down.
NannerAirCraft
Member
So Solarcity revenue will be in the Q4 report right? How much revenue do you guys see them adding?
Similar threads
- Locked
