Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

2017Q1 results

This site may earn commission on affiliate links.

schonelucht

Well-Known Member
Mar 10, 2014
5,080
9,788
Nederland
This thread discusses 2017Q1 results. To start the discussion of, here is my model for profit for the automotive side of TSLA.

To stress again, the below is strictly for the automotive side of TSLA. The energy side is just too complicated and in flux for me to model. Some of the issues : the complicated nature of SCTY's balance sheet before the merger, the convoluted way results were necessarily reported in the 'broken' Q4 quarter (pre/post merger), the lack of visibility on the ramp up and the unknown impact of taking the gigafactory in operation.

So lets start. First off, I will ignore service revenue and costs. It's a stated goal of management that service is not a profit center and it never has been (sometimes losing a few millions, sometimes running at a profit of a few millions). The easiest way to deal with this, is thus not dealing with it at all since impact on profit will me immaterial.

Revenue and costs are broken down between straight sales and leasing. The easiest part is straight sales. This quarter Tesla sold 25 000 cars, last quarter 22 200. Assuming the mix between lease and straight sale remained more or less the same, numbers should be multiplied by 25 000/22 000 or +12.6%. Revenue was 1 739M last quarter, becomes 1 958M this quarter. Add to that the recognition of EAP revenue for 10 000 cars last quarter and 20 000 cars this quarter @5k gives an additional 150M revenue. Cost of goods was 1 373M becomes 1 546M. Total profit in Q1 on direct sales 562M

The above paragraph also assumes that the margins and ASP were more or less comparable between the quarters. I think they were. We had the P100D last quarter and the 100D this quarter as the latest hot thing depressing ASP+margins. There was also the shutdown which likely had Tesla running production a little less efficiently before and after to play catchup and switch things around. There is a general trend for Tesla to become more cost effective as it gets better at manufacturing but I think that effect has largely ran its course for the S and X which are now a mature product. I also think Tesla's focus is now 100% on Model 3 instead of optimizing efficiency of the S/X production line.

On to leasing. These numbers are a little harder because they depend on cumulative sales. In Q4 total leasing net value assets (standing at 3 143M) increased by 184M. Assuming net value increase gained the same +12.6% from overall quarterly deliveries Q1 total leasing net value increase was 207M or +6.6% from a base of 3 134M. We will use this number as the increase for leasing revenue and cost. In Q4 leasing revenue was 255M of which 58M was recognized revenue from expiring resale value guarantees. That leaves us with 197M sales revenue last quarter, becomes 210M this quarter. Cost of goods was 172M becomes 194M. Gross profit of 16M.

To estimate how much resale value guarantees Tesla can recognize this quarter I just took it proportional to the amount of expiring guarantees in the next period (+29%) So 58M becomes 75M.

Total gross profit 562M + 16M + 75M = 653M

Now on to the other costs. Research and development last quarter (excluding Solarcity) was 235M. Sales, general & administration was 382M and interest expense was 43M. Taking a flat R&D and +10% for both SG&A and interest expense we get to 702M.

Final verdict : loss of 49M on the automotive side of business.

Possiblue upsides : every 1% more efficient manufacturing on the Model X increases profit by 6M (9M for the S). ZEV credits can do anything from +0M to +50M. Possible downsides : not fully recognizing EAP may reduce profits by anything from 0M to -50M (it is hard to believe they will already not recognize majority of the features)

What do you guys think? Reasonable? Unreasonable? Where did I go wrong?
 
There are so many unknowns I'm not even bothering to guess. I think Tesla Energy will show a substantial positive suprise -- all those big Powerpack installations have to add up, and I still suspect Tesla has *extremely* high gross margins on them (they've been really secretive about their gross margins on that business, I think they're saving it until it has enough volume that the fixed costs aren't messing the numbers up). We'll see the first results of the expenses taken to merge SolarCity, which should hurt, and probably won't see any of the merger synergies yet, though we might get a clue as to what they're going to be for Q3. The accounting for PPAs/leases remains a hairball though it looks to me like they're trying to untangle it. Expenditures at Riverbend and at the Gigafactory are a wild card. I expect to see a fair amount of refinancing of SolarCity's high-interest-rate debt at lower rates (there was already some of this in Q4 last year), which could look bad or good in Q1 though it's good for the future. Volumes at SolarCity should continue to drop as the focus is on profitability, but it's unclear whether overhead will drop enough in Q1 to actually make the profitability visible... I'll be watching for whether the cash-and-bank-loan percentage of sales goes up.

I think your estimates on the Automotive side are reasonable but there are so many other things which could swing the results by over $100 million that it's just not worth guessing. I'm just hoping that the Q1 report will give us some clarity on the post-merger SolarCity situation so that I can make a reasonable projection regarding Q2!
 
Thanks for the thought and great work. Unfortunately I think Gigafactory expansion will swamp everything else, and we can't predict what that cost, so... break even +/- 5%...

I am terrible at accounting rules, but wouldn't this capex spending affect cash flow but not affect profitability?

Also, to the original post, the new accounting rules can now (optionally) go into effect, where TSLA can recognise the revenue of the leased vehicles now, iirc. I don't see that accounting change being reflected.
 
There is a general trend for Tesla to become more cost effective as it gets better at manufacturing but I think that effect has largely run its course for the S and X which are now a mature product. I also think Tesla's focus is now 100% on Model 3 instead of optimizing efficiency of the S/X production line.

Good job! Thanks!

In the context of the Q1 ER you might be correct, but in the bigger picture I think that is incorrect.

1. Some of the M3 volume volume will reduce the MS-MX costs, AP hardware and software for example. Some of those savings could impact their Q1 results but probably in a minor way for now.

2. I believe that they will eventually make some of the alien dreadnaught improvements to the MS-MX production, automated wiring harness installation for example.

3. And battery costs. At some point they will use 2170's produced at a Gigafactory, or at a similar cost at a Panasonic facility using the same equipment and vertical integration (probably at a savings of over 55%). That will also coincide with an increase in capacity of at least 7.6% (cheaper and better:D).
 
Last edited:
My prediction is that they will not recognize any revenue from EAP and of course none from FSD.

There is my theory on why from last week:

2017 Investor Roundtable:General Discussion

I have a theory about EAP revenue and Deliveries. I have a feeling they will not recognize any of the EAP revenue in Q1. I believe this will be in-part because they will have a nice surprise to the upside on deliveries. Mostly because the expectations had been lowered with the shutdown for retooling/maintenance in Feb. I think deliveries will come in very strong due to the overhang of 6000+ cars in transit and the much talked about 2000+ cars that missed being delivered in Q4 by a day. They said they added production days, probably weekends and 3rd shift to compensate, so deliveries could come in north of 24k and maybe even north of 25k based on some of the VIN data and international tracking. There has also been a big push to deliver cars to China to take advantage of some expiring incentives. I think they will actually hold the EAP money for Q2 when they do not have such a huge overhang and lowered expectations. The EAP number will also be much bigger by then and EAP will have easily exceeded AP1 capabilities, which will remove any doubts about whether they should be recognizing the revenues. If they recognize those revs or even part of them in Q1, there might be questions about whether they should be doing that with only 2 cameras active and no AP1 parity. I believe Q2 deliveries will be the ones that lag, mostly due to the ramp for Model 3 production. Tesla is planning on almost doubling the number of cars they produce per week from July - Sept. They are going to have to hire a lot of people and get them trained in Q2.
 
  • Like
Reactions: neroden
My prediction is that they will not recognize any revenue from EAP and of course none from FSD.

There is my theory on why from last week:

2017 Investor Roundtable:General Discussion

Another theory: while it may be rougher than AP1, doesn't AP2 now have all the bullet point features? Auto lane change, summon, autopark, high speed autosteer etc. One could maybe argue that just because they aren't using 4 cameras, if they get the job done should it matter? (Also I am annoyed they didn't just write in a software check to throw an error if 4 cameras were blocked-- that is all anyone can tell anyway. The fact that they didn't actually supports your theory).

Edit: Feature parity still missing: auto headlights and wipers, per Electrek. Still pretty minor.
 
Last edited:
Another theory: while it may be rougher than AP1, doesn't AP2 now have all the bullet point features? Auto lane change, summon, autopark, high speed autosteer etc. One could maybe argue that just because they aren't using 4 cameras, if they get the job done should it matter? (Also I am annoyed they didn't just write in a software check to throw an error if 4 cameras were blocked-- that is all anyone can tell anyway. The fact that they didn't actually supports your theory).

Seems like too much of a coincidence that they squeezed in 8.1 and autopilot updates for HW2 users yet not HW1 drivers before end of March.
 
Another theory: while it may be rougher than AP1, doesn't AP2 now have all the bullet point features? Auto lane change, summon, autopark, high speed autosteer etc. One could maybe argue that just because they aren't using 4 cameras, if they get the job done should it matter? (Also I am annoyed they didn't just write in a software check to throw an error if 4 cameras were blocked-- that is all anyone can tell anyway. The fact that they didn't actually supports your theory).

They could make a case or recognition in Q1, but why? They had record deliveries in Q1. That was part of my theory. If Q1 was over 25k, which I nailed, then they would not bother with recognition of the revenues until Q2. They could also have some good ZEV credit revs in Q1, so they might be saving it for when they are using 4 cameras and do have those last couple of promised features. There is no risk in waiting, but if they take credit for to much now, they run the risk of being seen as being somewhat desperate. Everyone knows that money is there and everyone knows they are close.

I think they will push to get AP2 beyond parity in Q2 and recognize the revenues then. That is just my theory.
 
I am terrible at accounting rules, but wouldn't this capex spending affect cash flow but not affect profitability?

Also, to the original post, the new accounting rules can now (optionally) go into effect, where TSLA can recognise the revenue of the leased vehicles now, iirc. I don't see that accounting change being reflected.
They're *probably* not adopting it early. No definitive statement but from the vague statements in the last SCTY and TSLA annual reports, it looks like they aren't going to adopt those standards until next year.
 
Gigafactory expansion should be capex and doesn't affect directly cost or revenue or profits

I am terrible at accounting rules, but wouldn't this capex spending affect cash flow but not affect profitability?

Some assets depreciate linearly in time from the moment you put them to use. When those assets are scaled to deliver 100GWh/year but in ramp up mode are used to turn out just 1GWh in a quarter, then the depreciation cost (added as cost of goods) is going to be massive relative to your small production. Meaning gross margin suffers. That is why Tesla Energy essentially is not turning a profit at all, despite calculations from some here that gross margins should be enormous : those posters fail to acknowledge fixed costs from depreciating assets.
 
Yeah. One of the issues here is that depreciation is bullshit.

OK. Let me be more specific. The concept of depreciation is great. But the half-assed linear models of depreciation used by accounting standards are typically bullshit. To do a proper economic model you must know the *true* depreciation curve for your assets. There's actually some discussion of this in Buffett's most recent letter...

So this will in fact affect reported profitability, but not so much real profitability.

Solar panel manufacturing equipment depreciates much *faster* than the lifetimes used and has much *less* salvage value. The battery manufacturing equipment? Some of it's probably like that too, but I'm guessing a lot of it depreciates *slower* than the lifeitmes used and has *more* salvage value, because battery manufacturing doesn't change as fast.

Anyway, using the half-assed standards for depreciation, you will indeed see low reported gross profit margins early on due to goofy depreciation schedules -- also due to the fact that you have to hire people before the actual production starts (so you have an overhang of operating costs right at the beginning) -- and other things which are essentially nonsense from a long-term-profit point of view.
 
i think the closer comparison is the q3 2016 quarter, especially since units are closer to what was produced then. also in q4 16 commentary tesla guided gaap and non-gaap gross margin to q3 2016 levels (30%).

from q3 2016 i think you add 1% for units and 2% for mix shift towards model x, take away 1% for forex movement.
this gets me very close to your auto revenue number: 1,917 x 1.01 x 1.02 x 0.99 = 1,955
i don't have a good model for lease revenue so i will just take q3 2016 + 1%: 231 x 1.01 = 233
total automotive revenue: 1,955 + 233 = 2,188
29.4% gross margin x 2,188 = 643 automotive gross profit

let me stop there because that number is 80 million higher than your figure of 562 million in gross profit.

quote from q4 2016 letter:
" In addition, both GAAP and non-GAAP automotive gross margin should recover in Q1 to Q3 2016 levels and then continue to expand in Q2 2017. "

also the guidance implies gaap and non-gaap gross profit to match q3 2016, i see that as hard to do without having meaningful zev credit sales again.

your thoughts?

This thread discusses 2017Q1 results. To start the discussion of, here is my model for profit for the automotive side of TSLA.

To stress again, the below is strictly for the automotive side of TSLA. The energy side is just too complicated and in flux for me to model. Some of the issues : the complicated nature of SCTY's balance sheet before the merger, the convoluted way results were necessarily reported in the 'broken' Q4 quarter (pre/post merger), the lack of visibility on the ramp up and the unknown impact of taking the gigafactory in operation.

So lets start. First off, I will ignore service revenue and costs. It's a stated goal of management that service is not a profit center and it never has been (sometimes losing a few millions, sometimes running at a profit of a few millions). The easiest way to deal with this, is thus not dealing with it at all since impact on profit will me immaterial.

Revenue and costs are broken down between straight sales and leasing. The easiest part is straight sales. This quarter Tesla sold 25 000 cars, last quarter 22 200. Assuming the mix between lease and straight sale remained more or less the same, numbers should be multiplied by 25 000/22 000 or +12.6%. Revenue was 1 739M last quarter, becomes 1 958M this quarter. Add to that the recognition of EAP revenue for 10 000 cars last quarter and 20 000 cars this quarter @5k gives an additional 150M revenue. Cost of goods was 1 373M becomes 1 546M. Total profit in Q1 on direct sales 562M

The above paragraph also assumes that the margins and ASP were more or less comparable between the quarters. I think they were. We had the P100D last quarter and the 100D this quarter as the latest hot thing depressing ASP+margins. There was also the shutdown which likely had Tesla running production a little less efficiently before and after to play catchup and switch things around. There is a general trend for Tesla to become more cost effective as it gets better at manufacturing but I think that effect has largely ran its course for the S and X which are now a mature product. I also think Tesla's focus is now 100% on Model 3 instead of optimizing efficiency of the S/X production line.

On to leasing. These numbers are a little harder because they depend on cumulative sales. In Q4 total leasing net value assets (standing at 3 143M) increased by 184M. Assuming net value increase gained the same +12.6% from overall quarterly deliveries Q1 total leasing net value increase was 207M or +6.6% from a base of 3 134M. We will use this number as the increase for leasing revenue and cost. In Q4 leasing revenue was 255M of which 58M was recognized revenue from expiring resale value guarantees. That leaves us with 197M sales revenue last quarter, becomes 210M this quarter. Cost of goods was 172M becomes 194M. Gross profit of 16M.

To estimate how much resale value guarantees Tesla can recognize this quarter I just took it proportional to the amount of expiring guarantees in the next period (+29%) So 58M becomes 75M.

Total gross profit 562M + 16M + 75M = 653M

Now on to the other costs. Research and development last quarter (excluding Solarcity) was 235M. Sales, general & administration was 382M and interest expense was 43M. Taking a flat R&D and +10% for both SG&A and interest expense we get to 702M.

Final verdict : loss of 49M on the automotive side of business.

Possiblue upsides : every 1% more efficient manufacturing on the Model X increases profit by 6M (9M for the S). ZEV credits can do anything from +0M to +50M. Possible downsides : not fully recognizing EAP may reduce profits by anything from 0M to -50M (it is hard to believe they will already not recognize majority of the features)

What do you guys think? Reasonable? Unreasonable? Where did I go wrong?
 
Last edited:
let me stop there because that number is 80 million higher than your figure of 562 million in gross profit.

Sounds reasonable and explainable for two reasons :

1) I didn't include any improvements on gross margin versus Q4, while you added (net) 2% from Q3 levels. While it's certainly possible that margins improved, the company has historically been too optimistic on its margin guidance. So I have a 'believe it when I see it' attitude.

2) Q3 included a lot of fully loaded (signature) X's with high ASP and thus fatter gross margins. This is likely not repeating itself this quarter.
 
  • Like
Reactions: neroden
it's hard for me to imagine they don't know their gross margin within 1-2% in february for a period that ends in march.

true on signatures, but also they produced 50% more x this quarter meaning they have become much more efficient at producing that vehicle.

Sounds reasonable and explainable for two reasons :

1) I didn't include any improvements on gross margin versus Q4, while you added (net) 2% from Q3 levels. While it's certainly possible that margins improved, the company has historically been too optimistic on its margin guidance. So I have a 'believe it when I see it' attitude.

2) Q3 included a lot of fully loaded (signature) X's with high ASP and thus fatter gross margins. This is likely not repeating itself this quarter.
 
Tesla’s SolarCity sells $241 million equity in its solar portfolio

On the solar side, it looks like this $241million bundled PPA sale in Electrek article from Dec 23rd was too late to get booked in Q4...should be a juicy bottom line addition in Q1. Based on Jason's direct response in the Q4 ER call, at the 1 hour mark:
"Jeffrey Osborne - Cowen & Co. LLC
Hey, good afternoon. I appreciate you squeezing me in. Two quick ones. One, was there a solar securitization in the quarter? And if so, how large was it? [snip]...
Jason S. Wheeler - Tesla, Inc.
Quick answer here, no, there's no solar securitization in Q4. There's one in Q1."
 
it's hard for me to imagine they don't know their gross margin within 1-2% in february for a period that ends in march. true on signatures, but also they produced 50% more x this quarter meaning they have become much more efficient at producing that vehicle.

Certainly. Maybe assigning a 0% chance to better margins happening as I did was wrong and you make a strong case. We will see!

On the solar side, it looks like this $241million bundled PPA sale in Electrek article from Dec 23rd was too late to get booked in Q4...should be a juicy bottom line addition in Q1.

How does such a securization work on the balance sheet? The cost of goods for said panels are already booked I assume? Am I right in thinking that it moves a bunch of deferred revenue of the books together with a reservation for future maintenance etc and the difference between both is added as straight profit? If so, what order of magnitude would you estimate the profit for this particular sale?
 
How does such a securization work on the balance sheet? The cost of goods for said panels are already booked I assume? Am I right in thinking that it moves a bunch of deferred revenue of the books together with a reservation for future maintenance etc and the difference between both is added as straight profit? If so, what order of magnitude would you estimate the profit for this particular sale?
No clue, but I found this press release from Sammons, which says the financing is non-recourse to Solar City. It also said the transaction was completed Dec 16th...so how was it not booked in Tesla's Q4? Was Jason referring to something else? Couldn't find any Tesla press release on the transaction.

Sammons Renewable Energy Leads $241 Million Solar Cash Equity Transaction with SolarCity - PR.com