General Discussion: 2018 Investor Roundtable

[neroden]

No plan survives contact with the enemy. When you are burning $9,000/minute and shorts trying to cut off your access to capital - you do what you have to do. Opinions of your CFO, family/friends take a back seat to survival.

You just need to figure out if you would rather go with $49,000 - $7,500 or $35,000 - $1,750.

I assure you the LR RWD PUP Model 3 is an excellent deal, even if its not $35,000. That $35,000 - $7,500 was a fantasy (always was) and I got 100+ disagrees over that point when I said Tesla would lose its shirt trying to make them this year.

If Tesla follows what I think to be the Pareto efficient solution -
200,000th car gets delivered on July 1st.
Tesla gets 24 weeks of Model 3 production at 5,000 cars = 120,000.

Best outcome is 120,000 P Model 3s if the demand is there, but it will be some kind of mix with minimum ASP of $60,000.

I got $60,000 as its unlikely people going in at $54,000K (LR AWD+PUP) are not picking up EAP at the minimum.

While I *mostly* agree with you I am quite sure that Tesla is going to start pushing straight for 10K as soon as they reach 5K. Although this does require CapEx, it should be possible to still be cash-flow-positive and profitable for Q3 and Q4, and obviously the more cars they can put out in the second half of the year, the better. (And the more people who get the full tax credit.) I think there will be more than 120,000 Model 3s produced in the second half of the year, though I'm not sure how much more.

 

[RobStark]

Yah, the pickup version of a semi. I think it looks pretty identical, other than front wheels are turned and render angle is a little different.

To me this doesn't look Semi sized.

It looks a bit longer than an F-150 but about 6-10" taller.

 

[neroden]

I understand your frustration. Thanks for sharing your situation - it seems you may not have the tax liability to take advantage of the credits. You want the hottest car, want it fast and want it cheap - not an easy objective to achieve.

The best solution I can think of is getting the current crop of Model 3s second hand by end of the year. Since the tax credit is in full effect, sellers essentially have to discount by at least that much to be competitive with new sales. So that will be roughly 40K plus tax. Only 5K more than you were hoping for and you LR + PUP as part of the deal.

This worked for Model S. For some reason there's always someone who buys the car and flips it within a few months for (initial cost minus tax credit minus a bit more). I have no idea who these people are. It seems financially extremely unsound.

 

[mongo]

Basically what I'm saying is that the initial buffer buildup is inherently a one-time delay in production; does that make sense? This is also why I said that the financing hit was insignificant; how much does a 10 day delay in production matter anyway?

Yah!
But if they gain inventory by supplier excess and not intentional production reduction, the buffer is free.
10 days of revenue is 400 million (5k/wk @ 56k ASP) ~80 million gross? Wow... at that rate, the 3 is 14 billion in revenue over a 50 week year....

 

[mongo]

To me this doesn't look Semi sized.

It looks a bit longer than an F-150 but about 6-10" taller.

Oh, I don't think they went semi dimensions, that would be heck to park. Just the overall design feel. (And we are talking about the truck on the projector screen, right?)

 

[Boomer19]

This worked for Model S. For some reason there's always someone who buys the car and flips it within a few months for (initial cost minus tax credit minus a bit more). I have no idea who these people are. It seems financially extremely unsound.

some people just can, but at least some people can also make out ok on that kind of deal. and of it worked for the S, there’s a higher probability it can happen with the 3 @ggies07

 

[generalenthu]

Folks reporting configurator opening up for them. It's getting crazy all around. Tesla might likely announce they're sold out for the year in their July delivery report and will prioritize the dual motor. ASP is going to surprise to the upside in q3 and q4.

 

[neroden]

Well, allow me to specify that JIT is a term that does not necessarily mean the moment the part arrives its picked used and put into production.

OK

Therefore your statement does not conflict with my other post stating than the lower the amount of buffer is the better for production flow, throughput and tied up capital. That balance is determined by the level of automation you install and synchronization of the different production steps.

In an auto production there is not such thing as pure JIT although its often claimed. If the two of us would walk into Fremont we will find places where parts are picked up from robots where you have small storage to make sure you have seamless supply. What we likely won't find is larger storages where material needs to be shifted to the line into that buffer to be put into production. That is what I am referring to.

All right, fair enough! I guess we don't really disagree...

When I first learned about JIT it was decades ago in the context of companies trying to absolutely minimize their inventory in stock of any parts whatsoever, and was being done for financial reasons, not operational -- specifically for capital efficiency. Squeezing pennies out by not having any inventory buffers. These were steady-state companies which had already synchronized the production rate of the whole supply chain.

There really are companies which make a point of having the parts arrive by truck the morning they're needed, which is really risky. You want to have at *least* a few days of buffer in case there's a catastrophe which closes the roads, etc. And more buffer for longer / riskier supply chains. That's the sort of JIT I was thinking of.

Companies which do this sort of extreme JIT have problems like the one Musk talked about where the floor mats were held up by a shootout at the border. This is the sort of thing which Musk is trying to get enough of a buffer to avoid. I think there's probably some pretty large raw material buffers at both Fremont and the Gigafactory, and I suspect they have large buffers for any parts delivered from overseas.

Musk's deliberate integration of processes usually done in different factories into one factory allows for a substantial buffer reduction by elininating those transportation-risk points.

The tied up capital part of production with buffers is not the biggest challenge but the so called "whip saw effect" triggered by buffers that makes your operation very difficult to manage.

Don't want to go deeper here but you may have picked that I worked in this environments....

Clearly you do want to synchronize the production speed of every element, regardless, to avoid "on-off" whipsawing. The buffer size is a matter of protecting against supply chain disruption. Companies with too-small buffers are very subject to supply chain disruption.

 

[neroden]

My thoughts.

1) The pickup truck is a distraction. Not sure why Elon felt the need to bring it up now. Such a product is years away (and behind a line of at least three other vehicles, all delayed already from our more rosy timelines not even 6 months ago). If it isn't an effort to prop up the stock then it looks suspiciously timed like one.

Nah, I think he's just getting ready to go into actual design. The semi and Model Y designs are probably into advanced feature-locked-down states, and the pickup is the only thing for the "blue sky" first-stage designers to work on at this point. So he's crowdsourcing features.

 

[neroden]

If the supplier's terms are net 60, and you sell the car built with the parts on the 59th day, is there any financing or capital involved?

OK, you seem to want to understand this. To try to explain it, let me flip it around. So, arguably in your scenario, there's no financing involved. But if you sell the car on the first day, you effectively have interest-free financing for 59 days, which you can use to pay for *something else*.

 

[Waiting4M3]

Could the video Elon tweeted a few days ago ("Vicious torque") be the pickup?

Elon Musk on Twitter

 

[Bet TSLA]

Sure, if someone is selling their used M3 for 35k I would gladly take that offer. Not sure people will sell it for that price though.....

You can buy a used Model S for under $35K. I highly recommend doing so while you're waiting for your Model 3. It won't depreciate much, and it's possible you'll decide you want to keep it instead of buying a 3. The only mistake you can make when buying a Tesla is waiting longer than necessary. Well, complaining about it instead of doing something about it is a mistake too, but let's not go there any more than we already have.

 

[Boomer19]

Could the video Elon tweeted a few days ago ("Vicious torque") be the pickup?

Elon Musk on Twitter

i think @mongo pointed out that was the M3 rear right quarter. the door missing from the frame as the slant body on the right side of the video and the floor panel underneath and the notch where the back bumper goes on the left side of the video

 

[JRP3]

Could the video Elon tweeted a few days ago ("Vicious torque") be the pickup?

Elon Musk on Twitter

No it's a Model 3.

 

[mongo]

OK, you seem to want to understand this. To try to explain it, let me flip it around. So, arguably in your scenario, there's no financing involved. But if you sell the car on the first day, you effectively have interest-free financing for 59 days, which you can use to pay for *something else*.

I like understanding

You sold the car day 1, that's a plus on the revenue side. But isn't it independent of how much parts inventory you have on the shelf?

If your supplier can only make 5k a week, and you slow production to get parts inventory, that hurts the bottom line. If you can only make 5k per week and your supplier can produce 6k, you can create a buffer without hurting revenue.

My assertation: If you are making cars as fast as you can, and you are on 60 day terms with the supplier, there is no bottom line difference if you have 1 or 40 days of inventory on the shelf, but 40 days let's you ride out supply issues better.

 

[neroden]

So I don't think any of us really know what's going on at the production line (we don't even know how many they have, as witnessed from the "Is GA1 the first Model 3 line or the Model S/X line" debate). I'm also not sure it matters. Clearly some of the innovations they tried in general assembly worked, others didn't, and they're trying new ones. Model 3 was designed for simplicity of general assembly and was deliberately designed to *allow* robots to do the general assembly; maybe that wasn't entirely successful, but it's still a shorter general assembly process than most cars, apparently. They'll be able to make good production rates with decent margins, and they'll be able to set up general assembly in Europe and Asia from kits.

Paint is probably the controlling bottleneck now. I doubt we'll get any color on that... (pun intended)... unless someone asks what the current production bottlenecks are at the next quarterly conference call.

 

[neroden]

I like understanding

You sold the car day 1, that's a plus on the revenue side. But isn't it independent of how much parts inventory you have on the shelf?

If your supplier can only make 5k a week, and you slow production to get parts inventory, that hurts the bottom line. If you can only make 5k per week and your supplier can produce 6k, you can create a buffer without hurting revenue.

My assertation: If you are making cars as fast as you can, and you are on 60 day terms with the supplier, there is no bottom line difference if you have 1 or 40 days of inventory on the shelf, but 40 days let's you ride out supply issues better.

It only matters when you are *changing* the quantity of inventory on the shelf (increasing it from 0 days to 40 days, or reducing it from 40 days to 0 days). That's why I called it a one-time thing. Got it?

 

[mongo]

It only matters when you are *changing* the quantity of inventory on the shelf (increasing it from 0 days to 40 days, or reducing it from 40 days to 0 days). That's why I called it a one-time thing. Got it?

Ahhhhhh, the light comes on. Even though none of the parts payments are due untill after the car is sold, carrying more inventory increases the time the inventory is on hand, thus pulling the payment in closer to the revenue, thus reducing the advantage/ cash on hand created by the quantity produced between sell date and due date.

In other words, if you sell on the due date there is zero surplus from the 60 day terms. If you sell the day the part is recieved, then you have 60 days with the deferred part cost adding to your cash. Or in the continuous case you have an extra 60 days of part cost in cash. Whereas with 60 days stock, you have zero extra cash. Credit as credit or credit as cash....

Thanks!

 

[MXWing]

While I *mostly* agree with you I am quite sure that Tesla is going to start pushing straight for 10K as soon as they reach 5K. Although this does require CapEx, it should be possible to still be cash-flow-positive and profitable for Q3 and Q4, and obviously the more cars they can put out in the second half of the year, the better. (And the more people who get the full tax credit.) I think there will be more than 120,000 Model 3s produced in the second half of the year, though I'm not sure how much more.

Pushing for 10K would make sense given only 2 quarters of full $7500 credit.

Capex gets paid for by not diluting in equities or raising through debt.

Musk is raising capital by "forcing" Model 3 owners to click that awd/performance box. They are also "forcing" clicks to FSD now to save $2000 or more in the future.

Besides, what's the cost of a tent? I mean $80 bucks a piece at Walmart?

 

[UnknownSoldier]

I know you're joking but it turns out those Sprung tents aren't cheap at all. I didn't realize it until I browsed around that company's website.