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

Tesla, TSLA & the Investment World: the Perpetual Investors' Roundtable

This site may earn commission on affiliate links.
Creative financing idea de jour:
  • Tesla could issue new TSLA shares to the owner with every new vehicle sold, at a discount: up to 10% of the ASP (at the discretion of the new owner), with a discount of 50%. For a $50k car that's ~20 shares worth ~$5k - for a $100k car it's ~40 shares worth ~$10k.
  • The 50% cash cost has to be paid immediately based on current stock price, and the new TSLA shares will vest after 2-3 years of continuous Tesla ownership - i.e. it's like a stock option with a hefty premium paid.
  • Trade-in to Tesla would transfer option value to the new vehicle. Post purchase upgrades and leasing would count towards these discounted stock options too.
Downsides: this would hurt GAAP profits and margins due to the 50% discount that is a GAAP expense, but would actually generate quite a bit of cash: with 100,000 new vehicles per quarter that's several hundreds of millions of dollars of cash flow per quarter. With realistic take-rates of this option annual dilution would still be around the level of inflation - but this could be further refined if it gets out of hand.

There's numerous upsides: beyond improving cash margins by 5%, it would more directly involve owners as investors, and it would increase the value of the vehicles without any cash costs. It would also help the "think of this car as an investment" notion that Tesla's FSD effort is trying to promote. The 2-3 years vesting schedule would also make sure it's a long term investment.
 
Tesla has increased the size of its offering by c.$0.3-0.4bn, both the equity and converts are increased.

Tesla have issued $750m of equity and this can be scaled up to $863m by the underwriters over allotment option.
Elon has scaled his purchase up to $25m from $10m.

The convert is now expected to pay a 2.0% coupon with $310 conversion price, and to raise $1.6bn (scaleable to $1.84bn by the underwriters). $262m is expected to be spent on derivatives to prevent potential dilution from the converts (up to a share price of $608).
 
Last edited:
Creative financing idea de jour:
  • Tesla could issue new TSLA shares to the owner with every new vehicle sold, at a discount: up to 10% of the ASP (at the discretion of the new owner), with a discount of 50%. For a $50k car that's ~20 shares worth ~$5k - for a $100k car it's ~40 shares worth ~$10k.
  • The 50% cash cost has to be paid immediately based on current stock price, and the new TSLA shares will vest after 2-3 years of continuous Tesla ownership - i.e. it's like a stock option with a hefty premium paid.
  • Trade-in to Tesla would transfer option value to the new vehicle. Post purchase upgrades and leasing would count towards these discounted stock options too.
Downsides: this would hurt GAAP profits and margins due to the 50% discount that is a GAAP expense, but would actually generate quite a bit of cash: with 100,000 new vehicles per quarter that's several hundreds of millions of dollars of cash flow per quarter. With realistic take-rates of this option annual dilution would still be around the level of inflation - but this could be further refined if it gets out of hand.

There's numerous upsides: beyond improving cash margins by 5%, it would more directly involve owners as investors, and it would increase the value of the vehicles without any cash costs. It would also help the "think of this car as an investment" notion that Tesla's FSD effort is trying to promote. The 2-3 years vesting schedule would also make sure it's a long term investment.
If diluting is your answer, then they can do that at no discount. Also, best investors are those who value the company on its merit not due to discounts. Employee ESPP is different.
 
Not sure if there are differences in quality, for Tesla, gethering data during AP or in shadow mode.

Note that "shadow mode" isn't actually an Autopilot on/off property: shadow mode is new but inactive functionality of the neural network that neither Autopilot nor manual driving is going to utilize in any way.

Tesla uses "shadow mode" as an alpha testing tool: when they test new features they first introduce them as shadow mode attributes across the whole fleet, and ask the fleet to report back false positives and false negatives, and about certain sequences that the FSD developers are interested in. According to Karpathy, new neural net features go through several rounds of shadow mode testing before they ever get activated for any human driver.

As Andrej Karpathy said it during his Autonomy Day presentation:

"no humans were harmed in the process"​

Every car of the HW2+ fleet helps train the networks - even those who don't have any EAP or FSD option purchased.

And yes, this process is exponential too.
 
The convert is now expected to pay a 2.0% coupon with $310 conversion price, and to raise $1.6bn (scaleable to $1.84bn by the underwriters). $262m is expected to be spent on derivatives to prevent potential dilution from the converts (up to a share price of $608).

More money for the company is excellent news. Low conversion value but with potential dilution prevented up to a share price of $608 it really doesn't matter.
 
As Andrej Karpathy said it during his Autonomy Day presentation:

"no humans were harmed in the process
And here I thought that line was harkening back to wgen Andrej attempted to test himself against ImageNet. Labeling is painful.

Also wanted to suppoet your previous comment on FSD pricing/ revenue. I see 4 classes of Tesla:
  1. Base AP only - f0 continuing revenue
  2. FSD used for friends and family - $0
  3. FSD privately owned and used on TN - 25% of revenue generated
  4. FSD owned by Tesla on TN - priceless*

* revenue*miles*hours*days*years - tires - electricity

I see choosing the split point of internal fleet vs external sales being one of the more difficult choices Tesla will face going forward.
 
It looks like this transaction is done.
So much for TSLAQ. Tesla just raised c.$2.5bn in less than one day after its worst quarter, at peak FUD and while its share was price spiralling lower. Tesla has and has always had cash on tap whenever it needs it, at the expense of largely insignificant dilution.

generate.php
 
Robotaxi infra: maybe some of the $2.6B can be used to accelerate it.

I am guessing the cost will include,
  • a depot or distribution center for Tesla owned cars
  • Tesla owned cars (lease returns would do so maybe no cash needed)
  • Auto charging port (for starter it can be done manually by the attendant of the depot)
  • Front end (app) and backend softwares
  • on route emergency service
  • Paid drivers initially (testing and demo period) till regulation allow driverless cab.
  • Other capital/cash will be needed just to retrofit current FSD option cars.
  • Potentially hire more engineers in AP team to validate the extreme corner cases (race of 9s)