:
"(Reuters) - Tesla Motors Inc(TSLA) said on Friday it must repay the principle on $411 million of 2018 convertible notes and expects to do so in the third quarter, adding to the cash pressure on the electric vehicle company.
"During the third quarter, we will be using substantial amounts of cash in connection with conversions of our 2018 Notes and we could pursue other actions to reduce our outstanding balance of convertible notes, which could require further outlays of cash," Tesla wrote in its quarterly filing with the U.S. Securities and Exchange Commission."
This "analysis" makes no sense. The 2018 notes are deep-in-the-money and convertible to stock at the option of the holder. The holders would be stupid to do anything other than convert them to stock. (Now I go do some more research...)
OK, so I dug into the convertibles and they're *super* complicated. They're not normal convertibles. They've got a weird partial cash payment thing going on. It's specified in a weird way, but it's economically (though not legally or for tax purposes) equivalent to this:
-- first the bond is converted to stock at the conversion rate
-- then a price is calculated based on the volume-weighted average price over 20 days
-- and an amount of stock equal to the face value of the bonds is repurchased by Tesla in cash at that price
-- Tesla then has the option to repurchase more of the stock, up to all of it, in cash as that same price
So Tesla is obligated to pay the face value back in cash, but any excess conversion value turns into stock. Unless Tesla decides to pay it off in cash, which they can. If they do that it's economically equivalent to a stock buyback, and is antidilutive.
Worth noting there are $660 million of the 2018 convertibles. If they only have to repay the principal on $411 million, then not everyone converted.
The $660 million converts to 5.3 million shares. Tesla set up a hedge transaction where they have an option to buy 5.3 million shares at $124.52, but they sold warrants (essentially options to buy but a bit different) to third parties to buy 5.3 million shares at 184.48. It's not clear but both transactions probably only trigger when the convertibles are converted. It looks like the warrants expire in March 2018, but they might not, I haven't seen their prospectus (and they seem not to be publicly traded so there may not be one).
Anyway, if the warrants are triggered *and* Tesla exercises their antidilutive options, Tesla realizes $317.788 million in cash. This happens in March 2018 at the latest. The maximum cash repayment requirement might be $660 million; subtracting the $317.788, the maximum real exposure is $342.212 million. Of course Tesla could choose to use additional cash to prevent the dilution (like a share buyback) but they wouldn't do so if there was a cash crunch.
"Fully diluted" calculations have been assuming that the 2018 convertibles, which are in-the-money, would be converted. If they are paid off in cash, it's antidilutive (and I have to go redo my valuation spreadsheet
)
...I'm not sure when those warrants will get exercised, but if they aren't going to be exercised before March 2018, and I were a bank, I'd lend against the expected income from them.
It's also possible that Tesla could simply politely request that the warrant-holders exercise their warrants. Or Tesla may even have a clause allowing them to force the warrant-holders to exercise -- like I say I haven't seen the terms.
Alternatively, Tesla could issue 5.3 million shares at $230 or $220 (dilutive) and use a portion of the proceeds to buy back the warrants for 5.3 million shares (which would be anti-dilutive, since the warrants are already considered dilution). The result would be non-dilutive, essentially accelerating the receipt of the cash by paying the time value of the warrant. The time value is about ~$18.50/share based on current Jan 2018 call prices, so about $98 million. Tesla could retain the option to buy at $124.52 and not exercise it until later. In short, Tesla could probably collect ~$880 million in cash now *without diluting the stock*, at the long-term expense of maybe $98 million. (If the warrant-holders don't want to sell, the same economic result can be accomplished by buying call options from a different bank.) This is enough to pay off the entire principal on the $660 million in convertibles and generate additional cash. (The convertible-holders will, however, get stock; paying them entirely in cash would cost $1.1 - $1.2 billion)
It's possible, of course, that the short-termers in the market would be stupid enough to think such a move of issuing stock and buying back the same amount in warrants was dilutive (it's not).