Thread by @ChrisBloomstran

Brett @wintonARK, Your taking the time to reply is again appreciated. I was going to let @ARKInvest's 2025 $TSLA $3,000 price target and your replies go but in thinking over the past several days a few of your answers just don’t resonate. Comments and a proposition follow...1/

While most of your replies lacked easily quantifiable or direct answers to my questions, your offhand response about Tesla’s expected 2025 share count suggests extremes of either promotional intent or lack of understanding. Perhaps it's neither and I'm missing something. 2/

The ARK model completely ignores coming dilution from the exercise of 165m option & RSU shares already issued (excluding any subsequent grants). This is a BIG deal. I guarantee you @elonmusk gets it. Seeing an even 1B shares outstanding in ARK's 2025 projection was baffling. 3/

Your reply to me was stunning: “Good catch re: the options and RSUs as well as just leaving the balance sheet in cash (above deployable Capex since in most cases they are bandwidth rather than capital constrained.) WILL LIKELY ADJUST TO DEPLOY EXCESS CASH INTO SOME BUYBACKS...4/

WE WILL UPDATE THESE LINES.” What??? Let’s be clear here. Holding your assumed share count at the projected 1 billion shares in 2025, to correct the oversight you are NOW ASSUMING Tesla will REPURCHASE 125 million shares to offset the dilution from option/RSU exercise? /5

Are you kidding me? Do you know how much money that is? In your replies to me you assume/infer:

1. 125 million shares repurchased between today’s price & ARK's $3,000 target. That’s a capital requirement of more than $80B at today’s share price and $375B at ARK's 2025 target. /6

Let’s use a midpoint of $228B. For perspective, Tesla raised $40B in cumulative capital from inception and now has $19.4B in cash, $12.7B of which was raised last year alone in “at-the-money” sales to those willing to pay extraordinary prices relative to fundamental metrics. /7

Your model and cash requirements were already impossibly stretched and now to fix the problem you are going to just whip up $228B for repurchases out of thin air? That happens to be the current market cap of $XOM and seven times Tesla’s trailing full-year 2020 revenues. /8

2. You acknowledge Tesla will need more factories, but that, “the model invests maximum cash that can be effectively deployed into EV factories (subject to scaling bandwidth constraints) and debt funds those factories at some level (60% varies)." Bandwidth constraints? 9/

"In most cases they are scaling bandwidth rather than capital constrained.” I’m at another loss here. You assume NO additional equity cap raise so cash flow and a mix of debt will finance all EV growth? Zero additional dilution? I just can’t believe it but, OK. However…10/

3. You clearly state in your reply that Tesla will be underwriting its own auto insurance by the beginning of 2023. I’ve already pointed out that to get to your assumed $23B in premiums by 2025, at least $8B in EQUITY capital will be required to underwrite at that rate. 11/

You also replied that Tesla would be, “insuring US-only” and that Tesla would directly underwrite auto on, “40% of EV production.” Just think about the mathematical impossibility here. The ARK “BEAR CASE” assumes 5 million GLOBAL EV unit sales in 2025. 12/

Let’s say unit sales are 3, 4 and 5 million from 2023-2025. That’s 12m global units. What % are US sales? Allow me to allow you to contradict yourself and assume $TSLA writes 40% of ALL units, 12m vehicles sold in the 3 years from which they begin underwriting. Crazy Train. 13/

40% of that global number is 4.8 million vehicles. That’s $4,792 in annual premium per vehicle. Back out non-US sales by half and premiums per vehicle double to $9,584 per car. Guess what my more expensive Mercedes SUV than your assumed average Tesla price EV costs to insure? 14/

Hint - it’s WAY less. I’m not going to again get into ARK's assumed 2025 pre-tax $9.2B insurance profit (implying on the order of a 100% return on equity). The forecast note and your replies demonstrate a lack of understanding about underwriting, allowed pricing and profit. 15/

Auto insurance profitability comes from regulated underwriting and investment income/gain. Your observations about “loss ratios” make no sense, ignore expenses, and fail to incorporate the required mechanics of an insurance underwriting operation. 16/

Unless you are a low-cost, well-capitalized provider its a terrible business. So Tesla is disrupting auto manufacturing, auto insurance, ride hail and energy, all marginal at best industries, save the best operators? But let’s revert back and stick with capital requirements. 17/

4. You acknowledge that capital must be “reserved against underwriting and this in not explicitly called out. Might build this in.” Might? Brett, you cannot have an underwriting operation without capital and that capital must be ring fenced. Auto insurance is brutal. 18/

Of course, insurers can under reserve as GE did in its reinsurance operation for years and send unrealistic dividends to the parent. In auto, however, there’s no way. Auto is extremely short-tail. Two-thirds of losses are paid in year one and 80% by year two of an accident. 19/

Your words again: “In almost all cases they have undeployed cash well in excess of that required to fund the insurance business. So not a material growth restriction.” Not material? $TSLA will need at least HALF of current cash and almost half of equity to play underwriter. 20/

Further suggesting lower CAC, better underwriting expertise & better retention is misguided. Do you know what’s been done on the tech side of underwriting with telematics (when allowed), the degree to which advertising scales or what percentage of auto policies are retained? /21

When I read about vertical integration in auto insurance I think about robbing Peter to pay Paul. The insurance commissioners aren’t stupid. What percentage of $PGR and GEICO policies are retained each year? $TSLA will be better? How about years of actuarial data? Medical? /22

5. Your next comment/projection about cash and capital is one of the most bizarre things I’ve ever read: “I”m touched that you’re concerned for our credibility. In the out years they have $100s of billions on their balance sheet. (The central tendency is $200b.)" Come again? /23

You continue: "As detailed above we will likely refine the balance sheet management loop to accommodate buybacks (or bitcoin;).” This is Twilight Zone stuff. I’d love for you to clarify what “the out years” are here and when Tesla will have $200B in cash on the balance sheet. /24

That’s nearly double the cash Berkshire accumulated over 56 years - and $BRK GENERATES cash. Loads of it. $TSLA needs cash, way more of it than believed to pull off your target. I might as well say $BRK will have $1.3T in cash in 2025. Support? No, but I’m building a model…/25

Over the entire existence of Tesla, the company raised cash by selling $27.3B of common stock, $12.7B of which was raised last year at what seem to be extremely high multiples to any fundamental metric (outside of ARK’s 2025 price target). /26

Tesla also bears $14.7B in outstanding debt & preferred. Tesla’s CUMULATIVE LOSS is $5.4B. With $19.4B in cash on the balance sheet, the ARK model suggests Tesla is finished selling stock to the public and capex will entirely be funded prospectively with cash flows and debt. /27

Despite requiring an extraordinary amount of capital to grow vehicle production (the auto industry generally requires roughly a dollar of capital to produce a dollar of annual revenues), in ARK's opinion Tesla is funding secured. Not accurate? /28

$TSLA's cumulative capital raised of $40B exceeds trailing annual revenues of $31.5B. Shareholder equity is $23.7B with less than $4.7B in net cash. You tell me that the company has sufficient equity capital to grow to 5-10 million vehicles in annual 2025 production, plus.../29

capitalize a yet non-existent private passenger and commercial auto insurance underwriter that will write $23B in premiums and earn a regulatorily impossible $9.2B EBIT profit, AND have, “$100s of billions of cash on the balance sheet with a central tendency of $200B?” /30

My central tendency here is to call BS. The math doesn’t add up. The logic doesn’t add up. This is either sheer puffery or lack of understanding. Either way its bad. The students doing inexperienced write-ups and pitches from my original message get the benefit of the doubt. /31

You just can’t “miss” 11% of what will be shares out by 2025 & then simply say we’ll adjust the model to "allow for" ~$228B of REPOS for a company doing $31.5B in sales that's proven to require every dollar in capital raised to unprofitably (so far) grow into its existence. /32

Layer in seeding and building a de novo auto insurance underwriter as large as number four player Allstate is today over an impossibly short three-year period. Further layer in growing the cash balance to hundreds of billions of dollars with a “central tendency of $200B.” /33

The reasoning, whether purposefully deceptive or otherwise, is spectacularly flawed. Your answers to my questions were evasive. How again do you model $TSLA writing $23B in 2025 premium and earning $9.2B pretax? You never directly answered the direct question. /34

I’m sure the world would love to see a full set of pro forma Tesla financials for 2021-2025 supporting ARK's forecast. I know I would. The model provided to the public doesn’t do that. Let's skip Monte Carlo simulations & bull/bear cases based on some set of probabilities. /35

Give us a bear, bull and target case set of financial statements that supports your bear, bull and target case share price in the model. Income statement, balance sheet, statement of cash flows and statement of shareholders’ equity. /36

At 30x earnings by 2025 Tesla will need to earn $100B to support $ARK's $3,000 share price target, a $3T market cap after now buying back 125m shares, midrange $288B worth to offset dilution? Let’s see how the cash balance grows to its “central tendency” of $200B by then. /37

I undoubtedly come off as brutally harsh here so welcome knowing where I'm misguided. Trust me I’m far nicer on students. The model and price target ARK put out is, however, offensive to common sense. It doesn't pass the test of reasonableness. In Missouri we say, "Show Me." /38

Rather than a continued back and forth on Twitter and trying to cram a cogent case into myriad 280-character segments, how about a podcast with the two of us discussing the prospects of Tesla over the next 4 3/4 years? It should make for a healthy, enlightening discussion. /39

It would be nice for listeners to not have to piece together back and forth via this channel. Listeners would enjoy an informative, civil debate. ARK has enjoyed enormous success for which you and all there should be commended. Come defend your new price target and model. END