Anatomy of a Bubble – Tesla and Bitcoin
endlessmetrics.substack.com
endlessmetrics.substack.com
The article claims that Tesla valuation is not based on fundamentals but it fails to present any fundamental model.
Here's one model that justifies price target of $600 (i.e. much higher than current $480 price): https://twitter.com/garyblack00/status/1300540006864486404
Now, you might disagree with Gary Black's numbers and think that they are too optimistic but at least there are numbers to disagree with.
This article, sadly, lacks any quantitative or qualitative analysis to disagree with. You should probably disregard people who's only argument is shouting "bubble" and comparing anything to bitcoin.
For the past 10 years Tesla has been growing revenue at 50% Compounded Annual Growth Rate.
If they keep growing at 30-50% for the next 5 to 10 years, they'll grow into a valuation that might seem excessive today. That's a very simple and yet more often than not unappreciated math behind compounded growth. Einstein understood it. Warren Buffet understands it. The author of this article doesn't.
You don't need great analysis to see that something that has appreciated 131 times in less than a year is a bubble, unless they cured cancer or seriously undersold at their IPO. And no one undersells by that much.
What has appreciated 131 times in a year?
A significant part of their revenue is selling regulatory credits to traditional manufacturers. In essence, buying a Tesla subsidizes someone else's purchase of an Escalade.
If Tesla continues to grow they will be taking market share away from traditional manufacturers, who will thus need fewer regulatory credits. So at some point each additional sale of a Tesla will result in no net gain in profit. Tesla will have to grow well beyond that point for additional sales to equal additional profit.
Unless you think that consumers have the money and desire to purchase an Escalade and a Tesla at the same time.
Net income is traditionally defined as Revenue - Expenses.
From what I can tell, the Twitter formula is deriving net income from the market price of the stock, which is flawed.
"If it's so bad, then where's your model?"
Sometimes no model is better. Not that I don't think this article is full of it.
Well, I mean, you could say this of practically any growing business. But that _very, very_ rarely actually happens.
Seems particularly dubious in this case; most of the Tesla bull numerology seems to take as an article of faith that both demand for cars will grow quite dramatically over the next decade, and (even more dubious) that the ASP for cars will grow dramatically over the next decade.
- 10-year lead on the EV market
- Undisputed best product in the EV market (not the cheapest).
- Proprietary charging stations rolled out globally that'd take years (even a decade?) for anyone to catch up with.
- Hands down best autonomous driving system
- Diversified into solar, upcoming diesel.
The audience of Tesla fans is, not my cup of tea. But I can't help but draw some parallels to the $30-40/share Apple days when it started growing and there were similar arguments against it. But in my head I thought, well, have you actually used the laptops? There was nothing close to them.
Their first practical car was introduced two years after the Nissan Leaf.
> Proprietary charging stations rolled out globally that'd take years (even a decade?) for anyone to catch up with.
I'm not sure how well they're doing there globally. There are about 200 fast charging locations in Ireland (mostly CCS/Chademo combo) and 1000 or so 22kW AC ones. Tesla supercharging stations? Two. (They also have a bunch of destination chargers, perhaps 20 or 30). I think they're doing worse in Ireland than in most European countries, but not THAT much worse, and of course the CCS infra in many countries is far better, too.
People repeatedly report quality issues with Teslas. Not some minor stuff but stuff like rain getting inside.
> best autonomous driving system
I'd put a big question mark next to this one, because they don't use LIDAR. It's laudable that they want to build camera-only autonomous driving systems, because that makes them more affordable, but this has a cost in quality/performance.
re: auto-pilot. My recent car purchase was almost solely on picking the best autopilot. Current day, short of the Super Cruise nothing compares. I read a lot of "it doesn't actually drive itself" but as of the last few months it literally goes onramp to offramp, handling all the merges and lane changes with zero intervention reliably. It's remarkable. I don't expect anyone is close to complex traffic signals and urban driving regardless of platform.
Yes. Enters and exits freeway without intervention or confirmation. On a recent road trip there were several times where it exited one freeway, circled around, and merged onto another without any intervention at all.
They don't necessarily build the best cars.
As a tesla owner I'd say they are better bult than most American cars, on par with German but behind the Japanese.
Five years ago I would have agreed that they don't have quality cars (even though they were still the best EV). I was refusing to go near them for several years after test driving a "classic" model S with rattles and squeaks and all that, but at least in my own car (2019, model X "raven") these kind of things are non-existent.
Edit: On topic, I also think that the stock is crazy high.
Yea the stock is insane, it'll be interesting to see where it goes.
Where I work (in Norway) about 10% of the employees have teslas and there are not more issues than most fossils except for the Toyotas perhaps :)
> If they keep growing at 30-50% for the next 5 to 10 years
What makes it a bubble is the above statement you just made which sounds reasonable but is actually absurd. I'm curious how in year 10 Tesla will grow 50%. Do you have any idea how big that number is? How will that be accomplished?
Curious, I just ran math on Tesla growth at 50% for 10 years. That means Tesla's revenue will be 1.2 trillion dollars and in year 10 it will GAIN 461 billion dollars worth of revenue growth.
For comparison's sake, in 2019 Telsa revenue grew 15% (by 3 billion dollars) to 24.5 billion.
I've listened to the bulls continuously say things Tesla can double every year for the next 10 years and so on, but there is no path there. It always involves moonshots like self-driving taxis, a new eletric grid powered by Tesla batteries ect ect. All of these are super low margin businesses with heavy competition that require heavy investment with unexpected payoff.
Comparing Tesla to Bitcoin is flawed. Tesla is a publicly traded stock that produces products and is subject to financial audits. Bitcoin is completely different and can't be valued like a traditional financial asset. See Robert Shiller's discussion.
Tesla will need to make a lot of profit in the future to justify a $442 billion dollar market cap. They'll need to go from breakeven to making around $66 billion a year (for a P/E ratio of 15).
Jeremy Siegel wrote a great article in 2000 about how tech companies would need to make massive profits to justify their bubble valuations. It's still a great read.
Also, I think it may be a very long time before 15 is considered a "normal" P/E ratio again, if ever. It seems like low interest rates are here to stay.
The bubble conclusion might be right. A good bubble argument should demonstrate how the market price has completely detached from fundamentals.
You provide a good starting point for a back of the envelope calculation. Suppose you think Tesla's long term sustainable P/E is 25. Then they need to start making 18 billion a year to justify their current valuation.
Amazon revenue for the twelve months ending June 30, 2020 was $321.782B, a 27.66% increase year-over-year.
Amazon annual revenue for 2019 was $280.522B, a 20.45% increase from 2018. Amazon annual revenue for 2018 was $232.887B, a 30.93% increase from 2017. Amazon annual revenue for 2017 was $177.866B, a 30.8% increase from 2016. https://www.macrotrends.net/stocks/charts/AMZN/amazon/revenu...
By comparison Walmart’s revenue for 2018 was $500.343B and it hit $542.026B for the twelve months ending July 31, 2020. https://www.macrotrends.net/stocks/charts/WMT/walmart/revenu...
I don't know enough about the data to have an opinion. I do know that there are times in history when companies have sustained P/E ratios above 100 for many years (see Japan in the 1980s). You might be right, but from what I've seen, 100x P/E ratios usually come back down to earth.
The simplest model to demonstrate this an a fund that extracts 1% of it’s value per year. Eventually it’s total returns hits the global GDP growth rate plus inflation. Theoretically stocks should have a lower P/E due to associated risks but eventually a P/E of 20 is very possible.
Yes, but P/E ratios also relate to a lot of other factors, like human psychology
> It’s likely that as the world gains more wealth that wealth will face diminishing returns when generating more wealth.
Maybe, but my (outdated) understanding of the literature is that the equity risk premium has not decreased, even as the world has gotten a lot richer
> The simplest model to demonstrate this an a fund that extracts 1% of it’s value per year. Eventually it’s total returns hits the global GDP growth rate plus inflation. Theoretically stocks should have a lower P/E due to associated risks but eventually a P/E of 20 is very possible.
I don't understand this logic (could be right, I just don't follow it)
Number go up, so people buy. Number will go down. Don't enter a burning building. Don't try to catch a falling knife.
Yes, we all know that.
Assets go through periods of being overvalued all the time. There's a big difference between that and a bubble.
Nor does the author even mention the main difference between Tesla and Bitcoin. Tesla is a company and there are tested traditions of valuation that generally hold over time such as price to earnings ratio.
Bitcoin has no earnings. It's more like a currency or gold. Therefore, valuation is more challenging, but still possible given some work. Had that work been done, some actionable conclusions might be drawn.
The work was done, and it was irrelevant, just like with Tesla. Sure, we have really straightforward models for stock valuation, so you can do things like say "Tesla will need to return $X over the next Y years to justify this valuation at a discount rate of Z%" but that doesn't mean the particular values of X, Y and Z are even remotely reasonable.
A bubble is when asset valuation is completely unmoored from fundamentals--the "greater fool" theory dominates market behavior. At the height of the 2017 Bitcoin bubble, there were plenty who would tell you with a straight face that the fundamentals justified the valuation--this is going to be a new global reserve currency! Later price action seem to indicate that they were gravely mistaken, but why is Tesla any different right now? Just because the valuation model is easier to understand doesn't mean one's assumptions about input variables aren't completely insane.
Someone should use Siegel's thought framework with 2020 tech company numbers.
Not saying some things he has promised aren't somewhat a pie in the sky...
Unfortunately, Elon Musk is Elon Musk.
Not many are rushing to buy econobox EVs.
The more expensive cars being included in "average price of a new car" are going to include pickup trucks, minivans, and SUVs. The equivalent model here is (AIUI) Tesla Model X, whose base price would be about $80k, whose competitors are ranging more around $30k-40k base.
EDIT: (HN Throttled, can't reply, replying here)
I agree there is a premium, but I do not agree it's considerable considering a) people buy a payment and b) we're in a low interest rate environment where people can stretch for a luxury car like a Tesla due to low carrying costs of the auto debt.
Why? https://youtu.be/XCvwCcEP74Q?t=798 (13:17-14:15 is the relevant time window)
If the Fed is unwavering enough to keep its new average inflation targeting scheme credible, it should not pop (though things may still cool off at some point).
The legacy OEMs must allocate their capital to maintain current products/dividends/pensions and have a seat at the EV table in the future. Not as easy as it sounds, especially considering technology creates winner-takes-all situations and cap-ex to participate in battery production at scale appears to be in the 10s of billions.
That would surely lead to a lower valuation, if anything? Utility companies are not generally massively profitable, nor are they typically valued at hundreds of times P/E.
Given any anecdote about the economy going down and stock market going up, it is worth noting what is going on with the money supply. There is a huge amount of liquidity being pushed into the USD supply by the fed and treasury. It needs to go somewhere. The (government's) hope is that it will go into hiring/spending and the real economy. Even if that is what happens to the majority of the new money supply (debatable) a large amount will go into financial assets for people who want to preserve their wealth. The expected result (IMO) would be stocks to go up in USD purely because of demand for them. This can alse been seen as inflation or maybe financial asset inflation. The fundamentals of the economy are not good, but the new money has to go somewhere. After paying for living expenses etc, there aren't many places to put it besides the stock market.
A final thought is that if consumer price inflation remains low and financial assets catch all the inflation, the "rich" or invested benefit the most. However, its arguable that financial asset inflation without CPI is a better result for most people vs high CPI and a stock market crash. I would be interested to know if this a framework used by the Fed or just me as a random guy on the internet.
Why is that important to remember? With initiatives in the US like the Green New Deal and other various green energy pushes; there will be a huge demand for grid-connected batteries. Edit: They (Tesla) has a massively profitable grid-connected battery in Australia that's paired with wind generation.
IMO, Tesla's high price is a gamble that American politics will create a large demand for batteries. But, I wouldn't call it a bubble.
> Today alone, Tesla’s market cap soared by $64 billion in eight hours, including after hours. That’s $8 billion in “value created” per hour.
>In the second quarter, Tesla’s total revenues were $6 billion, down by 5.8% from Q2 last year. Today, its value rose by $8 billion per hour. Over the past four quarters, Tesla’s total revenues were $26 billion. Today, its value rose by $64 billion.
And the same source on the stock split:
> The stock split did the job, based on the logic that a five-dollar bill broken into five ones makes each of those ones suddenly worth $1.16 — or $1.87 if you start counting since the announcement of the split on August 11.
And I think this really puts things into perspective:
> Tesla’s shares are now valued at about 20 times annual revenues. [...] In terms of market cap, this makes Tesla the seventh most valuable US company – not counting Alibaba.
A car company that sold just 400K cars in its best year is the seventh most valuable US company. Unbelievable.
Similar statements are a staple of what people say about growth companies. Investors seem to think the prospects are way better. Your mileage may vary.
VW, the third-largest car manufacturer in the world, sold 11 million cars last year (27x Tesla), for ~$280bn revenue (11x Tesla) and ~$19bn profit.
VW made 1 out of 8 of the ~77 million cars sold in 2019.
Tesla's valuation is currently 6x that of VW's. So how exactly are investors picturing Tesla's growth here? To justify this insane evaluation, they'd need to utterly dominate the market by sales, and with an incredible profit margin.
This isn't driven by investors. This is pure speculation.
They do have more than just the cars. Tesla says they expect their energy to business to eventually surpass the EV one. They have also an insurance company.
I don't have any Tesla stock.
Everyone thought they were crazy, that it was a waste of money to emit less garbage into the air, and NO ONE stepped in to compete on EVs. Meanwhile, Tesla were simply following the science, which said we NEED EV's. As the science has progressed, many governments in Europe and some US states have massively pushed toward EV, and early adopters who are also looking at science and the performance of EVs say: I'm rich, Tesla works for me now.
Does Tesla have too high of a valuation? Maybe, but their EV's are years ahead of the competition, so maybe not. Maybe the fault here is with other automakers for completely ignoring EV's for the first decade of Tesla's domination.
This statement from the article is incorrect and doesn't account for the stock split. Had you invested that much at the opening price you'd actually have closer to $5mil today.
Actually the amount in the article is correct
One is backed by multiple factories producing vehicles, with even more factories coming online in the next year, an entire industry facing "innovator's dilemma" with a petrol & diesels new sales ban coming into force 2030-2035 for most countries.
The other, a virtual commodity not backed by any physical items, issuers or a real economy.
How are they comparable?
The existence of factories is irrelevant if those factories are not spectacularly profitable.
And right now they're barely - not spectacularly - profitable. And the innovator's dilemma applies equally to the rest of the industry, which is already making EVs of its own.
Like everyone else I have no idea if Tesla is a reasonable bet. But it's clear the price today has become completely and irrationally detached from fundamentals to the point where the value might as well be from trading virtual items.
The people investing in Tesla are the same people who invested Bitcoin. Robinhood is leading all price movement
Tesla has factories, etc. Don't forget blockchain was also going to be useful someday.
Tesla has dubious growth potential for a number of reasons. Tesla, and Musk in general, specialize in appealing to the only remaining middle class with money, which is tech-y young adult males. You can only sell each one so many cars. It will eventually saturate its core market and growth will slow unless they can sell more stuff to the same people or the same stuff to different people.
This means that have to convince "different people"--who don't care about Nikolai Tesla or EVs--to stop driving SUVs and used Honda Accords.
Tesla dabbles in other applications of its battery tech but nobody has imagined an opportunity there as big as the automobile one.
This leaves the idea of finding new stuff to sell to Musk's core fanbase. But what do you sell to people with too much money and not enough needs? An investment vehicle. Tesla stock itself is the second product offering. I wouldn't be surprised if they've, in the backrooms, made arrangements with the shady foreign shops that helped pump bitcoin to $20k, knowing that nobody seems to investigate or prosecute fraud anymore. This would be consistent with Musk's pattern of drawing all his inspiration from whatever's cool on reddit.
"Stocks (in a bubble) can become inflated through mania-fueled narratives - not fundamentals."
I disagree on nuance. Narrative often drops off during a pure bubble. Rising prices fuel mania and further rising stock prices. This is a key element of bubbles: feedback. We have a debate about TSLA in 2019. You say it's overpriced. I buy TSLA. I make money. We argue again. It's hard for you to win the argument now.
Bubble narrative tends to get vague as time goes on. Bitcoin's narrative was that it would be a real currency. People would get bitcoin salaries and buy groceries with it. That isn't the narrative anymore.
The dotcom narrative, towards the end, became a vague "this is just how the economy works now." The mania was there. The stock prices were there. The narrative itself fell off. Uber is another good example. What's the narrative now?
Tesla has some differences. A "narrative" is actually driving stock prices. Tesla investors are betting on certain things being achieved. They may or may not be achieved, and that will determine whether or not it was a good investment.
Narrative 1 is (at current prices), break-even or less for investors: Tesla will keep its position in the EV market, but the EV market will grow massively.
Narrative 2 is: Self driving is coming soon. Tesla is in position and others are not.
The first half of the narrative is suspect. "Soon" can be a long time. The second half is (IMO), pretty solid. It takes time to make new models of cars. This is the other extreme to software. If all factories switched in one day, it still takes a decade to replace the full fleet. Meanwhile, factories long lead times. Etc.
If the software works and they get a regulatory green light, Tesla immediately controls an existing fleet and all their new cars are immediately self driving. It's a very big lead in a very, very saucy new market.
Now... You can think that's likely or unlikely. But, A Tesla share is really a bet on this narrative/prediction becoming reality. A bitcoin is no longer a bet on any prediction besides future prices.
"Sometimes, performance becomes completely untethered to any analyzable fundamentals."
This is true, but I think it mistakes what these "fundamentals" mean. It's like using BMI to determine if someone is fat. True, on average. Average is good enough for a stock analyst or an insurance actuary concerned with portfolios and populations. For any individual though, the "mirror test" is more accurate than BMI. The fundamentals of any day 1 startup always say "don't buy."
A more fundamental fundamental is (value)=([p]narrative 1)X([$}narrative 1) + ([p]narrative2)X([$}narrative2)
Insert your own probabilities(p) and values and you have your guess at Tesla's "real" value.
Whatever you think of current prices, you have to be extremely bearish to come up with a value that reflect TSLA's current "fundamentals".